Rally Without a Surge? Sequential Crises and Asymmetric Mobilization in Polarized Contexts

Rally Without a Surge? Sequential Crises and Asymmetric Mobilization in Polarized Contexts | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Rally Without a Surge? Sequential Crises and Asymmetric Mobilization in Polarized Contexts Andrea Szabó This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-8523383/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Classic ‘rally-round-the-flag’ theory posits uniform increases in incumbent support during national crises. Yet, in highly polarized democracies, we argue this conventional wisdom overlooks how crises can instead generate asymmetric dynamics across partisan blocs, often without a detectable aggregate increase in incumbent support. We study Hungary under long-term populist governance during two sequential shocks within a single electoral cycle: the onset of the COVID-19 pandemic (11 March 2020) and Russia’s full-scale invasion of Ukraine (24 February 2022), the latter occurring in the run-up to the 2022 parliamentary election. Using the Vox Populi archive of public polls, we estimate segmented interrupted time-series models with pollster fixed effects, survey-mode controls where recorded, and precision weights based on trimmed sample size. Across specifications, neither crisis produces an apparent surge in support for an incumbent among likely voters. Instead, the war breakpoint coincides with a discrete weakening on the opposition side (≈ 4–5 percentage points in the baseline specification), with little evidence of compensating post-war trend change; this inference is directionally robust, but conservative ARMA(1,1) models attenuate the estimated discontinuity. Pulse-window estimates in the mass electorate indicate at most a modest and short-lived COVID-era uptick for the incumbent, whereas opposition declines are larger around the war window. Three original surveys (2021, March 2022, May 2022) help interpret these aggregate patterns: pandemic pessimism concentrates among opposition publics, whereas the war is interpreted through a ‘peace versus war’ frame aligned with the government’s campaign narrative. Overall, our findings suggest that sequential crises in polarized contexts reshape electoral competition not through uniform national rallies, but via asymmetric mechanisms, including bloc-specific vulnerabilities that contract opposition support and differential issues framing that benefits the incumbent. This reconceptualization of rally effects has important implications for understanding crisis politics in an era of heightened polarization. rally-round-the-flag COVID-19 pandemic Russia–Ukraine war Hungary poll of polls Figures Figure 1 Figure 2 Figure 3 Introduction Multiple crises rarely arrive one at a time. What happens when electorates confront distinct shocks within the same electoral cycle? Classic rally-around-the-flag scholarship (e.g., Mueller 1973 ; Brody 1991 ) predicts a visible—often temporary—surge in incumbent support as citizens “rally” behind national leadership. Yet sequential crises create an underexplored challenge for this logic: earlier shocks can condition how voters interpret later ones, and polarization can shift rally dynamics from a national swing to bloc-specific responses. We study Hungary’s 2018–2022 cycle, which includes two major shocks—the COVID-19 pandemic and the Russia–Ukraine war—within a single election cycle. The aggregate pattern is puzzling: incumbent support remains broadly stable, but the war onset coincides with a pronounced drop-in opposition support. This paper addresses a critical, yet underexplored, puzzle in contemporary political behavior: how do sequential crises in highly polarized democracies impact electoral dynamics when classic 'rally-around-the-flag' effects are seemingly absent or uneven? We theorize that crisis politics in such settings operate not through a uniform incumbent surge, but primarily via opposition-side contraction and bloc-owned framing. We address this puzzle by developing a bloc-structured theory of rally effects. In polarized contexts with segmented partisan publics and fragmented media ecosystems, crises are more likely to be translated into divergent interpretive frames and differential issue ownership than into a uniform national response (e.g., Petrocik 1996 ; Riker 1986 ). A crisis can therefore consolidate turnout and vote readiness within one bloc while the other struggles with coordination, motivation, or cross-pressures (e.g., Sides and Vavreck 2013 ). When crises are sequential, the more election-proximate shock can also override or reframe the political consequences of earlier ones by shifting what is salient when choices are made. Empirically, we link aggregate patterns to individual-level mechanisms by combining poll-based interrupted time-series (ITS) models with three linked surveys: Autumn 2021 (pandemic experiences and turnout readiness), March 2022 (pre-election participation baseline), and May 2022 post-election (closed- and open-ended vote motivations). This design allows us to separate (i) aggregate discontinuities around crisis breakpoints from (ii) bloc-differentiated motivations and readiness that can generate stable-looking aggregates. We make three contributions. First, we advance a novel bloc-structured theory of rally effects, arguing that crises in polarized democracies generate selective, rather than uniform, responses, often manifesting as opposition contraction without an incumbent surge. Second, it theorizes and tests sequential-crisis dynamics, showing how an election-proximate shock can dominate the motivational environment even after a prolonged earlier crisis. Third, it demonstrates the value of combining poll-based ITS with linked surveys to disaggregate mechanisms that remain hidden in aggregate time series. Although we study Hungary, the theoretical claim is general: in polarized systems with segmented partisan publics, crises are likely to be translated into bloc-owned frames, so electoral consequences may operate less through average preference shifts than through asymmetric mobilization, differential vote readiness, and opposition-side weakening. We treat Hungary as a hard, informative test because polarization is high and both shocks were salient within a single electoral cycle. This matters beyond Hungary because sequential crises are increasingly common, and a bloc-asymmetric lens clarifies when stable-looking aggregates can mask opposition vulnerability rather than incumbent surges. Hungary offers a hard test for sequential-crisis rally dynamics because partisan sorting and a polarized information environment in principle, should make nationwide opinion movements difficult to detect. Citizens receive and interpret crisis information through bloc-consistent media and elite cues, and crisis evaluations can be rapidly re-framed as ordinary partisan conflict. At the same time, the 2022 election provides a clear electoral endpoint and a rare combination of a prolonged public-health emergency followed by an acute external security shock close to the vote. If classic rally effects were to appear as aggregate incumbent surges, we should observe them here. If instead crises work primarily by shifting relative turnout readiness and by redistributing issue salience across blocs, then aggregate stability can coexist with decisive electoral consequences. This logic motivates our emphasis on between-bloc contrasts, short-run discontinuities, and survey-based evidence on crisis-related motivation and considered alternatives. Our framework yields four testable hypotheses about how sequential crises shape electoral dynamics in polarized settings: (H1) Classic rally (baseline expectation): A broad-based rally would appear as a discrete upward shift in incumbent support at the onset of a major crisis, especially among likely voters, visible in aggregate polling trends. (H2) Asymmetric opposition weakening: Under bloc-structured rally dynamics, crisis effects should be disproportionately visible on the opposition side. We expect downturns in opposition support and/or increases in opposition-side volatility around key crisis moments, even if incumbent support remains largely stable. (H3) Bloc-structured framing and motivation: At the micro level, crisis-relevant frames should be bloc-owned. In 2022, we expect war-related considerations (security, peace) to feature disproportionately among government voters’ stated motivations, while pandemic experiences in late 2021 relate to turnout readiness mainly via pre-existing vote intentions rather than a uniform, crisis-induced mobilization across blocs. (H4) Recency in sequential crises: When two major shocks occur within a single electoral cycle, the more proximate crisis should dominate election-proximate motivations and framing, indicating sequential salience rather than an additive national rally based on all past shocks. Figure 1 summarizes our hypotheses and empirical tests, and highlights crisis timing relative to the 2022 election (COVID onset: 11 March 2020; war onset: 24 February 2022). Note The diagram summarizes hypothesized pathways linking two crisis shocks (COVID-19 and the Russia–Ukraine war) to four mechanisms (incumbent boost, opposition weakening, reallocation, and frame ownership) and to the empirical tests used in the paper (poll ITS, event-window models, and three survey waves). Crisis timing relative to the 2022 election: COVID onset (11 March 2020; about 25 months before the election) and war onset (24 February 2022; about 5–6 weeks before the election). Theoretical framework A. Rally dynamics and heterogeneity under polarization We conceptualize ‘rally-round-the-flag’ as a crisis-period shift in political support toward incumbent authorities and governing institutions. Classic accounts locate the mechanism in heightened perceived threat, reduced partisan contestation, and citizens’ incentives to signal unity in the face of an external shock (Mueller 1970 ; Brody 1991 ; Baum 2002 ; Baker and Oneal 2001 ). In contemporary democracies, however, rally-effects are typically brief and conditional: early increases in trust or approval often fade as the shock becomes politicized and as policy costs become salient (Schraff 2021 ; van der Meer 2023 ; Devine 2024 ). Rally responses are also heterogeneous across partisan publics. Who rallies depends on elite cues, partisan trust, and the informational environment, a phenomenon implying that aggregate stability can mask substantial bloc-level movement (Edwards and Swenson 1997 ; Hetherington and Nelson 2003 ; Hegewald and Schraff 2024; Prior 2007 ; Stroud 2011 ; Levendusky 2013 ; Druckman, Peterson, and Slothuus 2013 ). Where polarization is high and governing parties dominate agenda setting and media access, opposition identifiers may not “join” the rally; instead, they may respond with skepticism, disengagement, or delayed electoral decision-making, while incumbent identifiers consolidate (Iyengar, Sood, and Lelkes 2012; Mason 2018 ). This bloc-structured view motivates a reconceptualization of what constitutes rally evidence as rally evidence in electoral contexts. Rather than expecting a symmetric, society-wide surge for incumbents, we focus on selective reinforcement of the incumbent bloc and relative opposition-side weakening within the likely electorate. These dynamics matter most near elections, when small changes in turnout readiness and late-cycle justification can become electorally decisive. More broadly, heterogeneity in crisis responses is consistent with work on the conditioning role of political trust and macropolitical public mood in shaping evaluations of incumbents (Hetherington and Rudolph 2015 ; Stimson 1999 ). Building on this, we theorize that in highly polarized contexts, the 'rally-round-the-flag' effect transforms from a symmetrical, society-wide phenomenon into an asymmetric, bloc-specific redistribution of political support. This shift implies that incumbent gains might be subtle or absent at the aggregate level, while underlying mechanisms produce critical changes in partisan readiness, motivation, and issue framing. B. COVID-19 as a prolonged crisis: normalization, performance conflict, and turnout readiness COVID-19 differs from canonical short international crises because it is prolonged, policy-intensive, and distributive. Early pandemic phases often produced increases in political trust and executive support, consistent with a classic rally logic (Bol et al. 2021 ; Kritzinger et al. 2021 ; Schraff 2021 ). Yet as restrictions persist and costs concentrate, COVID governance becomes a site of partisan contestation. Across countries, the initial rally frequently attenuated or reversed as evaluations of competence, fairness, and economic management diverged across partisan publics (van der Meer 2023 ; Colloca et al. 2024 ). In election periods, prolonged crises can reshape participation readiness rather than simply shifting vote choice. For the incumbent bloc, crisis management can be integrated into competence and stability narratives. For challengers, persistent crisis governance may depress efficacy, reduce the perceived payoff of opposition voting, or exacerbate coordination problems, especially in polarized settings where distrust is high and alternative crisis programs are hard to communicate credibly (Brouard and Michel 2025). We therefore treat COVID as a shock that can reorder the likely electorate by changing turnout intentions and late-cycle engagement. Cross-national evidence from the pandemic likewise points to short-lived or conditional rally effects, often strongest early on and fading as performance evaluations and partisan conflict intensify (Bækgaard et al. 2020 ; Johansson et al. 2021 ; Safarpour and Baum 2022 ; Sosa-Villagarcia and Hurtado Lozada 2021; Lytkina and Reeskens 2024; Mitchell et al. 2025 ). C. External security shocks, framing advantage, and issue ownership External security crises more closely match canonical rally conditions: they are sudden, salient, and tied to national sovereignty and physical safety. Such events can create a temporary unity premium for incumbents, while penalizing challengers who are portrayed as risky, inexperienced, or destabilizing (Baum 2002 ; Chávez and Wright 2022). Evidence from the Russia–Ukraine war likewise suggests that security shocks can generate rally dynamics, sometimes around national leaders and sometimes around supranational actors (Steiner et al. 2023 ; Kizilova 2024 ; Devine and Valgarðsson 2024). We connect these dynamics to issue ownership: parties benefit electorally when voters perceive them as more competent to “handle” a salient issue, and crises can sharpen or activate those competence reputations (Petrocik 1996 ; Bélanger and Meguid 2008 ; Walgrave, Lefevere, and Tresch 2012 ; van der Brug 2004 ; Stubager and Slothuus 2013). In Hungary’s 2022 campaign, the governing party was positioned to claim peace and security ownership by coupling sovereignty messaging with risk framing of the opposition. This implies that war-related motivations and post-hoc justifications should be disproportionately salient within the incumbent electorate, even if aggregate vote shares do not exhibit a large surge. Evidence from external-security and military contexts shows that rally dynamics can depend on casualties, elite framing, and the distributional politics of sanctions and war—producing asymmetric reactions across partisan publics (Lai and Reiter 2005; Kuijpers 2019 ; Verdier and Woo 2011; Frye 2019 ; Morales 2021 ; RezaeeDaryakenari et al. 2025 ; Muhammad and Undzėnas 2025; Rožukalne et al. 2022 ; Grechanaya and Ceron 2024 ). D. Sequential crises and four testable expectations When major shocks occur sequentially, recency and salience imply that the later shock can re-weight which frames dominate late-cycle decision-making and participation readiness. The earlier crisis may still structure identities and trust, but its mobilizing force can attenuate as citizens adapt and as competing grievances accumulate (Zaller 1992 ; Iyengar and Kinder 1987 ; Schraff 2021 ; Colloca et al. 2024 ). Applied to Hungary’s 2022 election cycle, this framework yields four expectations summarized in Table 1 . We expect limited aggregate rally but a bloc-structured pattern in which COVID primarily reshapes turnout readiness and the composition of the likely electorate, while the war shock selectively reinforces incumbent frames and justifications. The strongest electoral signature should therefore appear as relative opposition-side weakening among likely voters rather than as a symmetric national surge. Recent work on crisis sequencing and the stability versus volatility of distrust underscores that political reactions may depend on how new shocks re-activate existing grievances and identities rather than simply adding to a uniform national response (van Alebeek et al. 2025 ; Frateur et al. 2025 ; Leininger and Schaub 2024). Research Design and Empirical Strategy Overview: four complementary empirical approaches Poll-based interrupted time-series (ITS), 2018–2022: Using all publicly available national vote-intention polls archived by Vox Populi ( https://kozvelemeny.org/ ), we estimate segmented regressions for Fidesz and for the unified opposition. This macro lens tests whether either crisis coincides with a discontinuity in party support, particularly among likely voters— once pollster house effects and survey mode are accounted for. Autumn 2021 survey (N ≈ 5,000): We measure pandemic experiences and perceived societal impact, and link these to turnout intention to assess whether COVID-19 was associated with differential participation readiness across partisan blocs before the war shock. March 2022 pre-election survey (N = 1,000): We capture baseline engagement and political information seeking to contextualize whether the campaign environment already reflected asymmetric mobilization and attention patterns. May 2022 post-election survey (N = 1,000): We measure crisis-related vote motivations and open-ended justifications to assess bloc-specific crisis frame ownership in the immediate electoral aftermath. Our empirical strategy is designed to differentiate aggregate stability from bloc-level reallocation within a sequential-crisis electoral cycle. This is achieved by integrating (i) poll-based interrupted time-series (ITS) estimates, which identify discontinuities in support trajectories around pre-specified crisis onsets (controlling for pollster house effects and secular trends), with three distinct probability-based surveys. These surveys include: (ii) an Autumn 2021 survey capturing late-pandemic experiences and perceived societal impact linked to turnout intention; (iii) a March 2022 pre-election survey assessing baseline engagement and information-seeking behavior; and (iv) a May 2022 post-election survey designed to elicit vote motives and open-ended justifications for evaluating crisis frame ownership and within-bloc volatility indicators. Each component plays a distinct inferential role, with the ITS providing macro-level evidence and the surveys collectively benchmarking attitudes (Autumn 2021), turnout readiness (March 2022), and crisis-related vote motivations/frames (May 2022). We treat each component as a partial test, prioritizing the convergence of findings across them rather than relying solely on any single estimate. Table 1 Hypotheses and empirical tests Hypothesis Empirical test Data Outcome(s) Main specification Where reported H1 (Classic rally): crisis onset increases incumbent support Poll time series 2018–2022 Fidesz support; opposition support Level + slope change ITS with controls Main text; Fig. 2 ; Tables 2 ; App. A1 H2 (Asymmetric rally): crisis effects appear as opposition weakening more than incumbent surge Autumn 2021 & March 2022 surveys 2021 (Aut) 2022 (Mar) Turnout readiness and engagement baseline Autumn 2021: pandemic experiences and turnout intention. March 2022: bloc gaps in engagement and participation track political interest/information seeking Main text; App. A3–A4 H3 (Bloc-owned war framing): war motives concentrated among government voters May 2022 post‑election survey May 2022 War/COVID vote motives (top‑two) and crisis frames in open‑ended responses Bloc contrasts in war/COVID motivations (top‑two) and open‑ended crisis framing; any‑source measures Main text; Fig. 3 ; App. A7 H4 (Between-bloc, not within-opposition sorting): war motive does not predict opposition volatility May 2022 post‑election survey May 2022 Decision timing, campaign instability, and cross‑pressures Bloc differences in decision timing and campaign volatility; within‑opposition associations with war motives; multivariate robustness checks Main text; App. A6–A9 Data source and unit of analysis We analyze two electorates when available. The primary series uses pollsters’ likely-voter or high-propensity filters, which best match the theoretical focus on participation readiness in election-proximate crises. As a complement, we also report pulse-window tests in the mass electorate to assess whether crisis effects are confined to likely voters or appear more broadly. Our macro-level dataset draws on the Vox Populi archive of Hungarian national voting-intention polls (January 2018–April 2022). Each observation corresponds to a unique poll result for a specific party/list from a specific polling organization at a specific fieldwork period. Outcomes All analyses treat the poll as the unit of observation. In figures we also display biweekly, sample-size–weighted averages for readability, but statistical estimates are based on poll-level observations. Our primary outcomes are poll-level support shares for (i) the incumbent Fidesz–KDNP alliance and (ii) the unified opposition list. We model these as separate dependent variables to allow for asymmetric reactions: a crisis can weaken the opposition without producing a commensurate incumbent surge in the aggregate. Crisis breakpoints and time trends Because rally dynamics may be short-lived rather than permanent, we complement the segmented models with pulse-window specifications that estimate average deviations in defined windows around each breakpoint. This approach provides a direct check on whether any effects are concentrated in the weeks immediately following crisis onset. Our baseline ITS specification is a segmented regression with a pre-crisis time trend, post-crisis level indicators, and post-crisis slope changes for each breakpoint. This formulation captures both abrupt discontinuities and medium-run trajectory shifts. We define two crisis breakpoints ex ante. COVID onset is March 11, 2020, the date the Hungarian government declared a state of emergency. War onset is February 24, 2022, the date of Russia’s full-scale invasion of Ukraine. Controls and robustness Additional robustness checks (alternative weighting, alternative breakpoint codings, survey-mode controls where available, and window-length sensitivity) are reported in the Online Appendix. We model the data as a pollster-by-time-period panel of poll-level observations and estimate weighted linear mixed models with pollster fixed effects (house effects). Serial dependence is handled by specifying correlated within-pollster residuals over the time index: our main specification uses an AR(1) structure, and we assess robustness using an ARMA(1,1) structure. We treat convergence in sign and magnitude across these error processes—rather than any single p-value—as the relevant evidentiary standard. To reduce bias from differences in question wording, sampling, and mode across pollsters, all ITS models include pollster fixed effects (house effects). We