{"paper_id":"396162ef-b197-4c62-96aa-dbff46a38594","body_text":"Data driven shoe design improves running economy beyond state-of-the-art Advanced Footwear Technology running shoes | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return\"[object Function]\"==o.call(a)}function e(a){return\"string\"==typeof a}function f(){}function g(a){return!a||\"loaded\"==a||\"complete\"==a||\"uninitialized\"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){(\"c\"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){\"img\"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),\"object\"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height=\"0\",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),\"img\"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||\"j\",e(a)?i(\"c\"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName(\"script\")[0],o={}.toString,p=[],q=0,r=\"MozAppearance\"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&\"[object Opera]\"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?\"object\":l?\"script\":\"img\",v=l?\"script\":u,w=Array.isArray||function(a){return\"[object Array]\"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split(\"!\"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split(\"=\"),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(\".\").pop().split(\"?\").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split(\"/\").pop().split(\"?\")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&\"css\"==i.url.split(\".\").pop().split(\"?\").shift()?\"c\":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (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];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results Data driven shoe design improves running economy beyond state-of-the-art Advanced Footwear Technology running shoes John Kuzmeski , Montgomery Bertschy , Laura Healey , View ORCID Profile Zach Barrons , View ORCID Profile Wouter Hoogkamer doi: https://doi.org/10.1101/2025.04.13.648601 John Kuzmeski a Integrative Locomotion Laboratory, Department of Kinesiology, University of Massachusetts , Amherst, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site For correspondence: jkuzmeski{at}umass.edu Montgomery Bertschy a Integrative Locomotion Laboratory, Department of Kinesiology, University of Massachusetts , Amherst, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Laura Healey b PUMA SE , Herzogenaurach, Germany Find this author on Google Scholar Find this author on PubMed Search for this author on this site Zach Barrons a Integrative Locomotion Laboratory, Department of Kinesiology, University of Massachusetts , Amherst, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Zach Barrons Wouter Hoogkamer a Integrative Locomotion Laboratory, Department of Kinesiology, University of Massachusetts , Amherst, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Wouter Hoogkamer Abstract Full Text Info/History Metrics Preview PDF Abstract Advanced Footwear Technology (AFT) has enabled remarkable improvements in running performance over traditional marathon racing shoes. However, reported differences between state-of-the-art AFT models are small, and vary across individuals. To assess if the benefits of AFT have been fully realized or if further running economy improvements can be unlocked using modern computational design and optimization techniques, we compared a prototype AFT shoe developed using a data-driven computational design process (PUMA Fast-R 3; FR3) against state-of-the-art AFT models. We quantified running economy for 15 trained runners (11M, 4F) in this prototype AFT shoe and three commercially available AFT models: the PUMA Fast-R 2 (FR2), the Nike Alphafly 3 (NIKE), and the Adidas Adios Pro Evo (ADI). Running economy in the FR3 was 3.15 ± 1.24%, 3.62 ± 1.25%, and 3.54 ± 1.16% better than in the FR2, NIKE and ADI (all p < 0.001), respectively, and every individual performed best in the FR3 shoes. While step parameters were similar between FR3 and FR2, the FR3 had a lower step frequency than the ADI (p = 0.013) and longer contact time than the NIKE and ADI (both p<0.001). Our results suggest that computational design analysis is a promising frontier in performance running shoe design, offering potential for further improvements and personalized AFT models. Introduction In the past decade, running shoe innovation has led to exciting improvements in running performance 1 – 5 . Hoogkamer et al. 6 were the first to study a novel type of marathon racing shoe featuring a modern, compliant and resilient foam and an embedded carbon fiber plate. They observed running economy improvements of 4% over traditional marathon racing shoes. Shoes with these and similar features are commonly referred to as “super shoes”, and as Advanced Footwear Technology (AFT) among researchers 7 . Since then, almost all running shoe brands have developed their own AFT model targeting improved running economy and performance. The individual contributions of reduced mass, increased longitudinal bending stiffness from carbon fiber plates, increased compliance and resilience from modern foams, taller stack height, and pronounced rocker profile are still topic of debate 8 – 14 , but virtually all state-of-the-art marathon racing shoes leverage some combination of these properties. While several studies have evaluated and