estimate both unweighted specifications and precision-weighted specifications that prioritize polls with larger effective sample sizes, with trimming to limit undue influence from extreme weights. Finally, we interpret the polling models as evidence about short-run shifts in support and turnout-proxy composition within a single case, not as definitive causal estimates isolated from all concurrent political dynamics. For that reason, we complement segmented (level-and-slope) specifications with short pulse-window models around each breakpoint and focus on (a) effect sizes and confidence intervals, (b) direction-of-change consistency across specifications, and (c) contrasts between incumbent and opposition blocs that are less sensitive to slow-moving confounders. Identification and interpretive scope. Our empirical strategy leverages the sharp timing of two widely salient shocks, with breakpoints defined ex ante based on institutional dates (the first declared state of emergency for COVID-19 and the day of the Russian invasion of Ukraine). While crisis timing is plausibly exogenous to short-run polling dynamics, other processes—campaign events, economic news, or strategic coordination—can unfold in parallel. We therefore combine several safeguards: pollster fixed effects to absorb stable house effects; time trends and breakpoint-specific trend terms to separate discontinuities from gradual movements; and serial-correlation checks (Online Appendix Table A2) to assess whether inference depends on residual dependence assumptions. Identification in this setting rests on timing rather than random assignment. We pre-specify breakpoints tied to widely recognized events and focus on whether the series exhibits discrete discontinuities and/or slope changes at those points. Importantly, the two shocks occur in distinct political environments—COVID early in the cycle and the war during an already campaign-like period—which provides additional leverage: if the models were merely capturing generic polling volatility, we would expect similar level shifts for both blocs across shocks. Instead, we assess whether discontinuities are bloc-specific and whether post-shock slopes compensate. We also treat the likely-voter series as a conservative test because it reduces compositional changes driven by turnout intention, and we weight by trimmed sample size to reflect measurement precision. Finally, we interpret the polling evidence alongside survey measures of crisis perceptions and vote motivations. Survey data and measures Our micro-level evidence comes from three probability-based national surveys fielded by the same survey program and post-stratified to population benchmarks. Together, they allow us to (i) gauge whether COVID-19 was associated with differential turnout readiness before the war shock, (ii) benchmark pre-election engagement patterns, and (iii) identify crisis frame ownership in voters’ reported motives and open-ended justifications. Autumn 2021 survey (N ≈ 5,000; COVID experiences and turnout intention) Our main outcome in this component is turnout intention, coded as a binary indicator for reporting high likelihood of voting. We use these measures to evaluate whether pandemic experiences were linked to participation readiness in a bloc-asymmetric manner. From the Autumn 2021 face-to-face survey, we construct two indices that capture COVID-19 exposure and perceptions. A “personal impact” index summarizes respondents’ reported household/health exposure and perceived personal threat, whereas a “societal impact” index summarizes perceived consequences for the economy and society. March 2022 pre-election survey (N ≈ 1000, baseline engagement) The March 2022 pre-election survey provides a baseline snapshot of political engagement shortly after the war shock and before the election. We construct a participation repertoire index (count of reported political activities) and measures of political information seeking to benchmark cross-bloc differences in attention and readiness. May 2022 post-election survey (N ≈ 1000, vote motivations and crisis frames) To probe whether the bloc-structured mechanism manifests as opposition-side volatility or demobilization signals, we operationalize four indicators: late deciding, campaign instability, considering an alternative list, and split-ticket voting. These are analyzed descriptively by bloc and then in weighted logistic models as robustness checks. For closed-ended motives, we create binary indicators for whether respondents selected war/peace/security and whether they selected COVID-related considerations among their top two reasons for vote choice. For open-ended responses, we code mentions of war/Ukraine/security/peace and other major categories to capture spontaneous framing. The May 2022 post-election survey is used to assess crisis framing and frame ownership. We draw on two sources: closed-ended top-two vote motives and open-ended vote justifications recorded verbatim and coded into substantive categories (see Online Appendix for codebook and procedures). Multivariable strategy and reporting Importantly, these regressions are not treated as causal identification devices. Their primary role is to ascertain whether the descriptive bloc asymmetries that motivate our argument are robust to plausible compositional differences across partisan publics. The main text prioritizes transparent descriptive contrasts and simple hypothesis-linked tests. Where multivariable adjustment is informative, we estimate weighted logistic regressions to verify that bloc differences in crisis framing, and volatility indicators persist when accounting for standard covariates (demographics and baseline political orientation measures). Results Polling time series among likely voters We begin by testing whether either crisis produced the classic aggregate “rally-round-the-flag” pattern—a discrete, nationwide increase in incumbent support at crisis onset (H1). Using deduplicated poll-level observations for likely voters (“Biztos”, N = 226) from the Vox Populi archive, we estimate segmented interrupted time-series (ITS) models with breakpoints at the start of the COVID-19 pandemic (11 March 2020) and Russia’s full-scale invasion of Ukraine (24 February 2022). Notably, the war shock (24 February 2022) occurred less than two weeks after the official campaign began (12 February 2022), and after the election date had been called on 11 January 2022—effectively placing the shock inside an already campaign-like information environment. All specifications include pollster fixed effects (house effects) and, where available, survey-mode controls. In these segmented ITS models, the post-crisis indicator captures an immediate level shift at the breakpoint (a discontinuity), while the post-crisis time term captures a change in slope relative to the pre-crisis trend. Because political shocks can generate brief “bumps” rather than durable re-alignments, we treat short-run level shifts (and, in the next section, pulse-window estimates) as the most direct evidence of rally-like reactions, and we interpret longer-run trend changes more cautiously. Across specifications we report unstandardized coefficients in percentage points and emphasize substantive magnitudes and uncertainty (confidence intervals) rather than significance alone; sensitivity to serial-correlation assumptions is summarized below and documented in the Online Appendix. Table 2 reports segmented ITS estimates for Fidesz support among likely voters. Neither crisis onset is associated with a discrete increase: the COVID-19 level change is small and statistically indistinguishable from zero (B = 0.549, p = 0.612), and the war-onset level change is essentially zero (B = − 0.014, p = 0.996). The pre-crisis trend is slightly negative (B = − 0.007 per day, p < 0.001), with only weak evidence that it attenuates after COVID (post-COVID slope change B = 0.004, p = 0.090). Overall, the likely-voter series provides little support for a classic, nationwide aggregate rally (H1). Opposition support follows a different trajectory. In Table 2 , the baseline specification implies an immediate post-invasion drop of roughly 4½ percentage points (B = − 4.449, p = 0.065), whereas the positive post-COVID drift is interrupted at the war breakpoint. We interpret this pattern as evidence consistent with opposition-side weakening (H2), rather than as an incumbent surge. Because these conclusions could be sensitive to time-series dependence, we next assess robustness to alternative residual-correlation structures. These ITS estimates speak to aggregate polling dynamics, not individual-level opinion change; they therefore cannot adjudicate micro-level mechanisms directly. Null average level shifts also do not rule out offsetting subgroup movements (e.g., compositional reweighting) that net out at the national aggregate. Stepping back, the segmented ITS results already differentiate between a classic incumbent rally and bloc-asymmetric adjustment. For Fidesz, neither crisis produces a clear level jump, and post-crisis slope changes are modest, consistent with a relatively stable incumbent baseline among likely voters. For the opposition, by contrast, COVID is associated mainly with gradual drift, whereas the war shock is concentrated in an immediate downward discontinuity. This pattern is precisely what our sequential framework highlights: crises can matter electorally through opposition contraction even when the incumbent does not visibly surge. The remaining analyses therefore ask whether this inference survives conservative residual structures (Online Appendix Table A2) and whether short-run pulse windows in the mass electorate show any transient incumbent uptick. Table 2 Interrupted time-series models (likely voters / “Biztos ” ). Predictor Fidesz (B (SE)) p Opposition (B (SE)) p Time trend (t) -0.007 (0.002) < 0.001 0.000 (0.002) 0.804 COVID level change (post_covid01) 0.549 (1.082) 0.612 0.486 (0.903) 0.591 Post-COVID slope change (t_covid) 0.004 (0.003) 0.090 0.008 (0.002) < 0.001 War level change (post_war01) -0.014 (2.871) 0.996 -4.449 (2.396) 0.065 Post-war slope change (t_war) 0.044 (0.110) 0.693 0.051 (0.092) 0.579 Model statistics N = 226; R²=0.612 (Adj. 0.579) N = 226; R²=0.529 (Adj. 0.488) Note. Unstandardized coefficients (percentage points) with standard errors in parentheses. Models include pollster fixed effects and survey-mode controls where available (mode controls are omitted from Online Appendix Table A2 for ensuring comparability across covariance specifications). Serial-correlation sensitivity Online Appendix Table A2 re-estimates the segmented ITS models using SPSS MIXED, allowing for within-pollster serial dependence and precision weights based on trimmed sample size (REGWGT = n_trim). Under an AR(1) residual structure, the estimated autocorrelation is small and not statistically distinguishable from zero (Fidesz: ρ = 0.070, p = 0.622; Opposition: ρ = 0.152, p = 0.346), and the substantive conclusions match the main specification. In particular, the war-onset shift remains near zero for Fidesz and remains a negative discontinuity for the opposition (AR(1): B = − 4.758, p = 0.030). Under ARMA(1,1), autocorrelation is absorbed by a more flexible process (e.g., ρ ≈ 0.94), and the estimated war-onset shift in the opposition series is attenuated and only marginal (B = − 3.570, p = 0.082). Overall, serial-correlation checks yield small, non-significant AR(1) estimates and do not alter the substantive conclusions; the more conservative ARMA(1,1) specification attenuates the war-level discontinuity for the opposition (Online Appendix Table A2). Opposition-side discontinuity around the war Opposition dynamics sharpen this asymmetry. Opposition support rises gradually during the pandemic period (post-COVID slope change ≈ 0.008 pp/day, p < 0.001) but drops immediately after the invasion (war level change ≈ − 4.5 to − 4.8 pp across Table 2 and Online Appendix Table A2). Across specifications there is little evidence of a compensating post-war slope change that would rapidly restore the pre-war trajectory. In sum, crisis politics are more visible as an opposition-side contraction than as abrupt gains in the incumbent series. This pattern is consistent with our sequential, bloc-structured account: prolonged COVID does not generate a clean rally, while the acute war shock coincides with a discrete weakening on the opposition side with little sign of an incumbent surge. We next examine whether these short-run dynamics differ between likely voters and the mass electorate using pulse-window models. Short-run pulse windows and voter-based heterogeneity To distinguish short-lived “bumps” from sustained shifts and to examine heterogeneity across voter bases, we next estimate non-overlapping pulse-window models in the mass electorate (“Összes”). These models focus on narrow windows around each crisis onset and provide a direct test of brief rally-like responses. For Fidesz, pulse-window estimates in the mass electorate suggest at most a modest short-run uptick around COVID onset, with no comparable short-run increase around the war; effects are weaker among likely voters. For the opposition, event windows point to declines, with larger point estimates around the war than around COVID—again consistent with war-timed opposition-side contraction. Comparing electorates reinforces this interpretation: the incumbent series is comparatively stable among committed voters, while opposition losses around the war are clearly visible in the mass electorate and appear as a war-timed discontinuity in the likely-voter segmented models (Table 2 ; Online Appendix Table A2). Table 3 Pulse-window models around COVID onset and war shock (mass electorate vs likely voters). Mass electorate (“Összes”, N = 250) Likely voters (“Biztos”, N = 226) — incumbent series (Fidesz) Predictor Fidesz (B (SE)) p Opposition (B (SE)) p Fidesz (B (SE)) p Time trend (t) 0.004 (0.001) < 0.001 0.009 (0.001) < 0.001 -0.004 (0.001) < 0.001 COVID 0–90 days (covid_0_90) 0.419 (0.901) 0.642 -3.084 (1.058) 0.004 1.912 (1.117) 0.088 COVID 91–180 days (covid_91_180) 1.825 (0.868) 0.037 -3.018 (1.019) 0.003 1.368 (1.183) 0.249 War level change (post_war01) 2.336 (1.412) 0.099 -4.219 (1.658) 0.012 0.954 (2.811) 0.735 Post-war slope change (t_war) 0.046 (0.060) 0.443 0.079 (0.071) 0.264 0.034 (0.110) 0.756 Model statistics R²=.613 (Adj. 0.580) Note. Models use pollster fixed effects and mode controls; time is indexed from COVID onset. Full diagnostics are reported in the Online Appendix. Note The figure uses deduplicated poll-level observations from the Vox Populi archive. For each biweekly interval (indexed by its start date), lines plot sample-size–weighted means of reported support (weights = poll n ) for the mass electorate (“Összes”) and likely voters (“Biztos”). Panel A shows Fidesz support; Panel B shows opposition support. Vertical dashed lines mark the onset of the COVID-19 pandemic (11 March 2020) and Russia’s full-scale invasion of Ukraine (24 February 2022). Differences in levels between bases reflect the likely-voter definition, and the inclusion of undecided/nonvoters in the mass-electorate series. Where no estimate is available for a given base in a biweekly period, the corresponding series is shown with gaps. The next section uses original survey data to probe the individual-level perceptions and motivations that can generate these bloc-asymmetric aggregate movements. From Polls to People: Survey Evidence Poll-level ITS models describe aggregate discontinuities but cannot by themselves identify the micro-level pathways—such as turnout readiness, decision timing, or bloc-owned interpretive frames—that could generate similar macro-patterns. We therefore triangulate the polling evidence with three linked surveys to assess (i) whether crisis experiences map onto vote readiness and (ii) whether the war is translated into partisan-bloc-owned motivations close to the election. The survey evidence is designed to adjudicate between a classic nationwide rally and an asymmetric mechanism: muted movement in the incumbent’s aggregate support paired with opposition vulnerability and bloc-owned crisis frames. We combine an Autumn 2021 module on COVID experiences and turnout intention, a March 2022 pre-election module on engagement and participation, and a May 2022 post-election module measuring stated vote motivations (closed-ended “top-two reasons” and open-ended justifications). Together, these data connect aggregate discontinuities to turnout readiness, decision timing, and crisis framing. COVID-19 Perceptions and Turnout Readiness (Autumn 2021) Using the Autumn 2021 face-to-face survey (N ≈ 5,000), we estimate logistic models predicting a top-two-box indicator of turnout intention (“certainly/probably would vote”), adjusting for standard sociodemographics, self-reported health, and COVID-related covariates (e.g., perceived personal impact, societal concern, vaccination status). Full model estimates are reported in Online Appendix Table A4. Turnout readiness is strongly structured by baseline partisanship: self-identified Fidesz voters report higher intended turnout than opposition voters even after covariate adjustment. This gap matters for interpreting later crisis dynamics, because an opposition advantage in general engagement does not necessarily translate into electoral leverage when vote readiness remains uneven across blocs. At the same time, pandemic perceptions do not translate into uniform mobilization. Viewing COVID-19 as a broader societal problem is associated with a lower likelihood of intending to vote, whereas perceived personal impact shows no independent association once covariates are included; vaccination status is positively associated with turnout intention. Online Appendix Table A4 reports these estimates, and interaction terms provide no robust evidence that these relationships differ systematically between Fidesz and opposition voters in this late-pandemic snapshot. Taken together, the late-pandemic survey evidence indicates that COVID-related concern can coincide with political withdrawal for some respondents—not a generalized rally-like mobilization—and that baseline vote readiness differs across partisan blocs. This pattern is consistent with our macro finding that COVID does not generate a clean, aggregate rally in incumbent support. Pre-election Participation Baseline (March 2022) We next use the March 2022 pre-election survey to assess whether partisan blocs differed in non-electoral participatory readiness on the eve of the election, shortly after the war’s outbreak. Negative binomial models of a participation repertoire index (0–8 activities; QM9) reveal that any bivariate opposition–government gap is not robust once we adjust for baseline political engagement. Controlling for political interest (0–100) and time spent on political information (minutes per day) eliminates the bloc difference (Opposition vs. Fidesz: B = 0.045, p = 0.770); full estimates are reported in Online Appendix Table A4. Interest and information-seeking remain strong predictors of participation, suggesting that pre-election participation differences primarily proxy general attentiveness rather than crisis-specific partisan mobilization capacity. In combination, the Autumn 2021 and March 2022 surveys narrow the set of plausible micro-mechanisms behind the polling discontinuities. COVID-related perceptions are not tightly coupled to turnout readiness in a way that would imply an automatic “rally → mobilization” dynamic, and participation differences prior to the election largely reflect baseline engagement and information acquisition. These results make a classic, uniform rally less likely and motivate a focus on how the subsequent acute crisis—the war—could reshape electoral alignment through bloc-owned framing and asymmetric vulnerability. Post-election Vote Motivations and Crisis Frames (May 2022) We then turn to the May 2022 post-election face-to-face survey (N ≈ 1,000) to connect the aggregate time-series pattern to individual-level motivations. Because post-election reports may involve post-hoc rationalization, we interpret these measures primarily as indicators of bloc-specific crisis frames—that is, the considerations respondents make available and legitimate when explaining their vote—rather than as clean causal drivers. The post-election survey indicates sharply asymmetric crisis framing consistent with bloc-owned motivations (H3). Figure 3 summarizes crisis-related vote motivations (see Online Appendix Table A7 for full distributions). War-related considerations are a dominant vote motive among government voters: 55.9% of Fidesz voters cite the war among their top-two reasons for voting, compared to 18.2% of opposition voters. Open-ended accounts reinforce this asymmetry: war justifications appear almost exclusively among government voters (13.2% of Fidesz voters; 0% among opposition voters). By contrast, COVID-19 is largely absent as a salient post-election vote cue. Across closed- and open-ended items, only 3.2% of respondents (22 of 687 valid cases) mention COVID-19 as a top-two motive. Mentions are rare overall, though slightly more frequent among Fidesz voters (4.8% vs. 0.4% among opposition voters). Given the low base rate, we treat this pattern as descriptive rather than as a basis for multivariable modeling. Bars show the share of respondents citing the war or COVID-19 among their top-two closed-ended reasons for their list vote, by vote bloc (Fidesz vs joint opposition). War is a dominant and sharply polarized motive (55.9% vs 18.2%), whereas COVID-19 is rare overall and marginally more common among Fidesz voters (4.8% vs 0.4%). To assess whether asymmetric war framing is accompanied by differential volatility or late deciding, we examine decision timing and campaign instability. Opposition voters are significantly more likely than Fidesz voters to report deciding in the final week/days before the election (12.1% vs. 6.4%; χ² = 6.66, p = 0.01, weighted), consistent with weaker baseline vote readiness on the opposition side. A broader indicator of deciding after the war onset (including “one month before the election”) shows only a modest and statistically non-significant bloc difference (20.2% vs. 15.8%; p = 0.14) and reported campaign instability during February–election day is similar across blocs (24.5% vs. 23.7%; p = 0.81). Within the opposition bloc, citing the war among the top-two motives are not associated with campaign instability or decision timing (all p > 0.60; see Online Appendix Table A7), suggesting that the war frame primarily differentiates between blocs rather than sorting voters within the opposition camp. The broader post-war decision-timing proxy includes the response category “one month before the election,” which can fall just before or just after 24 February 2022. We therefore treat this measure as potentially noisy, though still informative, for detecting war-related late deciding. Finally, patterns of considered alternatives (PE8) suggest subtle cross-pressures, particularly among eventual Fidesz voters. Fidesz voters are more likely than opposition voters to report having considered another list (13.1% vs. 4.9%; χ² = 11.51, p = 0.001; OR = 2.92; see Online Appendix Table A7). Among those who considered an alternative (N = 69 weighted), most eventual Fidesz voters named Fidesz itself as the alternative (≈ 67%), indicating intra-camp deliberation, but some considered the joint opposition (≈ 12%) or Our Homeland (Mi Hazánk; ≈9%). These cross-pressures point to modest volatility on the incumbent side rather than war-driven re-sorting within the opposition bloc. Online Appendix Tables A8–A9 report multivariable models that reinforce these descriptive patterns. Adjusting for demographics, settlement type, ideology, and political interest, Fidesz voting remains strongly associated with war-related motive claims and earlier decision timing, while within-opposition war mentions do not predict volatility proxies. Overall, the multivariable evidence supports the interpretation that the war frame differentiates between blocs rather than explaining within-bloc volatility. Taken together, the survey evidence provides micro-foundations for why the war breakpoint appears in the aggregate polls primarily as an opposition-side drop rather than a clean incumbent surge. The war becomes a strongly bloc-owned motivational frame among government voters (consistent with H3), while opposition voters exhibit weaker vote readiness and a higher propensity for last-minute decision-making. Across indicators, we find limited evidence that war motivations sort voters within the opposition bloc. In combination with the polling discontinuity, these findings support our sequential, bloc-structured account of rally effects in polarized contexts (H2–H3) and help reconcile muted aggregate movement in incumbent support with sharper war-timed opposition weakening. Discussion We emphasize that these patterns should be read as discontinuities consistent with a crisis-induced reallocation, not as definitive causal effects of the crises themselves. The ITS design strengthens temporal attribution by focusing on sharp breakpoints and pre-trend structure, but it cannot fully rule out all concurrent campaign and information shocks. Across four complementary empirical components, the evidence is most consistent with a selective and asymmetric crisis process. In the aggregate poll series, we do not observe a large, sustained incumbent jump at either crisis onset. Instead, the clearest macro-level movement is an opposition-side weakening around the war shock, coupled with micro-level evidence that the war became a highly salient motivational frame among incumbent supporters while remaining comparatively muted among opposition voters. This study re-evaluates rally dynamics under sequential crises in a highly polarized, dominant-party setting. Rather than assuming that external shocks generate a broad, aggregate surge in incumbent support, we theorize and test a bloc-structured pattern in which crises can reallocate participation readiness and electoral competition asymmetrically across partisan publics. Table 4 Summary of Findings Across Empirical Components Component (what it tests) “Classic rally” expectation What we find Mechanism implication Poll-of-polls ITS (2018–2022) Large, abrupt pro-incumbent shift after crises No clean Fidesz level shift; war coincides with an opposition-side level drop Rally can show up as opposition weakening/re-sorting , not only incumbent jumps Autumn 2021 survey (COVID perceptions → turnout intention) Pandemic threat mobilizes broadly COVID perceptions do not “mobilize” uniformly; effects look like composition + readiness differences Crisis effects filter through who is vote-ready , not national mood March 2022 pre-election baseline participation Opposition “anti-incumbent” energy should dominate engagement Apparent partisan gaps largely track political interest/info-seeking Participation capacity is not automatically electoral leverage without readiness May 2022 post-election motives + frames Crisis cues shared across electorate War is dominant and sharply polarized by bloc; COVID is rare as a motive War is interpreted and owned disproportionately by the pro-government camp; COVID has faded as an election cue Re-assessing the ‘dual rally’ hypothesis This sequential-crisis pattern speaks to an important boundary condition for rally theories. When trust and credibility are polarized, the key electoral consequence of crises may be the redistribution of readiness and cohesion across blocs rather than visible aggregate uplift for incumbents. Overall, the Hungarian case provides limited support for a simple “dual rally” expectation in which COVID-19 and the war each produce broad incumbent surges (H1). Instead, the evidence points toward an asymmetric mechanism: COVID-19 appears mainly as a prolonged background condition that does not translate into an election-proximate incumbent boost, whereas the war—by timing and by frame ownership—aligns more closely with an opposition-side contraction in the likely electorate (H2). Explaining the Landslide: How Muted Aggregate Rally Still Produces Decisive Electoral Outcomes Our survey evidence clarifies one pathway for such divergence: the war was disproportionately integrated into incumbent voters’ stated motivations and open-ended justifications, consistent with security/peace frame ownership (H3). In a context where turnout and persuasion are already asymmetric, such frame alignment can contribute to a landslide outcome even if the aggregate level of incumbent support does not surge. A broader implication concerns how decisive electoral outcomes can emerge even when poll averages look comparatively stable. Small shifts in the composition of the likely electorate—especially if concentrated on one side of a polarized competition—can widen electoral margins without producing a dramatic aggregate “rally spike.” These findings also help reconcile muted aggregate rally patterns with large electoral margins. In highly polarized settings, elections are often decided at the margin among weaker partisans and intermittent voters; modest bloc-specific demobilization or defection can therefore have outsized consequences when concentrated asymmetrically. A crisis that does not attract new supporters to the incumbent may still reduce the opposition’s effective coalition by lowering enthusiasm, increasing cross-pressures, or reweighting the salience of issues on which the incumbent holds framing or ownership advantages. From this perspective, the key electoral object is not the national “rally” level shift, but the distribution of readiness and cohesion across blocs as the campaign unfolds. The war shock in February 2022 is a particularly sharp instance because it entered an already campaign-like information environment, allowing government framing to structure interpretation quickly. This mechanism complements—rather than replaces—standard accounts centered on incumbency advantage and resource mobilization. Alternative explanations: opposition coordination, incumbency advantage, and fiscal mobilization Two considerations help situate the crisis-related patterns in the broader 2022 electoral context. First, unlike earlier opposition cycles in Hungary, the challengers entered the campaign as a coordinated joint list with a single prime-ministerial candidate. Coordination should, in principle, mitigate vote-splitting and strengthen the credibility of alternation. Yet it can also raise coordination costs and messaging tensions in heterogeneous coalitions, potentially depressing enthusiasm among peripheral supporters and leaving less room for rapid agenda shifts once the campaign enters its final phase. Second, the election unfolded under pronounced incumbency advantages. In competitive-authoritarian settings, incumbents can combine agenda control, asymmetric access to media, and the strategic deployment of state resources with rule-based advantages that shape the information environment and the perceived costs of defection. These tools are well documented in the comparative literature on electoral authoritarianism and hegemonic party dominance (Schedler 2002 ; Magaloni 2006 ; Levitsky and Way 2010 ; Schedler 2013 ). A related, and electorally salient, channel is pre-election fiscal mobilization. Political-budget-cycle theories emphasize that incumbents may tilt policy toward highly visible transfers, tax relief, or price interventions in the run-up to an election, especially where monitoring is imperfect and partisan cues structure attribution (Rogoff 1990 ; Brender and Drazen 2005 ; Shi and Svensson 2006 ). In Hungary, such measures can plausibly elevate baseline incumbent support and turnout readiness among government-leaning voters. These alternative explanations are not substitutes for the crisis account advanced here. If anything, they help clarify why the war shock could matter electorally even when aggregate rally effects appear muted. Structural incumbency advantages and fiscal mobilization can raise the level of government support, but they do not predict the timing of the within-opposition contraction around the war breakpoint. The war, by contrast, plausibly provided an affectively charged and frames-relevant cue that the government could claim as an issue-ownership advantage (peace/security), consistent with the post-election motivation patterns among Fidesz voters and the open-ended responses. Addressing Within-Opposition Dynamics and Explanatory Modeling Concerns At the same time, within-bloc heterogeneity and organizational dynamics remain plausible complementary explanations. Future work with panel data and validated turnout measures is needed to separate preference change, turnout propensity change, and measurement artifacts in crisis-period opposition contraction. Our argument is primarily between-bloc: war framing and crisis-period readiness differ sharply across the incumbent and opposition publics. Within the opposition, we find limited evidence that war motivations systematically predict late deciding or campaign instability. This suggests that the opposition’s weakening is less about widespread last-minute switching within the opposition and more about asymmetric readiness and frame disadvantage relative to the incumbent bloc. Alternative explanations and why timing matters. Several concurrent developments could in principle shape opposition support in early 2022, including inflationary pressures, campaign events, or organizational frictions within the heterogeneous opposition alliance. Most such forces, however, should generate gradual movements or affect both blocs in the same direction, rather than producing an abrupt, opposition-specific break aligned with the invasion date. The observed pattern—muted incumbent movement but a sharper opposition-side discontinuity, together with survey evidence that war-related considerations were disproportionately available to (and claimed by) incumbent voters—fits a framing/ownership mechanism that operates between blocs. We nevertheless treat the macro discontinuity as within-case evidence that is strengthened by convergence across designs, not as a definitive estimate that excludes all parallel shocks. Limitations and Future Research Avenues First, the macro evidence comes from a single-country interrupted time-series design with one treated unit and relies on sharp timing assumptions. Although we address pollster house effects, time-series dependence, and window-length sensitivity, unobserved concurrent events cannot be fully excluded. Second, post-election motive reports are subject to rationalization and identity-consistent storytelling; we therefore interpret motive patterns as indicators of frame availability and ownership rather than as direct causal drivers of vote choice. Third, some behavioral proxies (e.g., decision timing categories) are coarse. We treat these measures as suggestive indicators of volatility and emphasize convergence across measures and datasets rather than any single item. On balance, the evidence supports a conservative reading: the war shock is associated with a bloc-asymmetric shift that is consistent with rally dynamics among incumbent supporters, even as aggregate series remain muted because countervailing movements occur across blocs. Our designs cannot uniquely attribute the 2022 landslide to the war alone; incumbency advantages, pre-election transfers, and within-opposition coordination frictions plausibly contributed in parallel. What the analysis does establish is that the war’s timing relative to the election coincided with (i) changes in opposition support in the polling series and (ii) post-election war-related vote rationales that are highly concentrated among Fidesz voters. Accordingly, the article’s core contribution is best read as a boundary condition for rally effects in polarized, dominant-party settings: crises may matter primarily through partisan reallocation and turnout readiness rather than through a uniform national surge. Future work can adjudicate these pathways with cross-national designs and individual-level panels that track both preferences and turnout propensity across sequential shocks. Conclusion and Broad Contributions More broadly, our findings mandate a fundamental re-conceptualization of 'rally' effects. Instead of solely seeking aggregate incumbent uplift, scholars must consider 'rally' as a process of selective redistribution across partisan publics, driven by differential vulnerabilities and framing advantages. This new lens is crucial for future comparative research, guiding investigations into how institutional credibility, fragmented information environments, and opposition coordination critically condition crisis politics across diverse regimes and electoral cycles. We expect the sequential, bloc-asymmetric pattern documented here to be most likely when (a) partisan identities structure information exposure and elite credibility, (b) the incumbent can plausibly claim superior competence on the focal crisis dimension (e.g., security/peace), and (c) a major shock occurs close enough to an election that turnout readiness and issue salience are high. In such settings, crises may stabilize or even amplify incumbent advantages without generating a visible nationwide surge in average incumbent support. Comparative work that combines high-frequency opinion series with pre- and post-election survey modules across cases could test these scope conditions directly and clarify when aggregate rally effects are most likely to appear—or to remain hidden by bloc-level countervailing movements. Despite these constraints, the study offers a clear contribution to the rally literature: in polarized, dominant-party contexts, sequential crises may not produce additive incumbent surges. Instead, crisis timing and frame ownership can generate selective rally dynamics that are visible primarily as opposition-side weakening and asymmetric mobilization of the likely electorate. More broadly, our sequential-crisis lens travels significantly beyond the Hungarian case, offering crucial insights for understanding crisis politics in other polarized settings. We caution scholars against inferring 'no rally' solely from seemingly flat aggregate polls when multiple shocks occur close to an election. Instead, the critical theoretical and empirical question shifts to how each shock is strategically translated into a credible voting reason within competing partisan narratives , and which political camp is ultimately better positioned—through its elite messaging, media environment, and organizational capacity—to effectively own and communicate that frame . This approach provides a more nuanced and accurate understanding of how exogenous events reshape electoral behavior in the 21st century. Declarations Ethics. This article presents a secondary analysis of anonymized survey data and publicly available polling data. No new data was collected for this study, and no additional ethical approval was required for the secondary analyses. Funding. This research received no external funding. Author Contribution Andrea Szabó conceived the study, developed the theoretical framework, designed the research, conducted the analyses, prepared the figures and tables, and wrote the manuscript and online appendix. Data Availability Polling data are available from the Vox Populi archive. Replication code and derived datasets (including the aggregated time-series used for analysis) will be deposited in an open repository upon acceptance. Access to the original survey microdata is restricted; access may be granted upon reasonable request. References Bækgaard, M., Christensen, J., Madsen, J. K., & Kim Sass, M. (2020). Rallying around the flag in times of COVID-19: Societal lockdown and trust in democratic institutions. Journal of Behavioral Public Administration , 3 (2). https://doi.org/10.30636/jbpa.32.172 Baker, W. D., & Oneal, J. R. (2001). Patriotism or opinion leadership? The nature and origins of the rally ’round the flag effect. Journal of Conflict Resolution , 45 (5), 661–687. https://doi.org/10.1177/0022002701045005006 Baum, M. A. (2002). The constituent foundations of the rally-round-the-flag phenomenon. International Studies Quarterly , 46 (2), 263–298. https://doi.org/10.1111/1468-2478.00232 Bélanger, É., & Meguid, B. M. (2008). Issue salience, issue ownership, and issue-based vote choice. Electoral Studies , 27 (3), 477–491. https://doi.org/10.1016/j.electstud.2008.01.001 Bol, D., Giani, M., & Blais, A., and Peter John Loewen (2021). The effect of COVID-19 lockdowns on political support: Some good news for democracy? European Journal of Political Research , 60 (2), 497–505. https://doi.org/10.1111/1475-6765.12401 Brender, A., & Drazen, A. (2005). Political budget cycles in new versus established democracies. Journal of Monetary Economics , 52 (7), 1271–1295. https://doi.org/10.1016/j.jmoneco.2005.04.004 Brody, R. A. (1991). Assessing the President: The Media, Elite Opinion, and Public Support . Stanford University Press. Brouard, S., and Elie Michel (2025). Retrospective voting in times of pandemic: Comparative evidence from panel data. Political Science . https://doi.org/10.1080/00323187.2025.2500926 . Advance online publication. Chávez, K., and James Wright (2022). International crisis termination and presidential approval. Foreign Policy Analysis , 18 (3), orac005. https://doi.org/10.1093/fpa/orac005 Colloca, P., Roccato, M., & Russo, S. (2024). Rally ’round the flag effects are not for all: Trajectories of institutional trust among populist and non-populist voters. Social Science Research , 119 , 102986. https://doi.org/10.1016/j.ssresearch.2024.102986 Devine, D., & Viktor Orri, V. (2024). Stability and change in political trust: Evidence and implications from six panel studies. European Journal of Political Research , 63 (2), 478–497. https://doi.org/10.1111/1475-6765.12606 Druckman, J. N., Peterson, E., & Slothuus, R. (2013). How elite partisan polarization affects public opinion formation. American Political Science Review , 107 (1), 57–79. https://doi.org/10.1017/S0003055412000500 Edwards, G. C., III, and, & Swenson, T. (1997). Who rallies? The anatomy of a rally event. Journal of Politics , 59 (1), 200–212. https://doi.org/10.2307/2998222 Frateur, J., Bursens, P., Popelier, P., & Coroado, S. (2025). How to explain citizens’ declined political trust during the COVID-19 pandemic: a time-series qualitative comparative analysis (TsQCA). Comparative European Politics , 23 , 552–575. https://doi.org/10.1057/s41295-025-00412-z Frye, T. (2019). Economic sanctions and public opinion: Survey experiments from Russia. Comparative Political Studies , 52 (7), 967–994. https://doi.org/10.1177/0010414018806530 Grechanaya, T., & Ceron, A. (2024). Patriotism and national symbols in Russian and Ukrainian elections. Politics and Governance , 12 (2), 355–367. https://doi.org/10.17645/pag.7918 Hegewald, S., and Dominik Schraff (2024). Who rallies around the flag? Evidence from panel data during the Covid-19 pandemic. Journal of Elections Public Opinion and Parties , 34 (1), 158–179. https://doi.org/10.1080/17457289.2022.2120886 Hetherington, M. J., & Nelson, M. (2003). Anatomy of a rally effect: George W. Bush and the War on Terrorism. PS: Political Science & Politics 36(1): 37–42. https://doi.org/10.1017/S1049096503001665 Hetherington, M. J., & Rudolph, T. J. (2015). Why Trust Matters: Declining Political Trust and the Demise of American Liberalism . Princeton University Press. Iyengar, S., & Kinder, D. R. (1987). News That Matters: Television and American Opinion . University of Chicago Press. Iyengar, S., & Sood, G., and Yphtach Lelkes (2012). Affect, not ideology: A social identity perspective on polarization. Public Opinion Quarterly , 76 (3), 405–431. https://doi.org/10.1093/poq/nfs038 Johansson, B., Hopmann, D. N., & Shehata, A. (2021). When the rally-around-the-flag effect disappears, or when the COVID-19 pandemic becomes normalized. Journal of Elections Public Opinion and Parties , 31 (sup1), 321–334. https://doi.org/10.1080/17457289.2021.1924742 Kizilova, K., and Pippa Norris (2024). Rally around the flag effects in the Russian–Ukrainian war. European Political Science , 23 (2), 234–250. https://doi.org/10.1057/s41304-023-00450-9 Kritzinger, S., Foucault, M., Lachat, R., Partheymüller, J., & Eberl, M. S., and Marko Bubalo (2021). Rally round the flag’: The COVID-19 crisis and trust in the national government. West European Politics , 44 (5–6), 1205–1231. https://doi.org/10.1080/01402382.2021.1925017 Kuijpers, D. (2019). Rally around all the flags: The effect of military casualties on incumbent popularity in ten countries 1990–2014. Foreign Policy Analysis , 15 (3), 392–412. https://doi.org/10.1093/fpa/orz014 Lai, B., and Dan Reiter (2005). Rally ’round the Union Jack? Public opinion and the use of force in the United Kingdom, 1948–2001. International Studies Quarterly , 49 (2), 255–272. https://doi.org/10.1111/j.0020-8833.2005.00344.x Leininger, A., and Max Schaub (2024). Strategic alignment in times of crisis: Voting at the dawn of a global pandemic. Political Behavior , 46 , 1589–1607. https://doi.org/10.1007/s11109-023-09885-8 Levendusky, M. S. (2013). Why do partisan media polarize viewers? American Journal of Political Science , 57 (3), 611–623. https://doi.org/10.1111/ajps.12008 Levitsky, S., & Way, L. A. (2010). Competitive Authoritarianism: Hybrid Regimes after the Cold War . Cambridge University Press. Lytkina, E., and Tim Reeskens (2024). Rally around the government or a populist response? How concerns about COVID-19 and emotional responses relate to institutional trust and support for right-wing populism. American Behavioral Scientist , 69 (4), 478–504. https://doi.org/10.1177/00027642241240418 Magaloni, B. (2006). Voting for Autocracy: Hegemonic Party Survival and Its Demise in Mexico . Cambridge University Press. Mason, L. (2018). Uncivil Agreement: How Politics Became Our Identity . University of Chicago Press. https://doi.org/10.7208/chicago/9780226524689.001.0001 Mitchell, J., Bohman, A., Eger, M. A., & Hjerm, M. (2025). Rally around the flag? Explaining changes in Swedish public opinion toward NATO membership after Russia’s invasion of Ukraine. Acta Sociologica , 68 (1), 30–40. https://doi.org/10.1177/00016993241268185 Morales, J. (2021). Legislating during war: Conflict and politics. Journal of Public Economics , 193 , 104325. https://doi.org/10.1016/j.jpubeco.2020.104325 Mueller, J. E. (1970). Presidential popularity from Truman to Johnson. American Political Science Review , 64 (1), 18–34. https://doi.org/10.2307/1955610 Mueller, J. E. (1973). War, Presidents, and Public Opinion . Wiley. Muhammad, M., Domantas, & Undzėnas (2025). Entangled fates—the rally effect around Europe due to Russia’s war of aggression against Ukraine. Humanities and Social Sciences Communications , 12 (1), 1–10. https://doi.org/10.1038/s41599-025-05138-7 Petrocik, J. R. (1996). Issue Ownership in Presidential Elections, with a 1980 Case Study. American Journal of Political Science , 40 (3), 825–850. https://doi.org/10.2307/2111797 Prior, M. (2007). Post-Broadcast Democracy: How Media Choice Increases Inequality in Political Involvement and Polarizes Elections . Cambridge University