compared the first AFT model (Nike Vaporfly 4%) in different groups and under different conditions 15 – 23 , less have evaluated its successor (Nike Vaporfly next%) 24 – 28 , or other more recent state-of-the-art marathon racing shoes 29 – 34 . Insights from this body of literature suggest that the best AFT models have on average 3-4% better running economy than traditional marathon racing shoes, but that running economy differences between state-of-the-art AFT models are 1-1.5% on average, at best. For example, Denis et al. 29 compared running economy between Adidas Adios Pro 3 and 4 models, released in June 2022 and January 2025 respectively, and observed a significant running economy improvement of 1.3% on average. This plateau in running economy gains over recent years raises the question of whether the benefits of AFT have been fully realized or if further improvements can be unlocked using modern computational design and optimization techniques. Advancements in computational modeling and simulation provide product developers with powerful tools to virtually design, evaluate, and optimize a product before physical prototyping. Not only can such computational tools significantly reduce development time and resource costs, but they also enable a more comprehensive understanding of how the various design features influence product performance. In the case of footwear design, this process would typically begin with the creation of a parametric computer-aided design (CAD) model of a shoe or one of its components 35 . The CAD model is then converted into a parametric finite element model (FEM). Finite element analysis (FEA) is used to simulate forces experienced during running and evaluate key performance metrics 35 . Optimization algorithms can then iteratively tune design parameters of the FEM to meet specific performance goals (e.g., minimize mass and maximize energy return). While such techniques are commonly used in various engineering domains, it remains unclear to what extent these methods are being applied in the practice of footwear design, as companies seldom disclose their internal design workflows. Here, we compared running economy between three state-of-the-art AFT models and a prototype shoe model developed using data-driven computational design analysis. We hypothesized that running economy would be better in the prototype than in the commercially available state-of-the-art AFT models and anticipated no detectable differences between the existing AFT models. Methods Shoe conditions The prototype shoe (PUMA Fast-R 3) was developed by PUMA. To start, a virtual shoe model was created using the known geometry and material specifications of the PUMA Fast-R 2. Biomechanical data were recorded for 10 runners running at 14-18 km/h in the PUMA Fast-R 2. This data included time series of in-shoe pressure data (XSENSOR Technology Corporation, Calgary, Canada), ground reaction forces (Treadmetrix, Park City, UT, USA), and 3D kinematics from the ankle and metatarsal-phalangeal joint (Cameras: OptiTrack Color Prime, NaturalPoint Inc., Corvallis, OR, USA; Software: Theia 3D, Theia Markerless Inc., Kingston, Canada). These biomechanics data were used to simulate how the shoe behaves during running. Using dynamic computational simulations with the runners’ data, PUMA applied topology optimization to identify the optimal foam placement, material properties, carbon fiber layering, and overall geometry to minimize weight, maximize energy storage and return, and achieve the desired carbon plate stiffness. Upon receiving the final prototype shoe from PUMA, we compared it against three state-of-the-art, commercially available AFT models, without further involvement of PUMA. We included the Nike Alphafly 3 (NIKE), Adidas Adios Pro Evo 1 (ADI), and PUMA Fast-R 2 (FR2) ( Figure 1 ). At the start of the testing phase the NIKE and ADI shoes were the shoes worn for the male and female marathon world records, respectively. Download figure Open in new tab Figure 1. From left to right: Nike Alphafly 3 (NIKE), Adidas Adios Pro Evo 1 (ADI), PUMA Fast-R2 (FR2), PUMA Fast-R 3 (FR3). Material properties We tested each shoe model’s material properties with an Instron E10000 (Instron, Canton, MA, USA). We used a custom WaveMetrix script to control two different missile heads, a flat round missile with a diameter of 45mm to mimic the size of the calcaneus for rearfoot testing and a full foot last with heel to toe drop of 10 mm to assure even contact across the midsole. Each missile head was force-controlled to mimic the peak force for the corresponding region of the shoe: 1200 N for the rear foot and 2000 N for the full foot. Additionally, each time period was adjusted to mimic the loading duration for each area based on insole data from Honert et al. 36 : 140 ms for the rearfoot and 185 ms for the full foot. Each shoe was subjected to 100 half sine wave load cycles, with the final 10 load cycles averaged for analysis. Longitudinal bending stiffness was tested through a 3-point bending test. Each shoe was placed upon two parallel supports located 37 . The Instron tip was aligned with the metatarsal phalangeal joint, in the center of the supports, and followed a position controlled trapezoidal waveform while force was recorded 37 . This waveform depressed 7.5 mm into the midsole at 10 mm/s, was held that position for 0.5 seconds and then raised 7.5 mm at 10mm/s. This cycle was repeated 20 times, with the final 10 load cycles averaged for analysis. Shoe properties and mechanical testing results are summarized in Table 1 . View this table: View inline View popup Download powerpoint Table 1. Material properties and mechanical testing results for the Nike Alphafly 3 (NIKE), Adidas Adios Pro Evo 1 (ADI), PUMA Fast-R 2 (FR2), and PUMA Fast-R 3 (FR3) for the men’s US 9 sized shoes. View this table: View inline View popup Download powerpoint Table 2. Spatiotemporal and kinetic outcomes between the four shoe conditions. NIKE: Nike Alphafly 3, ADI: Adidas Adios Pro Evo 1, FR2: PUMA Fast-R 2, FR3: PUMA Fast-R 3. * denotes significant effect of footwear condition. Participants Fifteen runners participated in this study, 11 males (age: 31.5 ± 6.8 years old, height: 1.78 ± 0.04 m, mass: 67.5 ± 4.8 kg) and 4 females (age: 28.3 ± 3.4 years old, height: 1.66 ± 0.01 m, mass: 59.1 ± 2.2 kg). To be included in the study, participants were required to have recently run a 5-km in less than 21 minutes (females) or 19 minutes (males) or an equivalent race result over a different distance, and to fit a shoe size US W8, M9 or M11. The study was approved by the University of Massachusetts Amherst Institutional Review Board (#5924). All participants signed written informed consent forms before participating. Running Protocol Upon arriving at the lab for testing participants warmed up as they would for a high tempo workout, on a treadmill or outside. Once fully warmed up participants ran on a rigid force instrumented treadmill (Treadmetrix, Park City, UT, USA) for 2 minutes to familiarize themselves with the treadmill and mouthpiece. For the experimental protocol, the participants ran in each of the four shoe models twice in randomized, mirrored order (ABCDDCBA) for 5 minutes, followed by 5 minutes rest, for a total of eight running trials, while we measured expired air (TrueOne 2400, ParvoMedics, Salt Lake City, UT, USA). Running speed for each runner was set based on experience and recent 5k race results. Eight males ran at 16 km/h, three males ran at 14 km/h, three females ran at 14 km/h, one female ran at 12.9 km/h (7:30 min mile pace). The expired gas analysis system was calibrated before the start of the first trial. We recorded ground reaction force data at 1000 Hz during the last 30 s of each trial. Data analysis We calculated running economy as average metabolic rate (W/kg) over the last 2 minutes of each trial, based on the measured rates of oxygen uptake and carbon-dioxide production 38 . We filtered ground reaction force data using a dual-pass Butterworth filter with a 20 Hz cut-off 16 , 39 . Touch down and toe off were determined using a 25 N vertical ground reaction force threshold to calculate ground contact time and step frequency. We calculated peak vertical ground reaction force, and braking and propelling impulses during the contact phase. Statistics We used linear mixed-effects models to evaluate the effects of shoe model on running economy and kinetic outcomes. We used a traditional level of significance (α = 0.05), with Tukey’s HSD test corrections on post hoc analysis. Shoe conditions were used as the fixed effect, participants as a random intercept, and running economy and kinetic outcomes each as output variable. Results We observed a significant effect of shoe type on running economy ( Fig. 2 ). Post-hoc pairwise comparisons, revealed that the FR3 had a significantly better running economy (15.82 ± 1.72 W/kg) compared to the FR2 (16.33 ± 1.67 W/kg; p < 0.001), NIKE (16.42 ± 1.83 W/kg; p < 0.001), and ADI (16.41 ± 1.86 W/kg; p < 0.001). For all 15 participants running economy was best in the FR3. The average percent running improvements in the FR3 were 3.15 ± 1.24%, 3.62 ± 1.25%, and 3.54 ± 1.16% vs. the FR2, NIKE and ADI, respectively. There were no significant differences in running economy between the FR2 and ADI (p = 0.515), FR2 and NIKE (p = 0.406), or NIKE and ADI (p = 0.998). The respiratory exchange ratio was below 1.0 for all participants during all trials (0.92 ± 0.02). Download figure Open in new tab Figure 2. A) Running economy, quantified in metabolic rate, was best in the PUMA Fast-R 3 (FR3) vs. the Nike Alphafly 3 (NIKE), Adidas Adios Pro Evo 1 (ADI), and PUMA Fast-R 2 (FR2). B) Percent improvement in the FR3 relative to the other shoes varied substantially between individuals and shoes. There was a significant effect of shoe on step frequency (p < 0.001), contact time (p < 0.001), and braking/propelling impulses (p < 0.001). Step frequency was higher in the ADI than in the FR2 (p < 0.001) and FR3 (p = 0.013), but not than in the NIKE (p = 0.083). It was similar for the other between shoe comparisons. Contact time was similar in the FR2 and FR3 (p = 0.763), both longer than in the ADI and NIKE (all p < 0.001). ADI and NIKE were not significantly different from each other (p = 0.154). Braking and propelling impulses were higher in the FR3 than in the ADI (p < 0.001) and NIKE (p = 0.024), but not than in the FR2 (p = 0.794). They were also higher in the FR2 than in the ADI (p = 0.002), but not than in the NIKE (p = 0.170). They were similar in ADI and NIKE (p = 0.327). Discussion The primary finding of this study supports our hypothesis: the prototype shoe (FR3), developed