Press. RezaeeDaryakenari, B., Ghafouri, V., & Kasap, N. (2025). Who rallies round the flag? The impact of the US sanctions on Iranians’ attitude toward the government. Foreign Policy Analysis , 21 (1), orae033. https://doi.org/10.1093/fpa/orae033 Riker, W. H. (1986). The Art of Political Manipulation . Yale University Press. Rogoff, K. (1990). Equilibrium political budget cycles. American Economic Review , 80 (1), 21–36. Rožukalne, A., Kleinberga, V., Tīfentāle, A., & Strode, I. (2022). What is the flag we rally around? Trust in information sources at the outset of the COVID-19 pandemic in Latvia. Social Sciences , 11 (3), 123. https://doi.org/10.3390/socsci11030123 Safarpour, A., & Baum, M. A. (2022). Pandemic governors and public opinion: The effect of COVID-19 on support for U.S. governors. APSA Preprints. https://doi.org/10.33774/apsa-2022-vvgg3 Schedler, A. (2002). Elections without democracy: The menu of manipulation. Journal of Democracy , 13 (2), 36–50. https://doi.org/10.1353/jod.2002.0031 Schedler, A. (2013). The Politics of Uncertainty: Sustaining and Subverting Electoral Authoritarianism . Oxford University Press. Schraff, D. (2021). Political trust during the Covid-19 pandemic: Rally around the flag or lockdown effects? European Journal of Political Research , 60 (4), 1007–1017. https://doi.org/10.1111/1475-6765.12425 Shi, M., & Svensson, J. (2006). Political budget cycles: Do they differ across countries and why? Journal of Public Economics , 90 (8–9), 1367–1389. https://doi.org/10.1016/j.jpubeco.2005.09.009 Sides, J., & Vavreck, L. (2013). The Gamble: Choice and Chance in the 2012 Presidential Election . Princeton University Press. Sosa-Villagarcia, P., & Verónica, H. L. (2021). Covid-19 and presidential popularity in Latin America. Revista Latinoamericana de Opinión Pública , 10 (2), 71–91. https://doi.org/10.14201/rlop.23664 Steiner, N. D., Berlinschi, R., Farvaque, E., Fidrmuc, J., Harms, P., Mihailov, A., Neugart, M., & Stanek, P. (2023). Rallying around the EU flag: Russia's invasion of Ukraine and attitudes toward European integration. Journal of Common Market Studies , 61 (2), 283–301. https://doi.org/10.1111/jcms.13449 Stimson, J. A. (1999). Public Opinion in America: Moods, Cycles, and Swings . Westview. Stroud, N. J. (2011). Niche News: The Politics of News Choice . Oxford University Press. https://doi.org/10.1093/acprof:oso/9780199755509.001.0001 Stubager, R., and Rune Slothuus (2013). What Are the Sources of Political Parties’ Issue Ownership? Testing Four Explanations at the Individual Level. Political Behavior , 35 (3), 567–588. https://doi.org/10.1007/s11109-012-9204-2 Taber, C. S., & Milton Lodge (2006). Motivated skepticism in the evaluation of political beliefs. American Journal of Political Science , 50 (3), 755–769. https://doi.org/10.1111/j.1540-5907.2006.00214.x van Alebeek, C. R. A., Tom, W. G., van der Meer, & Hakhverdian, A. (2025). Stable or variable distrust? Disentangling the relationship between political trust and electoral behavior. European Political Science Review First View , 1–17. https://doi.org/10.1017/S1755773925100210 van der Brug, W. (2004). Issue Ownership and Party Choice. Electoral Studies , 23 (2), 209–233. https://doi.org/10.1016/S0261-3794(02)00061-6 van der Meer, T., & Steenvoorden, E., and Ebe Ouattara (2023). Fear and the COVID-19 Rally Round the Flag: A Panel Study on Political Trust. West European Politics , 46 (6), 1089–1105. https://doi.org/10.1080/01402382.2023.2171220 Verdier, D., and Bryan Woo (2011). Why rewards are better than sanctions: Economic statecraft and the rally around the flag effect. European Journal of Political Economy , 27 (2), 217–232. https://doi.org/10.1016/j.ejpoleco.2010.08.004 Walgrave, S., Lefevere, J., & Tresch, A. (2012). The associative dimension of issue ownership. Public Opinion Quarterly , 76 (4), 771–782. https://doi.org/10.1093/poq/nfs023 Zaller, J. R. (1992). The Nature and Origins of Mass Opinion . Cambridge University Press. https://doi.org/10.1017/CBO9780511818691 Additional Declarations No competing interests reported. Supplementary Files SequentialrallyeffectsONLINEAPPENDIX.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-8523383","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":583182606,"identity":"73e2bdb5-cb20-44a0-80bf-6f8ecbd6402e","order_by":0,"name":"Andrea Szabó","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAyklEQVRIiWNgGAWjYHACNgh1AIg/MCQwGJCkhXFGAqlamHmI0SLff/jZgw8Mdnl8tw8f/Gz7I43BXCIBvxaDA8fMDWcwJBdLnktLls5JyGGwnEFIC2ODmTQPA3PihjM8BkAtFQwGNwhokW9m/wbUUg/Uwv/5twUxWhiO8YBsOQyyhU2aAegwgloMzvCUG84wOJ448wybmWVPWhqPwZkHBBzWf3zbgw8V1Yl9Z5gf3/hhkyxncJyQwyB2IZg8xKgfBaNgFIyCUUAAAAANCkG51Wx2HwAAAABJRU5ErkJggg==","orcid":"","institution":"Eötvös Loránd University","correspondingAuthor":true,"prefix":"","firstName":"Andrea","middleName":"","lastName":"Szabó","suffix":""}],"badges":[],"createdAt":"2026-01-05 16:23:40","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-8523383/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-8523383/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":101664564,"identity":"b8030aee-ca78-442f-9474-074e4acd4b63","added_by":"auto","created_at":"2026-02-02 11:28:14","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":64646,"visible":true,"origin":"","legend":"\u003cp\u003esummarizes our hypotheses and empirical tests, and highlights crisis timing relative to the 2022 election (COVID onset: 11 March 2020; war onset: 24 February 2022).\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-8523383/v1/700db681b807406830830ae0.png"},{"id":101664563,"identity":"81546c32-d1a9-48ce-8e8e-c944c786fe7a","added_by":"auto","created_at":"2026-02-02 11:28:14","extension":"jpeg","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":266000,"visible":true,"origin":"","legend":"\u003cp\u003eBiweekly trends in incumbent and opposition support, 2018–2022\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNote.\u003c/strong\u003e The figure uses deduplicated poll-level observations from the Vox Populi archive. For each biweekly interval (indexed by its start date), lines plot \u003cstrong\u003esample-size–weighted means\u003c/strong\u003e of reported support (weights = poll \u003cem\u003en\u003c/em\u003e) for the mass electorate (“Összes”) and likely voters (“Biztos”). Panel A shows Fidesz support; Panel B shows opposition support. Vertical dashed lines mark the onset of the COVID-19 pandemic (11 March 2020) and Russia’s full-scale invasion of Ukraine (24 February 2022). Differences in levels between bases reflect the likely-voter definition, and the inclusion of undecided/nonvoters in the mass-electorate series. Where no estimate is available for a given base in a biweekly period, the corresponding series is shown with gaps.\u003c/p\u003e","description":"","filename":"floatimage2.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-8523383/v1/0d9e09da481fec05aa8e33ac.jpeg"},{"id":101664562,"identity":"76ba89a9-d034-4b11-b483-e0f80952e49c","added_by":"auto","created_at":"2026-02-02 11:28:14","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":24587,"visible":true,"origin":"","legend":"\u003cp\u003eCrisis-related vote motivations in the post-election survey (May 2022).\u003c/p\u003e\n\u003cp\u003eBars show the share of respondents citing the war or COVID-19 among their top-two closed-ended reasons for their list vote, by vote bloc (Fidesz vs joint opposition). War is a dominant and sharply polarized motive (55.9% vs 18.2%), whereas COVID-19 is rare overall and marginally more common among Fidesz voters (4.8% vs 0.4%).\u003c/p\u003e","description":"","filename":"floatimage3.png","url":"https://assets-eu.researchsquare.com/files/rs-8523383/v1/3226b37cc40672007e8f43fb.png"},{"id":106728571,"identity":"9ac53197-d931-4b05-b5af-6108f4485aba","added_by":"auto","created_at":"2026-04-12 18:43:18","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1755978,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-8523383/v1/c3e29be4-4ac4-44b6-8577-cdee447e7d81.pdf"},{"id":101754108,"identity":"039a256b-ecf2-41e6-a370-1d08b180543b","added_by":"auto","created_at":"2026-02-03 10:41:38","extension":"docx","order_by":0,"title":"","display":"","copyAsset":false,"role":"supplement","size":57781,"visible":true,"origin":"","legend":"","description":"","filename":"SequentialrallyeffectsONLINEAPPENDIX.docx","url":"https://assets-eu.researchsquare.com/files/rs-8523383/v1/0b90c2b11485fbe441267ad5.docx"}],"financialInterests":"No competing interests reported.","formattedTitle":"Rally Without a Surge? Sequential Crises and Asymmetric Mobilization in Polarized Contexts","fulltext":[{"header":"Introduction","content":"\u003cp\u003eMultiple crises rarely arrive one at a time. What happens when electorates confront distinct shocks within the same electoral cycle? Classic rally-around-the-flag scholarship (e.g., Mueller \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e1973\u003c/span\u003e; Brody \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e1991\u003c/span\u003e) predicts a visible—often temporary—surge in incumbent support as citizens “rally” behind national leadership. Yet sequential crises create an underexplored challenge for this logic: earlier shocks can condition how voters interpret later ones, and polarization can shift rally dynamics from a national swing to bloc-specific responses.\u003c/p\u003e \u003cp\u003eWe study Hungary’s 2018–2022 cycle, which includes two major shocks—the COVID-19 pandemic and the Russia–Ukraine war—within a single election cycle. The aggregate pattern is puzzling: incumbent support remains broadly stable, but the war onset coincides with a pronounced drop-in opposition support. This paper addresses a critical, yet underexplored, puzzle in contemporary political behavior: how do sequential crises in highly polarized democracies impact electoral dynamics when classic 'rally-around-the-flag' effects are seemingly absent or uneven? We theorize that crisis politics in such settings operate not through a uniform incumbent surge, but primarily via opposition-side contraction and bloc-owned framing.\u003c/p\u003e \u003cp\u003eWe address this puzzle by developing a bloc-structured theory of rally effects. In polarized contexts with segmented partisan publics and fragmented media ecosystems, crises are more likely to be translated into divergent interpretive frames and differential issue ownership than into a uniform national response (e.g., Petrocik \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e1996\u003c/span\u003e; Riker \u003cspan citationid=\"CR41\" class=\"CitationRef\"\u003e1986\u003c/span\u003e). A crisis can therefore consolidate turnout and vote readiness within one bloc while the other struggles with coordination, motivation, or cross-pressures (e.g., Sides and Vavreck \u003cspan citationid=\"CR49\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). When crises are sequential, the more election-proximate shock can also override or reframe the political consequences of earlier ones by shifting what is salient when choices are made.\u003c/p\u003e \u003cp\u003eEmpirically, we link aggregate patterns to individual-level mechanisms by combining poll-based interrupted time-series (ITS) models with three linked surveys: Autumn 2021 (pandemic experiences and turnout readiness), March 2022 (pre-election participation baseline), and May 2022 post-election (closed- and open-ended vote motivations). This design allows us to separate (i) aggregate discontinuities around crisis breakpoints from (ii) bloc-differentiated motivations and readiness that can generate stable-looking aggregates.\u003c/p\u003e \u003cp\u003eWe make three contributions. First, we advance a novel bloc-structured theory of rally effects, arguing that crises in polarized democracies generate selective, rather than uniform, responses, often manifesting as opposition contraction without an incumbent surge. Second, it theorizes and tests sequential-crisis dynamics, showing how an election-proximate shock can dominate the motivational environment even after a prolonged earlier crisis. Third, it demonstrates the value of combining poll-based ITS with linked surveys to disaggregate mechanisms that remain hidden in aggregate time series.\u003c/p\u003e \u003cp\u003eAlthough we study Hungary, the theoretical claim is general: in polarized systems with segmented partisan publics, crises are likely to be translated into bloc-owned frames, so electoral consequences may operate less through average preference shifts than through asymmetric mobilization, differential vote readiness, and opposition-side weakening. We treat Hungary as a hard, informative test because polarization is high and both shocks were salient within a single electoral cycle. This matters beyond Hungary because sequential crises are increasingly common, and a bloc-asymmetric lens clarifies when stable-looking aggregates can mask opposition vulnerability rather than incumbent surges.\u003c/p\u003e \u003cp\u003eHungary offers a hard test for sequential-crisis rally dynamics because partisan sorting and a polarized information environment in principle, should make nationwide opinion movements difficult to detect. Citizens receive and interpret crisis information through bloc-consistent media and elite cues, and crisis evaluations can be rapidly re-framed as ordinary partisan conflict. At the same time, the 2022 election provides a clear electoral endpoint and a rare combination of a prolonged public-health emergency followed by an acute external security shock close to the vote. If classic rally effects were to appear as aggregate incumbent surges, we should observe them here. If instead crises work primarily by shifting relative turnout readiness and by redistributing issue salience across blocs, then aggregate stability can coexist with decisive electoral consequences. This logic motivates our emphasis on between-bloc contrasts, short-run discontinuities, and survey-based evidence on crisis-related motivation and considered alternatives.\u003c/p\u003e \u003cp\u003eOur framework yields four testable hypotheses about how sequential crises shape electoral dynamics in polarized settings:\u003c/p\u003e\u003cdiv class=\"BlockQuote\"\u003e\u003cp\u003e(H1) Classic rally (baseline expectation): A broad-based rally would appear as a discrete upward shift in incumbent support at the onset of a major crisis, especially among likely voters, visible in aggregate polling trends.\u003c/p\u003e\u003cp\u003e(H2) Asymmetric opposition weakening: Under bloc-structured rally dynamics, crisis effects should be disproportionately visible on the opposition side. We expect downturns in opposition support and/or increases in opposition-side volatility around key crisis moments, even if incumbent support remains largely stable.\u003c/p\u003e\u003cp\u003e(H3) Bloc-structured framing and motivation: At the micro level, crisis-relevant frames should be bloc-owned. In 2022, we expect war-related considerations (security, peace) to feature disproportionately among government voters’ stated motivations, while pandemic experiences in late 2021 relate to turnout readiness mainly via pre-existing vote intentions rather than a uniform, crisis-induced mobilization across blocs.\u003c/p\u003e\u003cp\u003e(H4) Recency in sequential crises: When two major shocks occur within a single electoral cycle, the more proximate crisis should dominate election-proximate motivations and framing, indicating sequential salience rather than an additive national rally based on all past shocks.\u003c/p\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e \u003cp\u003eFigure \u003cspan refid=\"Fig1\" class=\"InternalRef\"\u003e1\u003c/span\u003e summarizes our hypotheses and empirical tests, and highlights crisis timing relative to the 2022 election (COVID onset: 11 March 2020; war onset: 24 February 2022).\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eNote\u003c/strong\u003e \u003c/p\u003e\u003cp\u003eThe diagram summarizes hypothesized pathways linking two crisis shocks (COVID-19 and the Russia–Ukraine war) to four mechanisms (incumbent boost, opposition weakening, reallocation, and frame ownership) and to the empirical tests used in the paper (poll ITS, event-window models, and three survey waves). Crisis timing relative to the 2022 election: COVID onset (11 March 2020; about 25 months before the election) and war onset (24 February 2022; about 5–6 weeks before the election).\u003c/p\u003e \u003cp\u003e\u003c/p\u003e \n\n \n\n \n\n \n\n \n\n \n\n "},{"header":"Theoretical framework","content":"\u003ch3\u003eA. Rally dynamics and heterogeneity under polarization\u003c/h3\u003e\u003cp\u003eWe conceptualize ‘rally-round-the-flag’ as a crisis-period shift in political support toward incumbent authorities and governing institutions. Classic accounts locate the mechanism in heightened perceived threat, reduced partisan contestation, and citizens’ incentives to signal unity in the face of an external shock (Mueller \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e1970\u003c/span\u003e; Brody \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e1991\u003c/span\u003e; Baum \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Baker and Oneal \u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2001\u003c/span\u003e). In contemporary democracies, however, rally-effects are typically brief and conditional: early increases in trust or approval often fade as the shock becomes politicized and as policy costs become salient (Schraff \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; van der Meer \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Devine \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eRally responses are also heterogeneous across partisan publics. Who rallies depends on elite cues, partisan trust, and the informational environment, a phenomenon implying that aggregate stability can mask substantial bloc-level movement (Edwards and Swenson \u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e1997\u003c/span\u003e; Hetherington and Nelson \u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e2003\u003c/span\u003e; Hegewald and Schraff 2024; Prior \u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e2007\u003c/span\u003e; Stroud \u003cspan citationid=\"CR53\" class=\"CitationRef\"\u003e2011\u003c/span\u003e; Levendusky \u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e2013\u003c/span\u003e; Druckman, Peterson, and Slothuus \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2013\u003c/span\u003e). Where polarization is high and governing parties dominate agenda setting and media access, opposition identifiers may not “join” the rally; instead, they may respond with skepticism, disengagement, or delayed electoral decision-making, while incumbent identifiers consolidate (Iyengar, Sood, and Lelkes 2012; Mason \u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e2018\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eThis bloc-structured view motivates a reconceptualization of what constitutes rally evidence as rally evidence in electoral contexts. Rather than expecting a symmetric, society-wide surge for incumbents, we focus on selective reinforcement of the incumbent bloc and relative opposition-side weakening within the likely electorate. These dynamics matter most near elections, when small changes in turnout readiness and late-cycle justification can become electorally decisive.\u003c/p\u003e\u003cp\u003eMore broadly, heterogeneity in crisis responses is consistent with work on the conditioning role of political trust and macropolitical public mood in shaping evaluations of incumbents (Hetherington and Rudolph \u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e2015\u003c/span\u003e; Stimson \u003cspan citationid=\"CR52\" class=\"CitationRef\"\u003e1999\u003c/span\u003e). Building on this, we theorize that in highly polarized contexts, the 'rally-round-the-flag' effect transforms from a symmetrical, society-wide phenomenon into an asymmetric, bloc-specific redistribution of political support. This shift implies that incumbent gains might be subtle or absent at the aggregate level, while underlying mechanisms produce critical changes in partisan readiness, motivation, and issue framing.\u003c/p\u003e\u003ch2\u003eB. COVID-19 as a prolonged crisis: normalization, performance conflict, and turnout readiness\u003c/h2\u003e\u003cp\u003eCOVID-19 differs from canonical short international crises because it is prolonged, policy-intensive, and distributive. Early pandemic phases often produced increases in political trust and executive support, consistent with a classic rally logic (Bol et al. \u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Kritzinger et al. \u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Schraff \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2021\u003c/span\u003e). Yet as restrictions persist and costs concentrate, COVID governance becomes a site of partisan contestation. Across countries, the initial rally frequently attenuated or reversed as evaluations of competence, fairness, and economic management diverged across partisan publics (van der Meer \u003cspan citationid=\"CR58\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Colloca et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eIn election periods, prolonged crises can reshape participation readiness rather than simply shifting vote choice. For the incumbent bloc, crisis management can be integrated into competence and stability narratives. For challengers, persistent crisis governance may depress efficacy, reduce the perceived payoff of opposition voting, or exacerbate coordination problems, especially in polarized settings where distrust is high and alternative crisis programs are hard to communicate credibly (Brouard and Michel 2025). We therefore treat COVID as a shock that can reorder the likely electorate by changing turnout intentions and late-cycle engagement.\u003c/p\u003e\u003cp\u003eCross-national evidence from the pandemic likewise points to short-lived or conditional rally effects, often strongest early on and fading as performance evaluations and partisan conflict intensify (Bækgaard et al. \u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Johansson et al. \u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Safarpour and Baum \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Sosa-Villagarcia and Hurtado Lozada 2021; Lytkina and Reeskens 2024; Mitchell et al. \u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e2025\u003c/span\u003e).\u003c/p\u003e\u003ch3\u003eC. External security shocks, framing advantage, and issue ownership\u003c/h3\u003e\u003cp\u003eExternal security crises more closely match canonical rally conditions: they are sudden, salient, and tied to national sovereignty and physical safety. Such events can create a temporary unity premium for incumbents, while penalizing challengers who are portrayed as risky, inexperienced, or destabilizing (Baum \u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Chávez and Wright 2022). Evidence from the Russia–Ukraine war likewise suggests that security shocks can generate rally dynamics, sometimes around national leaders and sometimes around supranational actors (Steiner et al. \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2023\u003c/span\u003e; Kizilova \u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e2024\u003c/span\u003e; Devine and Valgarðsson 2024).