through a data-driven computational design process, yields significantly better running economy compared to three commercially available state-of-the-art AFT models (NIKE, ADI, FR2). Specifically, running in the FR3 resulted in an average group improvement in running economy between 3.1 and 3.6% in relation to these state-of-the-art models, with each participant having the best running economy in the FR3. Furthermore, there was no significant difference in running economy between each of the commercially available state-of-the-art models with variation across participants in which of these shoes yielded the second-best running economy. The most striking result from this study was the consistency of the running economy response to the prototype shoe ( Fig. 2 ). Every participant recorded their best running economy while running in the FR3. While a similar consistent improvement has been reported for AFT before, this was relative to traditional racing shoes 6 , 15 . Others have reported inconsistent responses for running economy in AFT vs. traditional racing shoes, with some individuals showing no improvements or slightly worse running economy in AFT 19 , 20 . Running economy differences between AFT models are typically small and inconsistent between participants 29 , 31 , 32 . A common explanation for these inconsistencies is that there are “responders” and “non-responders” 40 – 42 . The unanimous positive response to the FR3 in our population highlights a further advancement in running economy for racing shoes, a step similar to the introduction of AFT, suggesting computational design analysis is a promising new frontier in performance running shoes design. The observed between shoe differences in running economy were not mirrored by changes in spatiotemporal and kinetic variables. Participants ran with a lower step frequency in the FR3 than in the ADI, while contact time and braking/propelling impulses were larger in the FR3 than in the ADI and NIKE. Longer contact times could partially explain the improved running economy 43 . However, FR3 also led to a significantly improved running economy over its predecessor (FR2) despite no spatiotemporal or kinetic differences. While the difference between traditional marathon racing shoes and AFT in appearance and properties is striking, the differences between the FR3 and the three other shoes is much more subtle. The computational design process focused on optimizing footwear characteristics known or expected to improve running economy such as mass, carbon fiber plate geometry and stiffness, and foam properties and distribution for maximal mechanical energy storage and return. While others have combined shoe finite element analysis with musculoskeletal models to estimate bone loading 44 or have used musculoskeletal simulations to predict differences in running economy from different midsole foams 45 , the current process only focused on shoe performance. Through the optimization process, the mass of the FR3 (167g) came out 82 g lighter than the FR2, which is relatively heavy (249g) compared to the competitor AFT models (NIKE 197g, ADI 133g). The 1% per 100g rule 46 , 47 indicates a ∼0.8% better running economy in the FR3 than in the FR2, which leaves ∼2.3% improvement in running economy to be accounted for by improvements in plate geometry and stiffness and foam energy storage and return. Moreover, the FR3 is substantially less stiff in longitudinal bending than the FR2 (16.3 vs. 26.4 N/mm). Since we assessed this with a three-point bending test, part of the compliance comes from foam deformation rather than plate bending 48 . Compression stiffness was lower in the FR3 than in the FR2 for both rear foot and full foot compression. Percent energy return in the FR3 was higher than in the FR2, both in compression and in bending. While not the lightest, as compared to the three other shoes, the FR3 had the lowest compression stiffness, the lowest bending stiffness, and the highest percent energy return, suggesting that the observed improved running economy is a combination of all of these factors. While all participants had their best running economy in the FR3, running economy in the three state-of-the-art AFT models varied substantially between individuals. The largest difference among the three AFT models was 4.1% for one individual (ADI vs. FR2), and across individuals running economy relative to the FR3 varied from 1.6-6.0% in NIKE, 1.3-5.6% in ADI and 1.7-5.5% in FR2. Such inter-individual variability is in-line with other studies using multiple trials in each shoe to assess running economy difference, report running economy improvements beyond traditional racing shoes ranging from ∼1% to ∼7% 6 (for review see 49 ), but smaller than controversial findings showing >10% improvements and deteriorations in running economy between AFT models for some individuals 49 , 50 . This 4 - 5% range in individual running economy values in the same shoes and the observed individual changes between the AFT models, confirms that some runners might have more benefits from specific AFT models than other runners, supporting the potential of personalized AFT design 51 , 52 . Here we show that AFT shoes developed using a data-driven computational design process improve running economy beyond these interindividual differences, as all participants