\u003c/p\u003e\u003cp\u003eWe connect these dynamics to issue ownership: parties benefit electorally when voters perceive them as more competent to “handle” a salient issue, and crises can sharpen or activate those competence reputations (Petrocik \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e1996\u003c/span\u003e; Bélanger and Meguid \u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e2008\u003c/span\u003e; Walgrave, Lefevere, and Tresch \u003cspan citationid=\"CR60\" class=\"CitationRef\"\u003e2012\u003c/span\u003e; van der Brug \u003cspan citationid=\"CR57\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Stubager and Slothuus 2013). In Hungary’s 2022 campaign, the governing party was positioned to claim peace and security ownership by coupling sovereignty messaging with risk framing of the opposition. This implies that war-related motivations and post-hoc justifications should be disproportionately salient within the incumbent electorate, even if aggregate vote shares do not exhibit a large surge.\u003c/p\u003e\u003cp\u003eEvidence from external-security and military contexts shows that rally dynamics can depend on casualties, elite framing, and the distributional politics of sanctions and war—producing asymmetric reactions across partisan publics (Lai and Reiter 2005; Kuijpers \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Verdier and Woo 2011; Frye \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Morales \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; RezaeeDaryakenari et al. \u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Muhammad and Undzėnas 2025; Rožukalne et al. \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Grechanaya and Ceron \u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003ch3\u003eD. Sequential crises and four testable expectations\u003c/h3\u003e\u003cp\u003eWhen major shocks occur sequentially, recency and salience imply that the later shock can re-weight which frames dominate late-cycle decision-making and participation readiness. The earlier crisis may still structure identities and trust, but its mobilizing force can attenuate as citizens adapt and as competing grievances accumulate (Zaller \u003cspan citationid=\"CR61\" class=\"CitationRef\"\u003e1992\u003c/span\u003e; Iyengar and Kinder \u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e1987\u003c/span\u003e; Schraff \u003cspan citationid=\"CR47\" class=\"CitationRef\"\u003e2021\u003c/span\u003e; Colloca et al. \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2024\u003c/span\u003e).\u003c/p\u003e\u003cp\u003eApplied to Hungary’s 2022 election cycle, this framework yields four expectations summarized in Table\u0026nbsp;\u003cspan refid=\"Tab1\" class=\"InternalRef\"\u003e1\u003c/span\u003e. We expect limited aggregate rally but a bloc-structured pattern in which COVID primarily reshapes turnout readiness and the composition of the likely electorate, while the war shock selectively reinforces incumbent frames and justifications. The strongest electoral signature should therefore appear as relative opposition-side weakening among likely voters rather than as a symmetric national surge.\u003c/p\u003e\u003cp\u003eRecent work on crisis sequencing and the stability versus volatility of distrust underscores that political reactions may depend on how new shocks re-activate existing grievances and identities rather than simply adding to a uniform national response (van Alebeek et al. \u003cspan citationid=\"CR56\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Frateur et al. \u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e2025\u003c/span\u003e; Leininger and Schaub 2024).\u003c/p\u003e"},{"header":"Research Design and Empirical Strategy","content":"\u003ch2\u003eOverview: four complementary empirical approaches\u003c/h2\u003e\u003cp\u003ePoll-based interrupted time-series (ITS), 2018–2022: Using all publicly available national vote-intention polls archived by Vox Populi (\u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://kozvelemeny.org/\u003c/span\u003e\u003cspan address=\"https://kozvelemeny.org/\" targettype=\"URL\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e), we estimate segmented regressions for Fidesz and for the unified opposition. This macro lens tests whether either crisis coincides with a discontinuity in party support, particularly among likely voters— once pollster house effects and survey mode are accounted for.\u003c/p\u003e\u003cp\u003eAutumn 2021 survey (N ≈ 5,000): We measure pandemic experiences and perceived societal impact, and link these to turnout intention to assess whether COVID-19 was associated with differential participation readiness across partisan blocs before the war shock.\u003c/p\u003e\u003cp\u003eMarch 2022 pre-election survey (N = 1,000): We capture baseline engagement and political information seeking to contextualize whether the campaign environment already reflected asymmetric mobilization and attention patterns.\u003c/p\u003e\u003cp\u003eMay 2022 post-election survey (N = 1,000): We measure crisis-related vote motivations and open-ended justifications to assess bloc-specific crisis frame ownership in the immediate electoral aftermath.\u003c/p\u003e\u003cp\u003eOur empirical strategy is designed to differentiate aggregate stability from bloc-level reallocation within a sequential-crisis electoral cycle. This is achieved by integrating (i) poll-based interrupted time-series (ITS) estimates, which identify discontinuities in support trajectories around pre-specified crisis onsets (controlling for pollster house effects and secular trends), with three distinct probability-based surveys. These surveys include: (ii) an Autumn 2021 survey capturing late-pandemic experiences and perceived societal impact linked to turnout intention; (iii) a March 2022 pre-election survey assessing baseline engagement and information-seeking behavior; and (iv) a May 2022 post-election survey designed to elicit vote motives and open-ended justifications for evaluating crisis frame ownership and within-bloc volatility indicators.\u003c/p\u003e\u003cp\u003eEach component plays a distinct inferential role, with the ITS providing macro-level evidence and the surveys collectively benchmarking attitudes (Autumn 2021), turnout readiness (March 2022), and crisis-related vote motivations/frames (May 2022). We treat each component as a partial test, prioritizing the convergence of findings across them rather than relying solely on any single estimate.\u003c/p\u003e\u003cp\u003e \u003c/p\u003e\u003cdiv class=\"gridtable\"\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e\u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e\u003ctable float=\"Yes\" id=\"Tab1\" border=\"1\"\u003e\u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 1\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eHypotheses and empirical tests\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e\u003ccolgroup cols=\"6\"\u003e\u003c/colgroup\u003e\u003cthead\u003e\u003ctr\u003e\u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eHypothesis\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eEmpirical test\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eData\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eOutcome(s)\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003eMain specification\u003c/p\u003e \u003c/th\u003e\u003cth align=\"left\" colname=\"c6\"\u003e \u003cp\u003eWhere reported\u003c/p\u003e \u003c/th\u003e\u003c/tr\u003e\u003c/thead\u003e\u003ctbody\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH1 (Classic rally): crisis onset increases incumbent support\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePoll time series\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2018–2022\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eFidesz support; opposition support\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eLevel + slope change ITS with controls\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMain text; Fig.\u0026nbsp;\u003cspan refid=\"Fig2\" class=\"InternalRef\"\u003e2\u003c/span\u003e; Tables\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e; App. A1\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH2 (Asymmetric rally): crisis effects appear as opposition weakening more than incumbent surge\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eAutumn 2021 \u0026amp; March 2022 surveys\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e2021 (Aut)\u003c/p\u003e \u003cp\u003e2022 (Mar)\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eTurnout readiness and engagement baseline\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eAutumn 2021: pandemic experiences and turnout intention. March 2022: bloc gaps in engagement and participation track political interest/information seeking\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMain text; App. A3–A4\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH3 (Bloc-owned war framing): war motives concentrated among government voters\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMay 2022 post‑election survey\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMay 2022\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eWar/COVID vote motives (top‑two) and crisis frames in open‑ended responses\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eBloc contrasts in war/COVID motivations (top‑two) and open‑ended crisis framing; any‑source measures\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMain text; Fig.\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e; App. A7\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003ctr\u003e\u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eH4 (Between-bloc, not within-opposition sorting): war motive does not predict opposition volatility\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eMay 2022 post‑election survey\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eMay 2022\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eDecision timing, campaign instability, and cross‑pressures\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003eBloc differences in decision timing and campaign volatility; within‑opposition associations with war motives; multivariate robustness checks\u003c/p\u003e \u003c/td\u003e\u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eMain text; App. A6–A9\u003c/p\u003e \u003c/td\u003e\u003c/tr\u003e\u003c/tbody\u003e\u003c/table\u003e\u003c/div\u003e\u003cp\u003e\u003c/p\u003e\u003ch2\u003eData source and unit of analysis\u003c/h2\u003e\u003cp\u003eWe analyze two electorates when available. The primary series uses pollsters’ likely-voter or high-propensity filters, which best match the theoretical focus on participation readiness in election-proximate crises. As a complement, we also report pulse-window tests in the mass electorate to assess whether crisis effects are confined to likely voters or appear more broadly.\u003c/p\u003e\u003cp\u003eOur macro-level dataset draws on the Vox Populi archive of Hungarian national voting-intention polls (January 2018–April 2022). Each observation corresponds to a unique poll result for a specific party/list from a specific polling organization at a specific fieldwork period.\u003c/p\u003e\u003ch3\u003eOutcomes\u003c/h3\u003e\u003cp\u003eAll analyses treat the poll as the unit of observation. In figures we also display biweekly, sample-size–weighted averages for readability, but statistical estimates are based on poll-level observations.\u003c/p\u003e\u003cp\u003eOur primary outcomes are poll-level support shares for (i) the incumbent Fidesz–KDNP alliance and (ii) the unified opposition list. We model these as separate dependent variables to allow for asymmetric reactions: a crisis can weaken the opposition without producing a commensurate incumbent surge in the aggregate.\u003c/p\u003e\u003ch3\u003eCrisis breakpoints and time trends\u003c/h3\u003e\u003cp\u003eBecause rally dynamics may be short-lived rather than permanent, we complement the segmented models with pulse-window specifications that estimate average deviations in defined windows around each breakpoint. This approach provides a direct check on whether any effects are concentrated in the weeks immediately following crisis onset.\u003c/p\u003e\u003cp\u003eOur baseline ITS specification is a segmented regression with a pre-crisis time trend, post-crisis level indicators, and post-crisis slope changes for each breakpoint. This formulation captures both abrupt discontinuities and medium-run trajectory shifts.\u003c/p\u003e\u003cp\u003eWe define two crisis breakpoints ex ante. COVID onset is March 11, 2020, the date the Hungarian government declared a state of emergency. War onset is February 24, 2022, the date of Russia’s full-scale invasion of Ukraine.\u003c/p\u003e\u003ch2\u003eControls and robustness\u003c/h2\u003e\u003cp\u003eAdditional robustness checks (alternative weighting, alternative breakpoint codings, survey-mode controls where available, and window-length sensitivity) are reported in the Online Appendix.\u003c/p\u003e\u003cp\u003eWe model the data as a pollster-by-time-period panel of poll-level observations and estimate weighted linear mixed models with pollster fixed effects (house effects). Serial dependence is handled by specifying correlated within-pollster residuals over the time index: our main specification uses an AR(1) structure, and we assess robustness using an ARMA(1,1) structure. We treat convergence in sign and magnitude across these error processes—rather than any single p-value—as the relevant evidentiary standard.\u003c/p\u003e\u003cp\u003eTo reduce bias from differences in question wording, sampling, and mode across pollsters, all ITS models include pollster fixed effects (house effects). We estimate both unweighted specifications and precision-weighted specifications that prioritize polls with larger effective sample sizes, with trimming to limit undue influence from extreme weights.\u003c/p\u003e\u003cp\u003eFinally, we interpret the polling models as evidence about short-run shifts in support and turnout-proxy composition within a single case, not as definitive causal estimates isolated from all concurrent political dynamics. For that reason, we complement segmented (level-and-slope) specifications with short pulse-window models around each breakpoint and focus on (a) effect sizes and confidence intervals, (b) direction-of-change consistency across specifications, and (c) contrasts between incumbent and opposition blocs that are less sensitive to slow-moving confounders.\u003c/p\u003e\u003cp\u003e \u003cb\u003eIdentification and interpretive scope.\u003c/b\u003e \u003c/p\u003e\u003cp\u003eOur empirical strategy leverages the sharp timing of two widely salient shocks, with breakpoints defined ex ante based on institutional dates (the first declared state of emergency for COVID-19 and the day of the Russian invasion of Ukraine). While crisis timing is plausibly exogenous to short-run polling dynamics, other processes—campaign events, economic news, or strategic coordination—can unfold in parallel. We therefore combine several safeguards: pollster fixed effects to absorb stable house effects; time trends and breakpoint-specific trend terms to separate discontinuities from gradual movements; and serial-correlation checks (Online Appendix Table A2) to assess whether inference depends on residual dependence assumptions.\u003c/p\u003e\u003cp\u003eIdentification in this setting rests on timing rather than random assignment. We pre-specify breakpoints tied to widely recognized events and focus on whether the series exhibits discrete discontinuities and/or slope changes at those points. Importantly, the two shocks occur in distinct political environments—COVID early in the cycle and the war during an already campaign-like period—which provides additional leverage: if the models were merely capturing generic polling volatility, we would expect similar level shifts for both blocs across shocks. Instead, we assess whether discontinuities are bloc-specific and whether post-shock slopes compensate. We also treat the likely-voter series as a conservative test because it reduces compositional changes driven by turnout intention, and we weight by trimmed sample size to reflect measurement precision. Finally, we interpret the polling evidence alongside survey measures of crisis perceptions and vote motivations.\u003c/p\u003e\u003ch2\u003eSurvey data and measures\u003c/h2\u003e\u003cp\u003eOur micro-level evidence comes from three probability-based national surveys fielded by the same survey program and post-stratified to population benchmarks. Together, they allow us to (i) gauge whether COVID-19 was associated with differential turnout readiness before the war shock, (ii) benchmark pre-election engagement patterns, and (iii) identify crisis frame ownership in voters’ reported motives and open-ended justifications.\u003c/p\u003e\u003ch2\u003eAutumn 2021 survey (N ≈ 5,000; COVID experiences and turnout intention)\u003c/h2\u003e\u003cp\u003eOur main outcome in this component is turnout intention, coded as a binary indicator for reporting high likelihood of voting. We use these measures to evaluate whether pandemic experiences were linked to participation readiness in a bloc-asymmetric manner.\u003c/p\u003e\u003cp\u003eFrom the Autumn 2021 face-to-face survey, we construct two indices that capture COVID-19 exposure and perceptions. A “personal impact” index summarizes respondents’ reported household/health exposure and perceived personal threat, whereas a “societal impact” index summarizes perceived consequences for the economy and society.\u003c/p\u003e\u003ch2\u003eMarch 2022 pre-election survey (N ≈ 1000, baseline engagement)\u003c/h2\u003e\u003cp\u003eThe March 2022 pre-election survey provides a baseline snapshot of political engagement shortly after the war shock and before the election. We construct a participation repertoire index (count of reported political activities) and measures of political information seeking to benchmark cross-bloc differences in attention and readiness.\u003c/p\u003e\u003ch2\u003eMay 2022 post-election survey (N ≈ 1000, vote motivations and crisis frames)\u003c/h2\u003e\u003cp\u003eTo probe whether the bloc-structured mechanism manifests as opposition-side volatility or demobilization signals, we operationalize four indicators: late deciding, campaign instability, considering an alternative list, and split-ticket voting. These are analyzed descriptively by bloc and then in weighted logistic models as robustness checks.\u003c/p\u003e\u003cp\u003eFor closed-ended motives, we create binary indicators for whether respondents selected war/peace/security and whether they selected COVID-related considerations among their top two reasons for vote choice. For open-ended responses, we code mentions of war/Ukraine/security/peace and other major categories to capture spontaneous framing.\u003c/p\u003e\u003cp\u003eThe May 2022 post-election survey is used to assess crisis framing and frame ownership. We draw on two sources: closed-ended top-two vote motives and open-ended vote justifications recorded verbatim and coded into substantive categories (see Online Appendix for codebook and procedures).\u003c/p\u003e\u003ch2\u003eMultivariable strategy and reporting\u003c/h2\u003e\u003cp\u003eImportantly, these regressions are not treated as causal identification devices. Their primary role is to ascertain whether the descriptive bloc asymmetries that motivate our argument are robust to plausible compositional differences across partisan publics.\u003c/p\u003e\u003cp\u003eThe main text prioritizes transparent descriptive contrasts and simple hypothesis-linked tests. Where multivariable adjustment is informative, we estimate weighted logistic regressions to verify that bloc differences in crisis framing, and volatility indicators persist when accounting for standard covariates (demographics and baseline political orientation measures).\u003c/p\u003e"},{"header":"Results","content":"\u003cdiv id=\"Sec18\" class=\"Section2\"\u003e \u003ch2\u003ePolling time series among likely voters\u003c/h2\u003e \u003cp\u003eWe begin by testing whether either crisis produced the classic aggregate \u0026ldquo;rally-round-the-flag\u0026rdquo; pattern\u0026mdash;a discrete, nationwide increase in incumbent support at crisis onset (H1). Using deduplicated poll-level observations for likely voters (\u0026ldquo;Biztos\u0026rdquo;, N\u0026thinsp;=\u0026thinsp;226) from the Vox Populi archive, we estimate segmented interrupted time-series (ITS) models with breakpoints at the start of the COVID-19 pandemic (11 March 2020) and Russia\u0026rsquo;s full-scale invasion of Ukraine (24 February 2022). Notably, the war shock (24 February 2022) occurred less than two weeks after the official campaign began (12 February 2022), and after the election date had been called on 11 January 2022\u0026mdash;effectively placing the shock inside an already campaign-like information environment. All specifications include pollster fixed effects (house effects) and, where available, survey-mode controls.\u003c/p\u003e \u003cp\u003eIn these segmented ITS models, the post-crisis indicator captures an immediate level shift at the breakpoint (a discontinuity), while the post-crisis time term captures a change in slope relative to the pre-crisis trend. Because political shocks can generate brief \u0026ldquo;bumps\u0026rdquo; rather than durable re-alignments, we treat short-run level shifts (and, in the next section, pulse-window estimates) as the most direct evidence of rally-like reactions, and we interpret longer-run trend changes more cautiously. Across specifications we report unstandardized coefficients in percentage points and emphasize substantive magnitudes and uncertainty (confidence intervals) rather than significance alone; sensitivity to serial-correlation assumptions is summarized below and documented in the Online Appendix.\u003c/p\u003e \u003cp\u003eTable\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e reports segmented ITS estimates for Fidesz support among likely voters. Neither crisis onset is associated with a discrete increase: the COVID-19 level change is small and statistically indistinguishable from zero (B\u0026thinsp;=\u0026thinsp;0.549, p\u0026thinsp;=\u0026thinsp;0.612), and the war-onset level change is essentially zero (B\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.014, p\u0026thinsp;=\u0026thinsp;0.996). The pre-crisis trend is slightly negative (B\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;0.007 per day, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001), with only weak evidence that it attenuates after COVID (post-COVID slope change B\u0026thinsp;=\u0026thinsp;0.004, p\u0026thinsp;=\u0026thinsp;0.090). Overall, the likely-voter series provides little support for a classic, nationwide aggregate rally (H1).\u003c/p\u003e \u003cp\u003eOpposition support follows a different trajectory. In Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e, the baseline specification implies an immediate post-invasion drop of roughly 4\u0026frac12; percentage points (B\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;4.449, p\u0026thinsp;=\u0026thinsp;0.065), whereas the positive post-COVID drift is interrupted at the war breakpoint. We interpret this pattern as evidence consistent with opposition-side weakening (H2), rather than as an incumbent surge. Because these conclusions could be sensitive to time-series dependence, we next assess robustness to alternative residual-correlation structures. These ITS estimates speak to aggregate polling dynamics, not individual-level opinion change; they therefore cannot adjudicate micro-level mechanisms directly. Null average level shifts also do not rule out offsetting subgroup movements (e.g., compositional reweighting) that net out at the national aggregate.