had their best running economy in the FR3, but we see great potential for using data-driven computational design processes to optimize AFT for an individual runner. Running economy is one of the three major determinants of marathon performance 53 , 54 , with recent insights suggesting a role for physiological resilience as well 55 . Indeed, improvements in running economy translate to improvements in running performance, but this relation is not one-to-one 56 . Kipp et al 57 quantified this relation accounting for the inherent curvilinear relation between running speed and running economy 58 , 59 , and the additional metabolic cost of overcoming air resistance during overground running, which increases disproportionally with increasing speed 60 . This relation indicates that the actual performance benefits from an improvement in running economy (e.g., time saved in a race) depend on the running velocity. Specifically, at faster running speeds the percent time savings are less than the percent improvement in running economy, which is also what is commonly reported when analyzing race result improvements since the introduction of AFT 1 – 5 . To put the observed running economy improvements in context, we used the framework by Kipp et al. 57 to calculate the theoretical improvements in marathon time from a 3.15% improvement in running economy (FR3 vs. FR2) for different baseline marathon times, for a sample runner (mass 65.6 kg and height 1.75 m) 61 , 62 ( Table 3 ). Future race result analyses are needed to evaluate these predicted time improvements. View this table: View inline View popup Download powerpoint Table 3. Predicted marathon time improvements from a 3.15% improvement in running economy for different baseline marathon times. Note that we only evaluated running economy for runners running at 12.9 (n = 1), 14 (n = 6) and 16 (n = 8) km/h, which limits extrapolation of our findings to elite runners (18km/h or faster) and recreational runners (12 km/h or slower) 27 . Furthermore, these footwear conditions may have a different effect if the participants are able to adapt over the course of several days or weeks, however all participants were used to regularly running in AFT models. By design, we compared state-of-the-art AFT models as is, with many differences in foam and plate properties, geometry and mass. As such our findings provide limited insights into the exact features that explain the observed differences in running economy. It might be that using more trials per condition (e.g., four) would have resulted in significant running economy differences between the three commercially available AFT models 63 , or that a larger sample size would have resulted in additional significant between shoe differences in spatiotemporal measures (e.g., step frequency in NIKE vs. ADI). Finally, we only assessed running economy over 5-minute trials and while changes in lab-based measures of running economy translate to running performance as discussed above, we did not assess how these shoes might compare under fatigued conditions to be anticipated 35 kilometers into a marathon. Similarly, we assessed shoes that were relatively fresh, used less than 40 km use per pair, we did not assess how these shoes compare after substantial use 64 . Conclusion To assess if the benefits of AFT have been fully realized, we compared a prototype AFT shoe developed using a data-driven computational design process (FR3) to commercially available AFT models. The FR3 was developed starting from a virtual shoe model with the known geometry and material specifications of the FR2. Time series of in-shoe pressure data, ground reaction forces, and ankle and metatarsal-phalangeal joint kinematics from 10 runners were used to simulate how the shoe behaves during running and topology optimization was applied to identify the optimal foam placement, material properties, carbon fiber layering, and overall geometry to minimize weight, maximize energy storage and return, and achieve the desired carbon plate stiffness. Running economy improved over 3% in the FR3 vs. the FR2 and two competitor AFT models (NIKE and ADI). While step parameters were similar between FR3 and FR2, the FR3 had a lower step frequency than the ADI (p = 0.013) and longer contact time than the NIKE and ADI (both p<0.001). Our results suggest that computational design analysis is a promising frontier in performance running shoe design, offering potential for further improvements and personalized AFT models. Author contributions Conceptualization: John Kuzmeski, Montgomery Bertschy, Laura Healey, Zach Barrons, Wouter Hoogkamer; Experimental design: John Kuzmeski, Montgomery Bertschy, Zach Barrons, Wouter Hoogkamer; Data collection: John Kuzmeski, Montgomery Bertschy, Zach Barrons; Data analysis: John Kuzmeski, Montgomery Bertschy, Zach Barrons; Writing - original draft preparation: John Kuzmeski, Montgomery Bertschy, Laura Healey, Wouter Hoogkamer; Writing - review and editing: John Kuzmeski, Montgomery Bertschy, Laura Healey, Zach Barrons, Wouter Hoogkamer Funding PUMA SE provided the shoes and funding for this project. Disclosure statement Wouter Hoogkamer has received research grants from PUMA and Saucony. Laura Healey is an employee of PUMA. PUMA did not have any influence on the views presented in this article. Acknowledgements We thank the subjects for participating, Joshua Cohen and Caroline Kolmodin for helping with data collection, and Dr. Meghan Huber and Dr. Key Nahan for proofreading this manuscript. Footnotes Funding PUMA SE provided the shoes and funding for this project. Disclosure statement Wouter Hoogkamer has received research grants from PUMA and Saucony. Laura Healey is an employee of PUMA. PUMA did not have any influence on the views presented in this article. References 1. ↵ Bermon , S. , Garrandes , F. , Szabo , A. , Berkovics , I. & Adami , P. E. Effect of Advanced Shoe Technology on the Evolution of Road Race Times in Male and Female Elite Runners . Frontiers in Sports and Active Living 3 , ( 2021 ). 2. Mason , J. , Starc , L. , Morin , J.-B. , McClelland , E. L. & Zech , A. Can the recent sex-specific evolutions in elite running performances be attributed to advanced footwear technology? Frontiers in Sports and Active Living 6 , ( 2024 ). 3. Needles , B. J. & Grabowski , A. M. Does running speed affect the performance improvements experienced by elite distance runners wearing advanced footwear technology spikes? Journal of Applied Physiology 137 , 1690 – 1698 ( 2024 ). OpenUrl CrossRef PubMed 4. Rodrigo Carranza , V. , González-Mohíno , F. , Santos-Concejero , J. & Gonzalez Rave , J. M. Impact of advanced footwear technology on elite men’s in the evolution of road race performance Impact of advanced footwear technology on elite men’s in the evolution of road race performance . Journal of Sports Sciences 40 , ( 2023 ). 5. ↵ Senefeld , J. W. , Haischer , M. H. , Jones , A. M. , Wiggins , C. C. , Beilfuss , R. , Joyner , M. J. & Hunter , S. K. Technological advances in elite marathon performance . Journal of Applied Physiology 130 , 2002 – 2008 ( 2021 ). OpenUrl CrossRef PubMed 6. ↵ Hoogkamer , W. , Kipp , S. , Frank , J. H. , Farina , E. M. , Luo , G. & Kram , R. A Comparison of the Energetic Cost of Running in Marathon Racing Shoes . Sports Medicine 48 , 1009 – 1019 ( 2018 ). OpenUrl CrossRef PubMed 7. ↵ Frederick , E. C. Let’s just call it advanced footwear technology (AFT) . Footwear Science 14 , 131 – 131 ( 2022 ). OpenUrl CrossRef 8. ↵ Barrons , Z. B. , Wannop , J. W. & Stefanyshyn , D. J. The Influence of Footwear Midsole Thickness on Running Economy and Frontal Plane Ankle Stability . Footwear Science 15 , 155 – 160 ( 2023 ). OpenUrl CrossRef 9. Bertschy , M. , Lino , H. , Healey , L. & Hoogkamer , W. Is increasing the effective leg length of a human runner metabolically beneficial? 2025.01.03.631222 Preprint at doi: 10.1101/2025.01.03.631222 ( 2025 ). OpenUrl Abstract / FREE Full Text 10. Kram , R. Ergogenic distance running shoes: how do we think they work and how can we understand them better? Footwear Science 14 , 139 – 146 ( 2022 ). OpenUrl CrossRef 11. Nigg , B. M. , Cigoja , S. & Nigg , S. R. Teeter-totter effect: a new mechanism to understand shoe-related improvements in long-distance running . British Journal of Sports Medicine 55 , 462 – 463 ( 2021 ). OpenUrl FREE Full Text 12. Ortega , J. A. , Healey , L. A. , Swinnen , W. & Hoogkamer , W. Energetics and Biomechanics of Running Footwear with Increased Longitudinal Bending Stiffness: A Narrative Review . Sports Medicine 51 , 873 – 894 ( 2021 ). OpenUrl CrossRef PubMed 13. Ruiz-Alias , S. A. , Pérez-Castilla , A. , Soto-Hermoso , V. M. & García-Pinillos , F. Influence of the carbon fiber plate curvature of advanced footwear technology on the running energetic cost and 3000-m performance . Sports Engineering 27, 21 ( 2024 ). 14. ↵ Shorten , M. R. Energy return in footwear – revisited . Footwear Science 0 , 1 – 14 ( 2024 ). OpenUrl 15. ↵ Barnes , K. R. & Kilding , A. E. A Randomized Crossover Study Investigating the Running Economy of Highly-Trained Male and Female Distance Runners in Marathon Racing Shoes versus Track Spikes . Sports Medicine 49 , 331 – 342 ( 2019 ). OpenUrl CrossRef PubMed 16. ↵ Healey , L. A. & Hoogkamer , W. Longitudinal bending stiffness does not affect running economy in Nike Vaporfly Shoes . Journal of Sport and Health Science 11 , 285 – 292 ( 2022 ). OpenUrl CrossRef 17. Hébert-Losier , K. , Finlayson Steven J. Lamb , Peter F. , Driller, Matthew W. Hanzlíková , Ivana , Dubois , Blaise , Esculier Jean-Francois & and Beaven , C. M. Kinematics of recreational male runners in “super”, minimalist and habitual shoes . Journal of Sports Sciences 40 , 1426 – 1435 ( 2022 ). OpenUrl CrossRef PubMed 18. Hébert-Losier , K. , Pfister , A. , Finlayson , S. J. , Esculier , J.-F. , Lamb , P. & Beaven , C. M. Are super shoes a super placebo? A randomised crossover trial in female recreational runners . Footwear Science 1 – 10 ( 2025 ). 19. ↵ Hébert-Losier , K. , Finlayson , S. J. , Driller , M. W. , Dubois , B. , Esculier , J.