\u003c/p\u003e \u003cp\u003eStepping back, the segmented ITS results already differentiate between a classic incumbent rally and bloc-asymmetric adjustment. For Fidesz, neither crisis produces a clear level jump, and post-crisis slope changes are modest, consistent with a relatively stable incumbent baseline among likely voters. For the opposition, by contrast, COVID is associated mainly with gradual drift, whereas the war shock is concentrated in an immediate downward discontinuity. This pattern is precisely what our sequential framework highlights: crises can matter electorally through opposition contraction even when the incumbent does not visibly surge. The remaining analyses therefore ask whether this inference survives conservative residual structures (Online Appendix Table A2) and whether short-run pulse windows in the mass electorate show any transient incumbent uptick.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab2\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 2\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eInterrupted time-series models (likely voters / \u0026ldquo;Biztos\u003cem\u003e\u0026rdquo;\u003c/em\u003e).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"5\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"char\" char=\".\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePredictor\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003eFidesz (B (SE))\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003ep\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eOpposition (B (SE))\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c5\"\u003e \u003cp\u003ep\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime trend (t)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.007 (0.002)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.000 (0.002)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.804\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCOVID level change (post_covid01)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.549 (1.082)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.612\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.486 (0.903)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.591\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePost-COVID slope change (t_covid)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.004 (0.003)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.090\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.008 (0.002)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWar level change (post_war01)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e-0.014 (2.871)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.996\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-4.449 (2.396)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.065\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePost-war slope change (t_war)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.044 (0.110)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c3\"\u003e \u003cp\u003e0.693\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.051 (0.092)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"char\" char=\".\" colname=\"c5\"\u003e \u003cp\u003e0.579\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel statistics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eN\u0026thinsp;=\u0026thinsp;226; R\u0026sup2;=0.612 (Adj. 0.579)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eN\u0026thinsp;=\u0026thinsp;226; R\u0026sup2;=0.529\u003c/p\u003e \u003cp\u003e (Adj. 0.488)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"5\"\u003e\u003cem\u003eNote.\u003c/em\u003e Unstandardized coefficients (percentage points) with standard errors in parentheses. Models include pollster fixed effects and survey-mode controls where available (mode controls are omitted from Online Appendix Table A2 for ensuring comparability across covariance specifications).\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec19\" class=\"Section2\"\u003e \u003ch2\u003eSerial-correlation sensitivity\u003c/h2\u003e \u003cp\u003eOnline Appendix Table A2 re-estimates the segmented ITS models using SPSS MIXED, allowing for within-pollster serial dependence and precision weights based on trimmed sample size (REGWGT\u0026thinsp;=\u0026thinsp;n_trim). Under an AR(1) residual structure, the estimated autocorrelation is small and not statistically distinguishable from zero (Fidesz: ρ\u0026thinsp;=\u0026thinsp;0.070, p\u0026thinsp;=\u0026thinsp;0.622; Opposition: ρ\u0026thinsp;=\u0026thinsp;0.152, p\u0026thinsp;=\u0026thinsp;0.346), and the substantive conclusions match the main specification. In particular, the war-onset shift remains near zero for Fidesz and remains a negative discontinuity for the opposition (AR(1): B\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;4.758, p\u0026thinsp;=\u0026thinsp;0.030).\u003c/p\u003e \u003cp\u003eUnder ARMA(1,1), autocorrelation is absorbed by a more flexible process (e.g., ρ\u0026thinsp;\u0026asymp;\u0026thinsp;0.94), and the estimated war-onset shift in the opposition series is attenuated and only marginal (B\u0026thinsp;=\u0026thinsp;\u0026minus;\u0026thinsp;3.570, p\u0026thinsp;=\u0026thinsp;0.082). Overall, serial-correlation checks yield small, non-significant AR(1) estimates and do not alter the substantive conclusions; the more conservative ARMA(1,1) specification attenuates the war-level discontinuity for the opposition (Online Appendix Table A2).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec20\" class=\"Section2\"\u003e \u003ch2\u003eOpposition-side discontinuity around the war\u003c/h2\u003e \u003cp\u003eOpposition dynamics sharpen this asymmetry. Opposition support rises gradually during the pandemic period (post-COVID slope change\u0026thinsp;\u0026asymp;\u0026thinsp;0.008 pp/day, p\u0026thinsp;\u0026lt;\u0026thinsp;0.001) but drops immediately after the invasion (war level change\u0026thinsp;\u0026asymp;\u0026thinsp;\u0026minus;\u0026thinsp;4.5 to \u0026minus;\u0026thinsp;4.8 pp across Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e and Online Appendix Table A2). Across specifications there is little evidence of a compensating post-war slope change that would rapidly restore the pre-war trajectory.\u003c/p\u003e \u003cp\u003eIn sum, crisis politics are more visible as an opposition-side contraction than as abrupt gains in the incumbent series. This pattern is consistent with our sequential, bloc-structured account: prolonged COVID does not generate a clean rally, while the acute war shock coincides with a discrete weakening on the opposition side with little sign of an incumbent surge. We next examine whether these short-run dynamics differ between likely voters and the mass electorate using pulse-window models.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec21\" class=\"Section2\"\u003e \u003ch2\u003eShort-run pulse windows and voter-based heterogeneity\u003c/h2\u003e \u003cp\u003eTo distinguish short-lived \u0026ldquo;bumps\u0026rdquo; from sustained shifts and to examine heterogeneity across voter bases, we next estimate non-overlapping pulse-window models in the mass electorate (\u0026ldquo;\u0026Ouml;sszes\u0026rdquo;). These models focus on narrow windows around each crisis onset and provide a direct test of brief rally-like responses.\u003c/p\u003e \u003cp\u003eFor Fidesz, pulse-window estimates in the mass electorate suggest at most a modest short-run uptick around COVID onset, with no comparable short-run increase around the war; effects are weaker among likely voters. For the opposition, event windows point to declines, with larger point estimates around the war than around COVID\u0026mdash;again consistent with war-timed opposition-side contraction.\u003c/p\u003e \u003cp\u003eComparing electorates reinforces this interpretation: the incumbent series is comparatively stable among committed voters, while opposition losses around the war are clearly visible in the mass electorate and appear as a war-timed discontinuity in the likely-voter segmented models (Table\u0026nbsp;\u003cspan refid=\"Tab2\" class=\"InternalRef\"\u003e2\u003c/span\u003e; Online Appendix Table A2).\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab3\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 3\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003ePulse-window models around COVID onset and war shock (mass electorate vs likely voters).\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"7\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c5\" colnum=\"5\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c6\" colnum=\"6\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c7\" colnum=\"7\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e\u0026nbsp;\u003c/th\u003e \u003cth align=\"left\" colspan=\"4\" nameend=\"c5\" namest=\"c2\"\u003e \u003cp\u003eMass electorate (\u0026ldquo;\u0026Ouml;sszes\u0026rdquo;, N\u0026thinsp;=\u0026thinsp;250)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colspan=\"2\" nameend=\"c7\" namest=\"c6\"\u003e \u003cp\u003eLikely voters (\u0026ldquo;Biztos\u0026rdquo;, N\u0026thinsp;=\u0026thinsp;226) \u0026mdash; incumbent series (Fidesz)\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003e\u003cem\u003ePredictor\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u003cem\u003eFidesz (B (SE))\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e\u003cem\u003eOpposition (B (SE))\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e\u003cem\u003eFidesz (B (SE))\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u003cem\u003ep\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eTime trend (t)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.004 (0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.009 (0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e-0.004 (0.001)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e\u0026lt;\u0026thinsp;0.001\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCOVID 0\u0026ndash;90 days (covid_0_90)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.419 (0.901)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.642\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-3.084 (1.058)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.004\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.912 (1.117)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.088\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eCOVID 91\u0026ndash;180 days (covid_91_180)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e1.825 (0.868)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.037\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-3.018 (1.019)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.003\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e1.368 (1.183)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.249\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eWar level change (post_war01)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e2.336 (1.412)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.099\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e-4.219 (1.658)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.012\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.954 (2.811)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.735\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePost-war slope change (t_war)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003e0.046 (0.060)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e0.443\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003e0.079 (0.071)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e \u003cp\u003e0.264\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003e0.034 (0.110)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e \u003cp\u003e0.756\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eModel statistics\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c5\"\u003e\u0026nbsp;\u003c/td\u003e \u003ctd align=\"left\" colname=\"c6\"\u003e \u003cp\u003eR\u0026sup2;=.613\u003c/p\u003e \u003cp\u003e (Adj. 0.580)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c7\"\u003e\u0026nbsp;\u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003ctfoot\u003e \u003ctr\u003e\u003ctd colspan=\"7\"\u003e\u003cem\u003eNote.\u003c/em\u003e Models use pollster fixed effects and mode controls; time is indexed from COVID onset. Full diagnostics are reported in the Online Appendix.\u003c/td\u003e\u003c/tr\u003e \u003c/tfoot\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003e \u003cstrong\u003eNote\u003c/strong\u003e \u003cp\u003eThe figure uses deduplicated poll-level observations from the Vox Populi archive. For each biweekly interval (indexed by its start date), lines plot sample-size\u0026ndash;weighted means of reported support (weights\u0026thinsp;=\u0026thinsp;poll \u003cem\u003en\u003c/em\u003e) for the mass electorate (\u0026ldquo;\u0026Ouml;sszes\u0026rdquo;) and likely voters (\u0026ldquo;Biztos\u0026rdquo;). Panel A shows Fidesz support; Panel B shows opposition support. Vertical dashed lines mark the onset of the COVID-19 pandemic (11 March 2020) and Russia\u0026rsquo;s full-scale invasion of Ukraine (24 February 2022). Differences in levels between bases reflect the likely-voter definition, and the inclusion of undecided/nonvoters in the mass-electorate series. Where no estimate is available for a given base in a biweekly period, the corresponding series is shown with gaps.\u003c/p\u003e \u003c/p\u003e \u003cp\u003eThe next section uses original survey data to probe the individual-level perceptions and motivations that can generate these bloc-asymmetric aggregate movements.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec22\" class=\"Section2\"\u003e \u003ch2\u003eFrom Polls to People: Survey Evidence\u003c/h2\u003e \u003cp\u003ePoll-level ITS models describe aggregate discontinuities but cannot by themselves identify the micro-level pathways\u0026mdash;such as turnout readiness, decision timing, or bloc-owned interpretive frames\u0026mdash;that could generate similar macro-patterns. We therefore triangulate the polling evidence with three linked surveys to assess (i) whether crisis experiences map onto vote readiness and (ii) whether the war is translated into partisan-bloc-owned motivations close to the election.\u003c/p\u003e \u003cp\u003e \u003cul\u003e \u003cli\u003e \u003cp\u003eThe survey evidence is designed to adjudicate between a classic nationwide rally and an asymmetric mechanism: muted movement in the incumbent\u0026rsquo;s aggregate support paired with opposition vulnerability and bloc-owned crisis frames.\u003c/p\u003e \u003c/li\u003e \u003cli\u003e \u003cp\u003eWe combine an Autumn 2021 module on COVID experiences and turnout intention, a March 2022 pre-election module on engagement and participation, and a May 2022 post-election module measuring stated vote motivations (closed-ended \u0026ldquo;top-two reasons\u0026rdquo; and open-ended justifications). Together, these data connect aggregate discontinuities to turnout readiness, decision timing, and crisis framing.\u003c/p\u003e \u003c/li\u003e \u003c/ul\u003e \u003c/p\u003e \u003cdiv id=\"Sec23\" class=\"Section3\"\u003e \u003ch2\u003eCOVID-19 Perceptions and Turnout Readiness (Autumn 2021)\u003c/h2\u003e \u003cp\u003eUsing the Autumn 2021 face-to-face survey (N\u0026thinsp;\u003cb\u003e\u0026asymp;\u003c/b\u003e\u0026thinsp;5,000), we estimate logistic models predicting a top-two-box indicator of turnout intention (\u0026ldquo;certainly/probably would vote\u0026rdquo;), adjusting for standard sociodemographics, self-reported health, and COVID-related covariates (e.g., perceived personal impact, societal concern, vaccination status). Full model estimates are reported in Online Appendix Table A4.\u003c/p\u003e \u003cp\u003eTurnout readiness is strongly structured by baseline partisanship: self-identified Fidesz voters report higher intended turnout than opposition voters even after covariate adjustment. This gap matters for interpreting later crisis dynamics, because an opposition advantage in general engagement does not necessarily translate into electoral leverage when vote readiness remains uneven across blocs.\u003c/p\u003e \u003cp\u003eAt the same time, pandemic perceptions do not translate into uniform mobilization. Viewing COVID-19 as a broader societal problem is associated with a lower likelihood of intending to vote, whereas perceived personal impact shows no independent association once covariates are included; vaccination status is positively associated with turnout intention. Online Appendix Table A4 reports these estimates, and interaction terms provide no robust evidence that these relationships differ systematically between Fidesz and opposition voters in this late-pandemic snapshot.\u003c/p\u003e \u003cp\u003eTaken together, the late-pandemic survey evidence indicates that COVID-related concern can coincide with political withdrawal for some respondents\u0026mdash;not a generalized rally-like mobilization\u0026mdash;and that baseline vote readiness differs across partisan blocs. This pattern is consistent with our macro finding that COVID does not generate a clean, aggregate rally in incumbent support.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e \u003cdiv id=\"Sec24\" class=\"Section2\"\u003e \u003ch2\u003ePre-election Participation Baseline (March 2022)\u003c/h2\u003e \u003cp\u003eWe next use the March 2022 pre-election survey to assess whether partisan blocs differed in non-electoral participatory readiness on the eve of the election, shortly after the war\u0026rsquo;s outbreak. Negative binomial models of a participation repertoire index (0\u0026ndash;8 activities; QM9) reveal that any bivariate opposition\u0026ndash;government gap is not robust once we adjust for baseline political engagement. Controlling for political interest (0\u0026ndash;100) and time spent on political information (minutes per day) eliminates the bloc difference (Opposition vs. Fidesz: B\u0026thinsp;=\u0026thinsp;0.045, p\u0026thinsp;=\u0026thinsp;0.770); full estimates are reported in Online Appendix Table A4. Interest and information-seeking remain strong predictors of participation, suggesting that pre-election participation differences primarily proxy general attentiveness rather than crisis-specific partisan mobilization capacity.\u003c/p\u003e \u003cp\u003eIn combination, the Autumn 2021 and March 2022 surveys narrow the set of plausible micro-mechanisms behind the polling discontinuities. COVID-related perceptions are not tightly coupled to turnout readiness in a way that would imply an automatic \u0026ldquo;rally \u0026rarr; mobilization\u0026rdquo; dynamic, and participation differences prior to the election largely reflect baseline engagement and information acquisition. These results make a classic, uniform rally less likely and motivate a focus on how the subsequent acute crisis\u0026mdash;the war\u0026mdash;could reshape electoral alignment through bloc-owned framing and asymmetric vulnerability.\u003c/p\u003e \u003cdiv id=\"Sec25\" class=\"Section3\"\u003e \u003ch2\u003ePost-election Vote Motivations and Crisis Frames (May 2022)\u003c/h2\u003e \u003cp\u003eWe then turn to the May 2022 post-election face-to-face survey (N\u0026thinsp;\u0026asymp;\u0026thinsp;1,000) to connect the aggregate time-series pattern to individual-level motivations. Because post-election reports may involve post-hoc rationalization, we interpret these measures primarily as indicators of bloc-specific crisis frames\u0026mdash;that is, the considerations respondents make available and legitimate when explaining their vote\u0026mdash;rather than as clean causal drivers.\u003c/p\u003e \u003cp\u003eThe post-election survey indicates sharply asymmetric crisis framing consistent with bloc-owned motivations (H3). Figure\u0026nbsp;\u003cspan refid=\"Fig3\" class=\"InternalRef\"\u003e3\u003c/span\u003e summarizes crisis-related vote motivations (see Online Appendix Table A7 for full distributions). War-related considerations are a dominant vote motive among government voters: 55.9% of Fidesz voters cite the war among their top-two reasons for voting, compared to 18.2% of opposition voters. Open-ended accounts reinforce this asymmetry: war justifications appear almost exclusively among government voters (13.2% of Fidesz voters; 0% among opposition voters).\u003c/p\u003e \u003cp\u003eBy contrast, COVID-19 is largely absent as a salient post-election vote cue. Across closed- and open-ended items, only 3.2% of respondents (22 of 687 valid cases) mention COVID-19 as a top-two motive. Mentions are rare overall, though slightly more frequent among Fidesz voters (4.8% vs. 0.4% among opposition voters). Given the low base rate, we treat this pattern as descriptive rather than as a basis for multivariable modeling.\u003c/p\u003e \u003cp\u003e \u003c/p\u003e \u003cp\u003eBars show the share of respondents citing the war or COVID-19 among their top-two closed-ended reasons for their list vote, by vote bloc (Fidesz vs joint opposition). War is a dominant and sharply polarized motive (55.9% vs 18.2%), whereas COVID-19 is rare overall and marginally more common among Fidesz voters (4.8% vs 0.4%).\u003c/p\u003e \u003cp\u003eTo assess whether asymmetric war framing is accompanied by differential volatility or late deciding, we examine decision timing and campaign instability. Opposition voters are significantly more likely than Fidesz voters to report deciding in the final week/days before the election (12.1% vs. 6.4%; χ\u0026sup2; = 6.66, p\u0026thinsp;=\u0026thinsp;0.01, weighted), consistent with weaker baseline vote readiness on the opposition side. A broader indicator of deciding after the war onset (including \u0026ldquo;one month before the election\u0026rdquo;) shows only a modest and statistically non-significant bloc difference (20.2% vs. 15.8%; p\u0026thinsp;=\u0026thinsp;0.14) and reported campaign instability during February\u0026ndash;election day is similar across blocs (24.5% vs. 23.7%; p\u0026thinsp;=\u0026thinsp;0.81). Within the opposition bloc, citing the war among the top-two motives are not associated with campaign instability or decision timing (all p\u0026thinsp;\u0026gt;\u0026thinsp;0.60; see Online Appendix Table A7), suggesting that the war frame primarily differentiates between blocs rather than sorting voters within the opposition camp.\u003c/p\u003e \u003cp\u003eThe broader post-war decision-timing proxy includes the response category \u0026ldquo;one month before the election,\u0026rdquo; which can fall just before or just after 24 February 2022. We therefore treat this measure as potentially noisy, though still informative, for detecting war-related late deciding.\u003c/p\u003e \u003cp\u003eFinally, patterns of considered alternatives (PE8) suggest subtle cross-pressures, particularly among eventual Fidesz voters. Fidesz voters are more likely than opposition voters to report having considered another list (13.1% vs. 4.9%; χ\u0026sup2; = 11.51, p\u0026thinsp;=\u0026thinsp;0.001; OR\u0026thinsp;=\u0026thinsp;2.92; see Online Appendix Table A7). Among those who considered an alternative (N\u0026thinsp;=\u0026thinsp;69 weighted), most eventual Fidesz voters named Fidesz itself as the alternative (\u0026asymp;\u0026thinsp;67%), indicating intra-camp deliberation, but some considered the joint opposition (\u0026asymp;\u0026thinsp;12%) or Our Homeland (Mi Haz\u0026aacute;nk; \u0026asymp;9%). These cross-pressures point to modest volatility on the incumbent side rather than war-driven re-sorting within the opposition bloc.