-F. & Beaven , C. M. Metabolic and performance responses of male runners wearing 3 types of footwear: Nike Vaporfly 4%, Saucony Endorphin racing flats, and their own shoes . Journal of Sport and Health Science 11 , 275 – 284 ( 2022 ). OpenUrl CrossRef 20. ↵ Hunter , I. , McLeod , A. , Valentine , D. , Low , T. , Ward , J. & Hager , R. Running economy, mechanics, and marathon racing shoes . Journal of Sports Sciences 37 , 2367 – 2373 ( 2019 ). OpenUrl CrossRef PubMed 21. Nielsen , A. , Franch , J. , Heyde , C. , Zee , M. de , Kersting , U. & Larsen , R. G. Carbon Plate Shoes Improve Metabolic Power and Performance in Recreational Runners . International Journal of Sports Medicine 43 , 804 – 810 ( 2022 ). OpenUrl CrossRef PubMed 22. Subramanium , A. , Vienneau , J. , Nigg , S. R. & Nigg , B. M. A methodological proof-of-concept of the teeter-totter effect . Journal of Sports Sciences 42 , 1432 – 1438 ( 2024 ). OpenUrl CrossRef PubMed 23. ↵ Whiting , C. S. , Hoogkamer , W. & Kram , R. Metabolic cost of level, uphill, and downhill running in highly cushioned shoes with carbon-fiber plates . Journal of Sport and Health Science 11 , 303 – 308 ( 2022 ). OpenUrl CrossRef 24. ↵ Hata , K. , Hamamura , Y. , Noro , H. , Yamazaki , Y. , Nagato , S. , Kanosue , K. & Yanagiya , T. Plantar Flexor Muscle Activity and Fascicle Behavior in Gastrocnemius Medialis During Running in Highly Cushioned Shoes With Carbon-Fiber Plates . Journal of Applied Biomechanics 1 , 1 – 9 ( 2024 ). OpenUrl CrossRef 25. Hata , K. , Noro , H. , Takeshita , T. , Yamazaki , Y. & Yanagiya , T. Leg stiffness during running in highly cushioned shoes with a carbon-fiber plate and traditional shoes . Gait & Posture 95 , 9 – 14 ( 2022 ). OpenUrl CrossRef PubMed 26. Martinez , E. , Hoogkamer , W. , Powell , D. W. & Paquette , M. R. The Influence of ‘Super-Shoes’ and Foot Strike Pattern on Metabolic Cost and Joint Mechanics in Competitive Female Runners . Medicine & Science in Sports & Exercise 56 , 1337 – 1344 ( 2024 ). OpenUrl CrossRef PubMed 27. ↵ Joubert , D. P. , Dominy , T. A. & Burns , G. T. Effects of Highly Cushioned and Resilient Racing Shoes on Running Economy at Slower Running Speeds . International Journal of Sports Physiology and Performance 18 ( 2 ), 164 – 170 . ( 2023 ). OpenUrl CrossRef 28. ↵ Hannigan , J. J. , Westley , L. & and Jin , L. Injury and performance-related running biomechanics in advanced footwear technology compared to minimalist footwear . Footwear Science 16 , 115 – 121 ( 2024 ). OpenUrl CrossRef 29. ↵ Denis , Y. , Knoke , H. , Braun , L. & Willwacher , S. New year, new gear: Does the next version of an advanced footwear model lead to better performance? Preprint at doi: 10.51224/SRXIV.482 ( 2024 ). OpenUrl CrossRef 30. Hudgins , J. H. , Pastina , J. T. , Gillis , I. L. , Abel , M. G. , Bergstrom , H. C. , Black , W. S. & Best , S. A. The Ability of Stryd Footpod Metrics to Reflect Changes in Metabolic Power Between Running Shoe Types . Journal of Sports Sciences 42 , 2229 – 2241 ( 2024 ). OpenUrl CrossRef PubMed 31. ↵ Joubert , D. P. & Jones , G. P. A comparison of running economy across seven highly cushioned racing shoes with carbon-fibre plates . Footwear Science 14 , 71 – 83 ( 2022 ). OpenUrl CrossRef 32. ↵ Perry , D. , Lino , H. , Bertschy , M. & Hoogkamer , W. Effects of longitudinal bending stiffness and midsole foam on running energetics . Footwear Science 17 , 3 – 9 ( 2025 ). OpenUrl CrossRef 33. Riedl , M. , von Diecken , C. & Ueberschär , O. One Shoe to Fit Them All? Effect of Various Carbon Plate Running Shoes on Running Economy in Male and Female Amateur Triathletes and Runners at Individual Training and Race Paces . Applied Sciences 14, 11535 ( 2024 ). 34. ↵ Werkhausen , A. , Lund-Hansen , M. , Wiedenbruch , L. , Peikenkamp , K. & Rice , H. Technologically advanced running shoes reduce oxygen cost and cumulative tibial loading per kilometer in recreational female and male runners . Scientific Reports 14, 11903 ( 2024 ). 35. ↵ Song , Y. , Shao , E. , Bíró , I. , Baker , J. S. & Gu , Y. Finite element modelling for footwear design and evaluation: A systematic scoping review . Heliyon 8 , ( 2022 ). 36. ↵ Honert , E. C. , Hoitz , F. , Blades , S. , Nigg , S. R. & Nigg , B. M. Estimating Running Ground Reaction Forces from Plantar Pressure during Graded Running . Sensors 22 , 3338 ( 2022 ). OpenUrl CrossRef PubMed 37. ↵ Roy , J.-P. R. & Stefanyshyn , D. J. Shoe Midsole Longitudinal Bending Stiffness and Running Economy, Joint Energy, and EMG . Medicine & Science in Sports & Exercise 38 , 562 – 569 ( 2006 ). OpenUrl CrossRef PubMed 38. ↵ Péronnet , F. & Massicotte , D. Table of nonprotein respiratory quotient: an update . Canadian Journal of Sport Sciences = Journal Canadien des Sciences du Sport 16 , 23 – 9 ( 1991 ). OpenUrl PubMed 39. ↵ Hoogkamer , W. , Kipp , S. & Kram , R. The Biomechanics of Competitive Male Runners in Three Marathon Racing Shoes: A Randomized Crossover Study . Sports Medicine 49 , 133 – 143 ( 2019 ). OpenUrl CrossRef PubMed 40. ↵ Brund , R. B. K. , Nielsen , A. , Godsk Larsen , R. , Kersting , U. G. & de Zee , M. Defining responders to altered running footwear based on measures of running economy . Footwear Science 17 , 29 – 35 ( 2025 ). OpenUrl CrossRef 41. Chollet , M. , Michelet , S. , Horvais , N. , Pavailler , S. & Giandolini , M. Individual physiological responses to changes in shoe bending stiffness: a cluster analysis study on 96 runners . European Journal of Applied Physiology 123 , 169 – 177 ( 2023 ). OpenUrl CrossRef PubMed 42. ↵ Heyde , C. , Nielsen , A. , Roecker , K. , Godsk Larsen , R. , de Zee , M. , Kersting , U. G. & and Brund , R. B. K. The percentage of recreational runners that might benefit from new running shoes. A likely scenario . Footwear Science 14 , 163 – 172 ( 2022 ). OpenUrl CrossRef 43. ↵ Kram , R. & Taylor , C. R. Energetics of running: a new perspective . Nature 346 , 265 – 267 ( 1990 ). OpenUrl CrossRef PubMed Web of Science 44. ↵ Song , Y. , Cen , X. , Sun , D. , Bálint , K. , Wang , Y. , Chen , H. , Gao , S. , Bíró , I. , Zhang , M. & Gu , Y. Curved carbon-plated shoe may further reduce forefoot loads compared to flat plate during running . Scientific Reports 14, 13215 ( 2024 ). 