\u003c/p\u003e \u003cp\u003eOnline Appendix Tables A8\u0026ndash;A9 report multivariable models that reinforce these descriptive patterns. Adjusting for demographics, settlement type, ideology, and political interest, Fidesz voting remains strongly associated with war-related motive claims and earlier decision timing, while within-opposition war mentions do not predict volatility proxies. Overall, the multivariable evidence supports the interpretation that the war frame differentiates between blocs rather than explaining within-bloc volatility.\u003c/p\u003e \u003cp\u003eTaken together, the survey evidence provides micro-foundations for why the war breakpoint appears in the aggregate polls primarily as an opposition-side drop rather than a clean incumbent surge. The war becomes a strongly bloc-owned motivational frame among government voters (consistent with H3), while opposition voters exhibit weaker vote readiness and a higher propensity for last-minute decision-making. Across indicators, we find limited evidence that war motivations sort voters within the opposition bloc. In combination with the polling discontinuity, these findings support our sequential, bloc-structured account of rally effects in polarized contexts (H2\u0026ndash;H3) and help reconcile muted aggregate movement in incumbent support with sharper war-timed opposition weakening.\u003c/p\u003e \u003c/div\u003e \u003c/div\u003e"},{"header":"Discussion","content":"\u003cp\u003eWe emphasize that these patterns should be read as discontinuities consistent with a crisis-induced reallocation, not as definitive causal effects of the crises themselves. The ITS design strengthens temporal attribution by focusing on sharp breakpoints and pre-trend structure, but it cannot fully rule out all concurrent campaign and information shocks.\u003c/p\u003e \u003cp\u003eAcross four complementary empirical components, the evidence is most consistent with a selective and asymmetric crisis process. In the aggregate poll series, we do not observe a large, sustained incumbent jump at either crisis onset. Instead, the clearest macro-level movement is an opposition-side weakening around the war shock, coupled with micro-level evidence that the war became a highly salient motivational frame among incumbent supporters while remaining comparatively muted among opposition voters.\u003c/p\u003e \u003cp\u003eThis study re-evaluates rally dynamics under sequential crises in a highly polarized, dominant-party setting. Rather than assuming that external shocks generate a broad, aggregate surge in incumbent support, we theorize and test a bloc-structured pattern in which crises can reallocate participation readiness and electoral competition asymmetrically across partisan publics.\u003c/p\u003e \u003cp\u003e \u003cdiv class=\"gridtable\"\u003e\u003ctable float=\"Yes\" id=\"Tab4\" border=\"1\"\u003e \u003ccaption language=\"En\"\u003e \u003cdiv class=\"CaptionNumber\"\u003eTable 4\u003c/div\u003e \u003cdiv class=\"CaptionContent\"\u003e \u003cp\u003eSummary of Findings Across Empirical Components\u003c/p\u003e \u003c/div\u003e \u003c/caption\u003e \u003ccolgroup cols=\"4\"\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c1\" colnum=\"1\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c2\" colnum=\"2\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c3\" colnum=\"3\"\u003e\u003c/div\u003e \u003cdiv align=\"left\" class=\"colspec\" colname=\"c4\" colnum=\"4\"\u003e\u003c/div\u003e \u003cthead\u003e \u003ctr\u003e \u003cth align=\"left\" colname=\"c1\"\u003e \u003cp\u003eComponent (what it tests)\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c2\"\u003e \u003cp\u003e\u0026ldquo;Classic rally\u0026rdquo; expectation\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c3\"\u003e \u003cp\u003eWhat we find\u003c/p\u003e \u003c/th\u003e \u003cth align=\"left\" colname=\"c4\"\u003e \u003cp\u003eMechanism implication\u003c/p\u003e \u003c/th\u003e \u003c/tr\u003e \u003c/thead\u003e \u003ctbody\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003ePoll-of-polls ITS (2018\u0026ndash;2022)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eLarge, abrupt pro-incumbent shift after crises\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eNo clean Fidesz level shift; war coincides with an opposition-side level drop\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eRally can show up as \u003cem\u003eopposition weakening/re-sorting\u003c/em\u003e, not only incumbent jumps\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eAutumn 2021 survey (COVID perceptions \u0026rarr; turnout intention)\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003ePandemic threat mobilizes broadly\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eCOVID perceptions do not \u0026ldquo;mobilize\u0026rdquo; uniformly; effects look like \u003cem\u003ecomposition\u0026thinsp;+\u0026thinsp;readiness\u003c/em\u003e differences\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eCrisis effects filter through \u003cem\u003ewho is vote-ready\u003c/em\u003e, not national mood\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMarch 2022 pre-election baseline participation\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eOpposition \u0026ldquo;anti-incumbent\u0026rdquo; energy should dominate engagement\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003eApparent partisan gaps largely track \u003cem\u003epolitical interest/info-seeking\u003c/em\u003e\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eParticipation capacity is not automatically electoral leverage without readiness\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003ctr\u003e \u003ctd align=\"left\" colname=\"c1\"\u003e \u003cp\u003eMay 2022 post-election motives\u0026thinsp;+\u0026thinsp;frames\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c2\"\u003e \u003cp\u003eCrisis cues shared across electorate\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c3\"\u003e \u003cp\u003e\u003cem\u003eWar\u003c/em\u003e is dominant and sharply polarized by bloc; \u003cem\u003eCOVID\u003c/em\u003e is rare as a motive\u003c/p\u003e \u003c/td\u003e \u003ctd align=\"left\" colname=\"c4\"\u003e \u003cp\u003eWar is \u003cem\u003einterpreted and owned\u003c/em\u003e disproportionately by the pro-government camp; COVID has faded as an election cue\u003c/p\u003e \u003c/td\u003e \u003c/tr\u003e \u003c/tbody\u003e \u003c/colgroup\u003e \u003c/table\u003e\u003c/div\u003e \u003c/p\u003e \u003cdiv id=\"Sec27\" class=\"Section2\"\u003e \u003ch2\u003eRe-assessing the \u0026lsquo;dual rally\u0026rsquo; hypothesis\u003c/h2\u003e \u003cp\u003eThis sequential-crisis pattern speaks to an important boundary condition for rally theories. When trust and credibility are polarized, the key electoral consequence of crises may be the redistribution of readiness and cohesion across blocs rather than visible aggregate uplift for incumbents.\u003c/p\u003e \u003cp\u003eOverall, the Hungarian case provides limited support for a simple \u0026ldquo;dual rally\u0026rdquo; expectation in which COVID-19 and the war each produce broad incumbent surges (H1). Instead, the evidence points toward an asymmetric mechanism: COVID-19 appears mainly as a prolonged background condition that does not translate into an election-proximate incumbent boost, whereas the war\u0026mdash;by timing and by frame ownership\u0026mdash;aligns more closely with an opposition-side contraction in the likely electorate (H2).\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec28\" class=\"Section2\"\u003e \u003ch2\u003eExplaining the Landslide: How Muted Aggregate Rally Still Produces Decisive Electoral Outcomes\u003c/h2\u003e \u003cp\u003eOur survey evidence clarifies one pathway for such divergence: the war was disproportionately integrated into incumbent voters\u0026rsquo; stated motivations and open-ended justifications, consistent with security/peace frame ownership (H3). In a context where turnout and persuasion are already asymmetric, such frame alignment can contribute to a landslide outcome even if the aggregate level of incumbent support does not surge.\u003c/p\u003e \u003cp\u003eA broader implication concerns how decisive electoral outcomes can emerge even when poll averages look comparatively stable. Small shifts in the composition of the likely electorate\u0026mdash;especially if concentrated on one side of a polarized competition\u0026mdash;can widen electoral margins without producing a dramatic aggregate \u0026ldquo;rally spike.\u0026rdquo;\u003c/p\u003e \u003cp\u003eThese findings also help reconcile muted aggregate rally patterns with large electoral margins. In highly polarized settings, elections are often decided at the margin among weaker partisans and intermittent voters; modest bloc-specific demobilization or defection can therefore have outsized consequences when concentrated asymmetrically. A crisis that does not attract new supporters to the incumbent may still reduce the opposition\u0026rsquo;s effective coalition by lowering enthusiasm, increasing cross-pressures, or reweighting the salience of issues on which the incumbent holds framing or ownership advantages. From this perspective, the key electoral object is not the national \u0026ldquo;rally\u0026rdquo; level shift, but the distribution of readiness and cohesion across blocs as the campaign unfolds. The war shock in February 2022 is a particularly sharp instance because it entered an already campaign-like information environment, allowing government framing to structure interpretation quickly. This mechanism complements\u0026mdash;rather than replaces\u0026mdash;standard accounts centered on incumbency advantage and resource mobilization.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec29\" class=\"Section2\"\u003e \u003ch2\u003eAlternative explanations: opposition coordination, incumbency advantage, and fiscal mobilization\u003c/h2\u003e \u003cp\u003eTwo considerations help situate the crisis-related patterns in the broader 2022 electoral context. First, unlike earlier opposition cycles in Hungary, the challengers entered the campaign as a coordinated joint list with a single prime-ministerial candidate. Coordination should, in principle, mitigate vote-splitting and strengthen the credibility of alternation. Yet it can also raise coordination costs and messaging tensions in heterogeneous coalitions, potentially depressing enthusiasm among peripheral supporters and leaving less room for rapid agenda shifts once the campaign enters its final phase.\u003c/p\u003e \u003cp\u003eSecond, the election unfolded under pronounced incumbency advantages. In competitive-authoritarian settings, incumbents can combine agenda control, asymmetric access to media, and the strategic deployment of state resources with rule-based advantages that shape the information environment and the perceived costs of defection. These tools are well documented in the comparative literature on electoral authoritarianism and hegemonic party dominance (Schedler \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2002\u003c/span\u003e; Magaloni \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Levitsky and Way \u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Schedler \u003cspan citationid=\"CR46\" class=\"CitationRef\"\u003e2013\u003c/span\u003e).\u003c/p\u003e \u003cp\u003eA related, and electorally salient, channel is pre-election fiscal mobilization. Political-budget-cycle theories emphasize that incumbents may tilt policy toward highly visible transfers, tax relief, or price interventions in the run-up to an election, especially where monitoring is imperfect and partisan cues structure attribution (Rogoff \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e1990\u003c/span\u003e; Brender and Drazen \u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e2005\u003c/span\u003e; Shi and Svensson \u003cspan citationid=\"CR48\" class=\"CitationRef\"\u003e2006\u003c/span\u003e). In Hungary, such measures can plausibly elevate baseline incumbent support and turnout readiness among government-leaning voters.\u003c/p\u003e \u003cp\u003eThese alternative explanations are not substitutes for the crisis account advanced here. If anything, they help clarify why the war shock could matter electorally even when aggregate rally effects appear muted. Structural incumbency advantages and fiscal mobilization can raise the level of government support, but they do not predict the timing of the within-opposition contraction around the war breakpoint. The war, by contrast, plausibly provided an affectively charged and frames-relevant cue that the government could claim as an issue-ownership advantage (peace/security), consistent with the post-election motivation patterns among Fidesz voters and the open-ended responses.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eAddressing Within-Opposition Dynamics and Explanatory Modeling Concerns\u003c/h3\u003e\n\u003cp\u003eAt the same time, within-bloc heterogeneity and organizational dynamics remain plausible complementary explanations. Future work with panel data and validated turnout measures is needed to separate preference change, turnout propensity change, and measurement artifacts in crisis-period opposition contraction.\u003c/p\u003e \u003cp\u003eOur argument is primarily between-bloc: war framing and crisis-period readiness differ sharply across the incumbent and opposition publics. Within the opposition, we find limited evidence that war motivations systematically predict late deciding or campaign instability. This suggests that the opposition\u0026rsquo;s weakening is less about widespread last-minute switching within the opposition and more about asymmetric readiness and frame disadvantage relative to the incumbent bloc.\u003c/p\u003e \u003cp\u003eAlternative explanations and why timing matters. Several concurrent developments could in principle shape opposition support in early 2022, including inflationary pressures, campaign events, or organizational frictions within the heterogeneous opposition alliance. Most such forces, however, should generate gradual movements or affect both blocs in the same direction, rather than producing an abrupt, opposition-specific break aligned with the invasion date. The observed pattern\u0026mdash;muted incumbent movement but a sharper opposition-side discontinuity, together with survey evidence that war-related considerations were disproportionately available to (and claimed by) incumbent voters\u0026mdash;fits a framing/ownership mechanism that operates between blocs. We nevertheless treat the macro discontinuity as within-case evidence that is strengthened by convergence across designs, not as a definitive estimate that excludes all parallel shocks.\u003c/p\u003e \u003cdiv id=\"Sec31\" class=\"Section2\"\u003e \u003ch2\u003eLimitations and Future Research Avenues\u003c/h2\u003e \u003cp\u003eFirst, the macro evidence comes from a single-country interrupted time-series design with one treated unit and relies on sharp timing assumptions. Although we address pollster house effects, time-series dependence, and window-length sensitivity, unobserved concurrent events cannot be fully excluded.\u003c/p\u003e \u003cp\u003eSecond, post-election motive reports are subject to rationalization and identity-consistent storytelling; we therefore interpret motive patterns as indicators of frame availability and ownership rather than as direct causal drivers of vote choice.\u003c/p\u003e \u003cp\u003eThird, some behavioral proxies (e.g., decision timing categories) are coarse. We treat these measures as suggestive indicators of volatility and emphasize convergence across measures and datasets rather than any single item.\u003c/p\u003e \u003cp\u003eOn balance, the evidence supports a conservative reading: the war shock is associated with a bloc-asymmetric shift that is consistent with rally dynamics among incumbent supporters, even as aggregate series remain muted because countervailing movements occur across blocs. Our designs cannot uniquely attribute the 2022 landslide to the war alone; incumbency advantages, pre-election transfers, and within-opposition coordination frictions plausibly contributed in parallel.\u003c/p\u003e \u003cp\u003eWhat the analysis does establish is that the war\u0026rsquo;s timing relative to the election coincided with (i) changes in opposition support in the polling series and (ii) post-election war-related vote rationales that are highly concentrated among Fidesz voters. Accordingly, the article\u0026rsquo;s core contribution is best read as a boundary condition for rally effects in polarized, dominant-party settings: crises may matter primarily through partisan reallocation and turnout readiness rather than through a uniform national surge. Future work can adjudicate these pathways with cross-national designs and individual-level panels that track both preferences and turnout propensity across sequential shocks.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec32\" class=\"Section2\"\u003e \u003ch2\u003eConclusion and Broad Contributions\u003c/h2\u003e \u003cp\u003eMore broadly, our findings mandate a fundamental re-conceptualization of 'rally' effects. Instead of solely seeking aggregate incumbent uplift, scholars must consider 'rally' as a process of selective redistribution across partisan publics, driven by differential vulnerabilities and framing advantages. This new lens is crucial for future comparative research, guiding investigations into how institutional credibility, fragmented information environments, and opposition coordination critically condition crisis politics across diverse regimes and electoral cycles.\u003c/p\u003e \u003cp\u003eWe expect the sequential, bloc-asymmetric pattern documented here to be most likely when (a) partisan identities structure information exposure and elite credibility, (b) the incumbent can plausibly claim superior competence on the focal crisis dimension (e.g., security/peace), and (c) a major shock occurs close enough to an election that turnout readiness and issue salience are high. In such settings, crises may stabilize or even amplify incumbent advantages without generating a visible nationwide surge in average incumbent support. Comparative work that combines high-frequency opinion series with pre- and post-election survey modules across cases could test these scope conditions directly and clarify when aggregate rally effects are most likely to appear\u0026mdash;or to remain hidden by bloc-level countervailing movements.\u003c/p\u003e \u003cp\u003eDespite these constraints, the study offers a clear contribution to the rally literature: in polarized, dominant-party contexts, sequential crises may not produce additive incumbent surges. Instead, crisis timing and frame ownership can generate selective rally dynamics that are visible primarily as opposition-side weakening and asymmetric mobilization of the likely electorate.\u003c/p\u003e \u003cp\u003eMore broadly, our sequential-crisis lens travels significantly beyond the Hungarian case, offering crucial insights for understanding crisis politics in other polarized settings. We caution scholars against inferring 'no rally' solely from seemingly flat aggregate polls when multiple shocks occur close to an election. Instead, the critical theoretical and empirical question shifts to \u003cem\u003ehow each shock is strategically translated into a credible voting reason within competing partisan narratives\u003c/em\u003e, and \u003cem\u003ewhich political camp is ultimately better positioned\u0026mdash;through its elite messaging, media environment, and organizational capacity\u0026mdash;to effectively own and communicate that frame\u003c/em\u003e. This approach provides a more nuanced and accurate understanding of how exogenous events reshape electoral behavior in the 21st century.\u003c/p\u003e \u003c/div\u003e"},{"header":"Declarations","content":"\u003cp\u003e \u003cb\u003eEthics.\u003c/b\u003e \u003c/p\u003e \u003cp\u003eThis article presents a secondary analysis of anonymized survey data and publicly available polling data. No new data was collected for this study, and no additional ethical approval was required for the secondary analyses.\u003c/p\u003e\u003ch2\u003eFunding.\u003c/h2\u003e \u003cp\u003eThis research received no external funding.\u003c/p\u003e\u003ch2\u003eAuthor Contribution\u003c/h2\u003e\u003cp\u003eAndrea Szab\u0026oacute; conceived the study, developed the theoretical framework, designed the research, conducted the analyses, prepared the figures and tables, and wrote the manuscript and online appendix.\u003c/p\u003e\u003ch2\u003eData Availability\u003c/h2\u003e\u003cp\u003ePolling data are available from the Vox Populi archive. Replication code and derived datasets (including the aggregated time-series used for analysis) will be deposited in an open repository upon acceptance. Access to the original survey microdata is restricted; access may be granted upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eB\u0026aelig;kgaard, M., Christensen, J., Madsen, J. K., \u0026amp; Kim Sass, M. (2020). Rallying around the flag in times of COVID-19: Societal lockdown and trust in democratic institutions. \u003cem\u003eJournal of Behavioral Public Administration\u003c/em\u003e, \u003cem\u003e3\u003c/em\u003e(2). \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.30636/jbpa.32.172\u003c/span\u003e\u003cspan address=\"10.30636/jbpa.32.172\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaker, W. D., \u0026amp; Oneal, J. R. (2001). Patriotism or opinion leadership? The nature and origins of the rally \u0026rsquo;round the flag effect. \u003cem\u003eJournal of Conflict Resolution\u003c/em\u003e, \u003cem\u003e45\u003c/em\u003e(5), 661\u0026ndash;687. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0022002701045005006\u003c/span\u003e\u003cspan address=\"10.1177/0022002701045005006\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBaum, M. A. (2002). The constituent foundations of the rally-round-the-flag phenomenon. \u003cem\u003eInternational Studies Quarterly\u003c/em\u003e, \u003cem\u003e46\u003c/em\u003e(2), 263\u0026ndash;298. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1468-2478.00232\u003c/span\u003e\u003cspan address=\"10.1111/1468-2478.00232\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eB\u0026eacute;langer, \u0026Eacute;., \u0026amp; Meguid, B. M. (2008). Issue salience, issue ownership, and issue-based vote choice. \u003cem\u003eElectoral Studies\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e(3), 477\u0026ndash;491. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.electstud.2008.01.001\u003c/span\u003e\u003cspan address=\"10.1016/j.electstud.2008.01.001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBol, D., Giani, M., \u0026amp; Blais, A., and Peter John Loewen (2021). The effect of COVID-19 lockdowns on political support: Some good news for democracy? \u003cem\u003eEuropean Journal of Political Research\u003c/em\u003e, \u003cem\u003e60\u003c/em\u003e(2), 497\u0026ndash;505. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1475-6765.12401\u003c/span\u003e\u003cspan address=\"10.1111/1475-6765.12401\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrender, A., \u0026amp; Drazen, A. (2005). Political budget cycles in new versus established democracies. \u003cem\u003eJournal of Monetary Economics\u003c/em\u003e, \u003cem\u003e52\u003c/em\u003e(7), 1271\u0026ndash;1295. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jmoneco.2005.04.004\u003c/span\u003e\u003cspan address=\"10.1016/j.jmoneco.2005.04.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrody, R. A. (1991). \u003cem\u003eAssessing the President: The Media, Elite Opinion, and Public Support\u003c/em\u003e. Stanford University Press.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrouard, S., and Elie Michel (2025). Retrospective voting in times of pandemic: Comparative evidence from panel data. \u003cem\u003ePolitical Science\u003c/em\u003e. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/00323187.2025.2500926\u003c/span\u003e\u003cspan address=\"10.1080/00323187.2025.2500926\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e. Advance online publication.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCh\u0026aacute;vez, K., and James Wright (2022). International crisis termination and presidential approval. \u003cem\u003eForeign Policy Analysis\u003c/em\u003e, \u003cem\u003e18\u003c/em\u003e(3), orac005. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/fpa/orac005\u003c/span\u003e\u003cspan address=\"10.1093/fpa/orac005\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eColloca, P., Roccato, M., \u0026amp; Russo, S. (2024). Rally \u0026rsquo;round the flag effects are not for all: Trajectories of institutional trust among populist and non-populist voters. \u003cem\u003eSocial Science Research\u003c/em\u003e, \u003cem\u003e119\u003c/em\u003e, 102986. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ssresearch.2024.102986\u003c/span\u003e\u003cspan address=\"10.1016/j.ssresearch.2024.102986\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDevine, D., \u0026amp; Viktor Orri, V. (2024). Stability and change in political trust: Evidence and implications from six panel studies. \u003cem\u003eEuropean Journal of Political Research\u003c/em\u003e, \u003cem\u003e63\u003c/em\u003e(2), 478\u0026ndash;497. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1475-6765.12606\u003c/span\u003e\u003cspan address=\"10.1111/1475-6765.12606\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDruckman, J. N., Peterson, E., \u0026amp; Slothuus, R. (2013). How elite partisan polarization affects public opinion formation. \u003cem\u003eAmerican Political Science Review\u003c/em\u003e, \u003cem\u003e107\u003c/em\u003e(1), 57\u0026ndash;79. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/S0003055412000500\u003c/span\u003e\u003cspan address=\"10.1017/S0003055412000500\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEdwards, G. C., III, and, \u0026amp; Swenson, T. (1997). Who rallies? The anatomy of a rally event. \u003cem\u003eJournal of Politics\u003c/em\u003e, \u003cem\u003e59\u003c/em\u003e(1), 200\u0026ndash;212. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2307/2998222\u003c/span\u003e\u003cspan address=\"10.2307/2998222\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFrateur, J., Bursens, P., Popelier, P., \u0026amp; Coroado, S. (2025). How to explain citizens\u0026rsquo; declined political trust during the COVID-19 pandemic: a time-series qualitative comparative analysis (TsQCA). \u003cem\u003eComparative European Politics\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e, 552\u0026ndash;575. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1057/s41295-025-00412-z\u003c/span\u003e\u003cspan address=\"10.1057/s41295-025-00412-z\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFrye, T. (2019). Economic sanctions and public opinion: Survey experiments from Russia. \u003cem\u003eComparative Political Studies\u003c/em\u003e, \u003cem\u003e52\u003c/em\u003e(7), 967\u0026ndash;994. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/0010414018806530\u003c/span\u003e\u003cspan address=\"10.1177/0010414018806530\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGrechanaya, T., \u0026amp; Ceron, A. (2024). Patriotism and national symbols in Russian and Ukrainian elections. \u003cem\u003ePolitics and Governance\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(2), 355\u0026ndash;367. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.17645/pag.7918\u003c/span\u003e\u003cspan address=\"10.17645/pag.7918\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHegewald, S., and Dominik Schraff (2024). Who rallies around the flag? Evidence from panel data during the Covid-19 pandemic. \u003cem\u003eJournal of Elections Public Opinion and Parties\u003c/em\u003e, \u003cem\u003e34\u003c/em\u003e(1), 158\u0026ndash;179. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/17457289.2022.2120886\u003c/span\u003e\u003cspan address=\"10.1080/17457289.2022.2120886\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHetherington, M. J., \u0026amp; Nelson, M. (2003). Anatomy of a rally effect: George W. Bush and the War on Terrorism. PS: Political Science \u0026amp; Politics 36(1): 37\u0026ndash;42. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/S1049096503001665\u003c/span\u003e\u003cspan address=\"10.1017/S1049096503001665\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHetherington, M. J., \u0026amp; Rudolph, T. J. (2015). \u003cem\u003eWhy Trust Matters: Declining Political Trust and the Demise of American Liberalism\u003c/em\u003e. Princeton University Press.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIyengar, S., \u0026amp; Kinder, D. R. (1987). \u003cem\u003eNews That Matters: Television and American Opinion\u003c/em\u003e. University of Chicago Press.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eIyengar, S., \u0026amp; Sood, G., and Yphtach Lelkes (2012). Affect, not ideology: A social identity perspective on polarization. \u003cem\u003ePublic Opinion Quarterly\u003c/em\u003e, \u003cem\u003e76\u003c/em\u003e(3), 405\u0026ndash;431. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/poq/nfs038\u003c/span\u003e\u003cspan address=\"10.1093/poq/nfs038\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJohansson, B., Hopmann, D. N., \u0026amp; Shehata, A. (2021). When the rally-around-the-flag effect disappears, or when the COVID-19 pandemic becomes normalized. \u003cem\u003eJournal of Elections Public Opinion and Parties\u003c/em\u003e, \u003cem\u003e31\u003c/em\u003e(sup1), 321\u0026ndash;334. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/17457289.2021.1924742\u003c/span\u003e\u003cspan address=\"10.1080/17457289.2021.1924742\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKizilova, K., and Pippa Norris (2024). Rally around the flag effects in the Russian\u0026ndash;Ukrainian war. \u003cem\u003eEuropean Political Science\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e(2), 234\u0026ndash;250. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1057/s41304-023-00450-9\u003c/span\u003e\u003cspan address=\"10.1057/s41304-023-00450-9\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKritzinger, S., Foucault, M., Lachat, R., Partheym\u0026uuml;ller, J., \u0026amp; Eberl, M. S., and Marko Bubalo (2021). Rally round the flag\u0026rsquo;: The COVID-19 crisis and trust in the national government. \u003cem\u003eWest European Politics\u003c/em\u003e, \u003cem\u003e44\u003c/em\u003e(5\u0026ndash;6), 1205\u0026ndash;1231. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/01402382.2021.1925017\u003c/span\u003e\u003cspan address=\"10.1080/01402382.2021.1925017\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKuijpers, D. (2019). Rally around all the flags: The effect of military casualties on incumbent popularity in ten countries 1990\u0026ndash;2014. \u003cem\u003eForeign Policy Analysis\u003c/em\u003e, \u003cem\u003e15\u003c/em\u003e(3), 392\u0026ndash;412. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/fpa/orz014\u003c/span\u003e\u003cspan address=\"10.1093/fpa/orz014\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLai, B., and Dan Reiter (2005). Rally \u0026rsquo;round the Union Jack? Public opinion and the use of force in the United Kingdom, 1948\u0026ndash;2001. \u003cem\u003eInternational Studies Quarterly\u003c/em\u003e, \u003cem\u003e49\u003c/em\u003e(2), 255\u0026ndash;272. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.0020-8833.2005.00344.x\u003c/span\u003e\u003cspan address=\"10.1111/j.0020-8833.2005.00344.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLeininger, A., and Max Schaub (2024). Strategic alignment in times of crisis: Voting at the dawn of a global pandemic. \u003cem\u003ePolitical Behavior\u003c/em\u003e, \u003cem\u003e46\u003c/em\u003e, 1589\u0026ndash;1607. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11109-023-09885-8\u003c/span\u003e\u003cspan address=\"10.1007/s11109-023-09885-8\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLevendusky, M. S. (2013). Why do partisan media polarize viewers? \u003cem\u003eAmerican Journal of Political Science\u003c/em\u003e, \u003cem\u003e57\u003c/em\u003e(3), 611\u0026ndash;623. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/ajps.12008\u003c/span\u003e\u003cspan address=\"10.1111/ajps.12008\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLevitsky, S., \u0026amp; Way, L. A. (2010). \u003cem\u003eCompetitive Authoritarianism: Hybrid Regimes after the Cold War\u003c/em\u003e. Cambridge University Press.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eLytkina, E., and Tim Reeskens (2024). Rally around the government or a populist response? How concerns about COVID-19 and emotional responses relate to institutional trust and support for right-wing populism. \u003cem\u003eAmerican Behavioral Scientist\u003c/em\u003e, \u003cem\u003e69\u003c/em\u003e(4), 478\u0026ndash;504. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/00027642241240418\u003c/span\u003e\u003cspan address=\"10.1177/00027642241240418\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMagaloni, B. (2006). \u003cem\u003eVoting for Autocracy: Hegemonic Party Survival and Its Demise in Mexico\u003c/em\u003e. Cambridge University Press.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMason, L. (2018). \u003cem\u003eUncivil Agreement: How Politics Became Our Identity\u003c/em\u003e. University of Chicago Press. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.7208/chicago/9780226524689.001.0001\u003c/span\u003e\u003cspan address=\"10.7208/chicago/9780226524689.001.0001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMitchell, J., Bohman, A., Eger, M. A., \u0026amp; Hjerm, M. (2025). Rally around the flag? Explaining changes in Swedish public opinion toward NATO membership after Russia\u0026rsquo;s invasion of Ukraine. \u003cem\u003eActa Sociologica\u003c/em\u003e, \u003cem\u003e68\u003c/em\u003e(1), 30\u0026ndash;40. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1177/00016993241268185\u003c/span\u003e\u003cspan address=\"10.1177/00016993241268185\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorales, J. (2021). Legislating during war: Conflict and politics. \u003cem\u003eJournal of Public Economics\u003c/em\u003e, \u003cem\u003e193\u003c/em\u003e, 104325. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jpubeco.2020.104325\u003c/span\u003e\u003cspan address=\"10.1016/j.jpubeco.2020.104325\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMueller, J. E. (1970). Presidential popularity from Truman to Johnson. \u003cem\u003eAmerican Political Science Review\u003c/em\u003e, \u003cem\u003e64\u003c/em\u003e(1), 18\u0026ndash;34. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2307/1955610\u003c/span\u003e\u003cspan address=\"10.2307/1955610\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMueller, J. E. (1973). \u003cem\u003eWar, Presidents, and Public Opinion\u003c/em\u003e. Wiley.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMuhammad, M., Domantas, \u0026amp; Undzėnas (2025). Entangled fates\u0026mdash;the rally effect around Europe due to Russia\u0026rsquo;s war of aggression against Ukraine. \u003cem\u003eHumanities and Social Sciences Communications\u003c/em\u003e, \u003cem\u003e12\u003c/em\u003e(1), 1\u0026ndash;10. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1038/s41599-025-05138-7\u003c/span\u003e\u003cspan address=\"10.1038/s41599-025-05138-7\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePetrocik, J. R. (1996). Issue Ownership in Presidential Elections, with a 1980 Case Study. \u003cem\u003eAmerican Journal of Political Science\u003c/em\u003e, \u003cem\u003e40\u003c/em\u003e(3), 825\u0026ndash;850. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.2307/2111797\u003c/span\u003e\u003cspan address=\"10.2307/2111797\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePrior, M. (2007). \u003cem\u003ePost-Broadcast Democracy: How Media Choice Increases Inequality in Political Involvement and Polarizes Elections\u003c/em\u003e. Cambridge University Press.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRezaeeDaryakenari, B., Ghafouri, V., \u0026amp; Kasap, N. (2025). Who rallies round the flag? The impact of the US sanctions on Iranians\u0026rsquo; attitude toward the government. \u003cem\u003eForeign Policy Analysis\u003c/em\u003e, \u003cem\u003e21\u003c/em\u003e(1), orae033. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/fpa/orae033\u003c/span\u003e\u003cspan address=\"10.1093/fpa/orae033\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRiker, W. H. (1986). \u003cem\u003eThe Art of Political Manipulation\u003c/em\u003e. Yale University Press.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRogoff, K. (1990). Equilibrium political budget cycles. \u003cem\u003eAmerican Economic Review\u003c/em\u003e, \u003cem\u003e80\u003c/em\u003e(1), 21\u0026ndash;36.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRožukalne, A., Kleinberga, V., Tīfentāle, A., \u0026amp; Strode, I. (2022). What is the flag we rally around? Trust in information sources at the outset of the COVID-19 pandemic in Latvia. \u003cem\u003eSocial Sciences\u003c/em\u003e, \u003cem\u003e11\u003c/em\u003e(3), 123. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.3390/socsci11030123\u003c/span\u003e\u003cspan address=\"10.3390/socsci11030123\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSafarpour, A., \u0026amp; Baum, M. A. (2022). Pandemic governors and public opinion: The effect of COVID-19 on support for U.S. governors. APSA Preprints. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.33774/apsa-2022-vvgg3\u003c/span\u003e\u003cspan address=\"10.33774/apsa-2022-vvgg3\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchedler, A. (2002). Elections without democracy: The menu of manipulation. \u003cem\u003eJournal of Democracy\u003c/em\u003e, \u003cem\u003e13\u003c/em\u003e(2), 36\u0026ndash;50. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1353/jod.2002.0031\u003c/span\u003e\u003cspan address=\"10.1353/jod.2002.0031\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchedler, A. (2013). \u003cem\u003eThe Politics of Uncertainty: Sustaining and Subverting Electoral Authoritarianism\u003c/em\u003e. Oxford University Press.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSchraff, D. (2021). Political trust during the Covid-19 pandemic: Rally around the flag or lockdown effects? \u003cem\u003eEuropean Journal of Political Research\u003c/em\u003e, \u003cem\u003e60\u003c/em\u003e(4), 1007\u0026ndash;1017. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/1475-6765.12425\u003c/span\u003e\u003cspan address=\"10.1111/1475-6765.12425\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShi, M., \u0026amp; Svensson, J. (2006). Political budget cycles: Do they differ across countries and why? \u003cem\u003eJournal of Public Economics\u003c/em\u003e, \u003cem\u003e90\u003c/em\u003e(8\u0026ndash;9), 1367\u0026ndash;1389. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.jpubeco.2005.09.009\u003c/span\u003e\u003cspan address=\"10.1016/j.jpubeco.2005.09.009\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSides, J., \u0026amp; Vavreck, L. (2013). \u003cem\u003eThe Gamble: Choice and Chance in the 2012 Presidential Election\u003c/em\u003e. Princeton University Press.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSosa-Villagarcia, P., \u0026amp; Ver\u0026oacute;nica, H. L. (2021). Covid-19 and presidential popularity in Latin America. \u003cem\u003eRevista Latinoamericana de Opini\u0026oacute;n P\u0026uacute;blica\u003c/em\u003e, \u003cem\u003e10\u003c/em\u003e(2), 71\u0026ndash;91. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.14201/rlop.23664\u003c/span\u003e\u003cspan address=\"10.14201/rlop.23664\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSteiner, N. D., Berlinschi, R., Farvaque, E., Fidrmuc, J., Harms, P., Mihailov, A., Neugart, M., \u0026amp; Stanek, P. (2023). Rallying around the EU flag: Russia's invasion of Ukraine and attitudes toward European integration. \u003cem\u003eJournal of Common Market Studies\u003c/em\u003e, \u003cem\u003e61\u003c/em\u003e(2), 283\u0026ndash;301. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/jcms.13449\u003c/span\u003e\u003cspan address=\"10.1111/jcms.13449\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStimson, J. A. (1999). \u003cem\u003ePublic Opinion in America: Moods, Cycles, and Swings\u003c/em\u003e. Westview.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStroud, N. J. (2011). \u003cem\u003eNiche News: The Politics of News Choice\u003c/em\u003e. Oxford University Press. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/acprof:oso/9780199755509.001.0001\u003c/span\u003e\u003cspan address=\"10.1093/acprof:oso/9780199755509.001.0001\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStubager, R., and Rune Slothuus (2013). What Are the Sources of Political Parties\u0026rsquo; Issue Ownership? Testing Four Explanations at the Individual Level. \u003cem\u003ePolitical Behavior\u003c/em\u003e, \u003cem\u003e35\u003c/em\u003e(3), 567\u0026ndash;588. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1007/s11109-012-9204-2\u003c/span\u003e\u003cspan address=\"10.1007/s11109-012-9204-2\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTaber, C. S., \u0026amp; Milton Lodge (2006). Motivated skepticism in the evaluation of political beliefs. \u003cem\u003eAmerican Journal of Political Science\u003c/em\u003e, \u003cem\u003e50\u003c/em\u003e(3), 755\u0026ndash;769. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1111/j.1540-5907.2006.00214.x\u003c/span\u003e\u003cspan address=\"10.1111/j.1540-5907.2006.00214.x\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan Alebeek, C. R. A., Tom, W. G., van der Meer, \u0026amp; Hakhverdian, A. (2025). Stable or variable distrust? Disentangling the relationship between political trust and electoral behavior. \u003cem\u003eEuropean Political Science Review First View\u003c/em\u003e, 1\u0026ndash;17. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/S1755773925100210\u003c/span\u003e\u003cspan address=\"10.1017/S1755773925100210\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan der Brug, W. (2004). Issue Ownership and Party Choice. \u003cem\u003eElectoral Studies\u003c/em\u003e, \u003cem\u003e23\u003c/em\u003e(2), 209\u0026ndash;233. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/S0261-3794(02)00061-6\u003c/span\u003e\u003cspan address=\"10.1016/S0261-3794(02)00061-6\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003evan der Meer, T., \u0026amp; Steenvoorden, E., and Ebe Ouattara (2023). Fear and the COVID-19 Rally Round the Flag: A Panel Study on Political Trust. \u003cem\u003eWest European Politics\u003c/em\u003e, \u003cem\u003e46\u003c/em\u003e(6), 1089\u0026ndash;1105. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1080/01402382.2023.2171220\u003c/span\u003e\u003cspan address=\"10.1080/01402382.2023.2171220\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVerdier, D., and Bryan Woo (2011). Why rewards are better than sanctions: Economic statecraft and the rally around the flag effect. \u003cem\u003eEuropean Journal of Political Economy\u003c/em\u003e, \u003cem\u003e27\u003c/em\u003e(2), 217\u0026ndash;232. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1016/j.ejpoleco.2010.08.004\u003c/span\u003e\u003cspan address=\"10.1016/j.ejpoleco.2010.08.004\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWalgrave, S., Lefevere, J., \u0026amp; Tresch, A. (2012). The associative dimension of issue ownership. \u003cem\u003ePublic Opinion Quarterly\u003c/em\u003e, \u003cem\u003e76\u003c/em\u003e(4), 771\u0026ndash;782. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1093/poq/nfs023\u003c/span\u003e\u003cspan address=\"10.1093/poq/nfs023\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZaller, J. R. (1992). \u003cem\u003eThe Nature and Origins of Mass Opinion\u003c/em\u003e. Cambridge University Press. \u003cspan class=\"ExternalRef\"\u003e\u003cspan class=\"RefSource\"\u003ehttps://doi.org/10.1017/CBO9780511818691\u003c/span\u003e\u003cspan address=\"10.1017/CBO9780511818691\" targettype=\"DOI\" class=\"RefTarget\"\u003e\u003c/span\u003e\u003c/span\u003e\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"rally-round-the-flag, COVID-19 pandemic, Russia–Ukraine war, Hungary, poll of polls","lastPublishedDoi":"10.21203/rs.3.rs-8523383/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-8523383/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003eClassic \u0026lsquo;rally-round-the-flag\u0026rsquo; theory posits uniform increases in incumbent support during national crises. Yet, in highly polarized democracies, we argue this conventional wisdom overlooks how crises can instead generate asymmetric dynamics across partisan blocs, often without a detectable aggregate increase in incumbent support. We study Hungary under long-term populist governance during two sequential shocks within a single electoral cycle: the onset of the COVID-19 pandemic (11 March 2020) and Russia\u0026rsquo;s full-scale invasion of Ukraine (24 February 2022), the latter occurring in the run-up to the 2022 parliamentary election. Using the Vox Populi archive of public polls, we estimate segmented interrupted time-series models with pollster fixed effects, survey-mode controls where recorded, and precision weights based on trimmed sample size. Across specifications, neither crisis produces an apparent surge in support for an incumbent among likely voters. Instead, the war breakpoint coincides with a discrete weakening on the opposition side (\u0026asymp;\u0026thinsp;4\u0026ndash;5 percentage points in the baseline specification), with little evidence of compensating post-war trend change; this inference is directionally robust, but conservative ARMA(1,1) models attenuate the estimated discontinuity. Pulse-window estimates in the mass electorate indicate at most a modest and short-lived COVID-era uptick for the incumbent, whereas opposition declines are larger around the war window. Three original surveys (2021, March 2022, May 2022) help interpret these aggregate patterns: pandemic pessimism concentrates among opposition publics, whereas the war is interpreted through a \u0026lsquo;peace versus war\u0026rsquo; frame aligned with the government\u0026rsquo;s campaign narrative. Overall, our findings suggest that sequential crises in polarized contexts reshape electoral competition not through uniform national rallies, but via asymmetric mechanisms, including bloc-specific vulnerabilities that contract opposition support and differential issues framing that benefits the incumbent. This reconceptualization of rally effects has important implications for understanding crisis politics in an era of heightened polarization.\u003c/p\u003e","manuscriptTitle":"Rally Without a Surge? Sequential Crises and Asymmetric Mobilization in Polarized Contexts","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-02-02 11:28:06","doi":"10.21203/rs.3.rs-8523383/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f027d318-94b2-4be3-b06a-3bb98d4aa9fe","owner":[],"postedDate":"February 2nd, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2026-04-11T21:53:39+00:00","versionOfRecord":[],"versionCreatedAt":"2026-02-02 11:28:06","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-8523383","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-8523383","identity":"rs-8523383","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}