45. ↵ Dorschky , E. , Krüger , D. , Kurfess , N. , Schlarb , H. , Wartzack , S. , Eskofier , B. M. & Van Den Bogert , A. J. Optimal control simulation predicts effects of midsole materials on energy cost of running . Computer Methods in Biomechanics and Biomedical Engineering 22 , 869 – 879 ( 2019 ). OpenUrl CrossRef PubMed 46. ↵ Franz , J. R. , Wierzbinski , C. M. & Kram , R. Metabolic cost of running barefoot versus shod: is lighter better . Medicine & Science in Sports & Exercise 44 , 1519 – 1525 ( 2012 ). OpenUrl CrossRef PubMed 47. ↵ Bachl N , Prokop L , Suckert R Frederick , E. , Daniels , J. & Hayes , J. The effect of shoe weight on the aerobic demands of running . In: Bachl N , Prokop L , Suckert R , editors. Current Topics in Sports Medicine . Vienna (Austria): Urban & Schwarzenberg 616 – 25 ( 1884 ). 48. ↵ Healey , L. , Bertschy , M. & Hoogkamer , W. Evaluating longitudinal bending stiffness testing for performance footwear . Footwear Science 13 , S5 – S6 ( 2021 ). OpenUrl CrossRef 49. ↵ Knopp , M. , Muñiz-Pardos , B. , Wackerhage , H. , Schönfelder , M. , Guppy , F. , Pitsiladis , Y. & Ruiz , D. Variability in Running Economy of Kenyan World-Class and European Amateur Male Runners with Advanced Footwear Running Technology: Experimental and Meta-analysis Results . Sports Medicine 53 , 1255 – 1271 ( 2023 ). OpenUrl CrossRef PubMed 50. ↵ Barrons , Z. B. , Rodrigo-Carranza , V. , Bertschy , M. & Hoogkamer , W. The Fallacy of Single Trials: The Need for Multiple Trials in Assessing Running Economy Responses in Advanced Footwear Technology . Sports Medicine ( 2024 ) doi: 10.1007/s40279-023-01991-1 . OpenUrl CrossRef 51. ↵ Frederick , E. C. , Stefanyshyn , Darren J. & and Hoogkamer , W. The rich research tapestry that foreshadowed advanced footwear technology and what it suggests for the future . Footwear Science 15 , S2 – S3 ( 2023 ). OpenUrl 52. ↵ Willwacher , S. & and Weir , G. The future of footwear biomechanics research . Footwear Science 15 , 145 – 154 ( 2023 ). OpenUrl CrossRef 53. ↵ Joyner , M. J. Modeling: optimal marathon performance on the basis of physiological factors . Journal of Applied Physiology 70 , 683 – 687 ( 1991 ). OpenUrl CrossRef PubMed Web of Science 54. ↵ Knopp , M. , Guppy , F. , Joyner , M. , Muniz-Pardos , B. , Wackerhage , H. , Schönfelder , M. , Pitsiladis , Y. & Ruiz , D. Validation of marathon performance model based on physiological factors in world-class East African runners: a case series . Translational Exercise Biomedicine 2 , 1 – 8 ( 2025 ). OpenUrl CrossRef 55. ↵ Jones , A. M. & Kirby , B. S. Physiological Resilience: What Is It and How Might It Be Trained? Scandinavian Journal of Medicine & Science in Sports 35, e70032 ( 2025 ). 56. ↵ Hoogkamer , W. , Kipp , S. , Spiering , B. A. & Kram , R. Altered Running Economy Directly Translates to Altered Distance-Running Performance . Medicine & Science in Sports & Exercise 48 , 2175 – 2180 ( 2016 ). OpenUrl CrossRef PubMed 57. ↵ Kipp , S. , Kram , R. & Hoogkamer , W. Extrapolating Metabolic Savings in Running: Implications for Performance Predictions . Frontiers in Physiology 10 , ( 2019 ). 58. ↵ Batliner , M. E. , Kipp , S. , Grabowski , A. M. , Kram , R. & Byrnes , W. C. Does Metabolic Rate Increase Linearly with Running Speed in all Distance Runners? Sports Medicine International Open 02 , E1 – E8 ( 2017 ). OpenUrl 59. ↵ Kipp , S. , Grabowski , A. M. & Kram , R. What determines the metabolic cost of human running across a wide range of velocities? Journal of Experimental Biology 221 , jeb184218 ( 2018 ). OpenUrl Abstract / FREE Full Text 60. ↵ Pugh , L. G. C. E. The influence of wind resistance in running and walking and the mechanical efficiency of work against horizontal or vertical forces . The Journal of Physiology 213 , 255 – 276 ( 1971 ). OpenUrl CrossRef PubMed Web of Science 61. ↵ Gordon , D. , Wightman Sarah , Basevitch , Itay , Johnstone , James , Espejo-Sanchez , Carolina , Beckford , Chelsea , Boal , Mariette , Scruton , Adrian , Ferrandino Mike & and Merzbach , V. Physiological and training characteristics of recreational marathon runners . Open Access Journal of Sports Medicine 8 , 231 – 241 ( 2017 ). OpenUrl CrossRef PubMed 62. ↵ da Silva , E. S. , Kram , R. & Hoogkamer , W. The metabolic cost of emulated aerodynamic drag forces in marathon running . Journal of Applied Physiology 133 , 766 – 776 ( 2022 ). OpenUrl CrossRef PubMed 63. ↵ Beltrami , F. G. , Baumann , G.-A. , Biedermann , K. & Spengler , C. M. On the Fallacy of Single Trials: A Proposal to Reduce Variability in Running Shoe Research . Sports Medicine ( 2025 ) doi: 10.1007/s40279-025-02180-y . OpenUrl CrossRef 64. ↵ Rodrigo-Carranza , V. , Hoogkamer , W. , González-Ravé , J. M. , Horta-Muñoz , S. , Serna-Moreno , M. D. C. , Romero-Gutierrez , A. & González-Mohíno , F. Influence of different midsole foam in advanced footwear technology use on running economy and biomechanics in trained runners . Scandinavian Journal of Medicine & Science in Sports 34 , e14526 ( 2024 ). OpenUrl CrossRef View the discussion thread. Back to top Previous Next Posted April 19, 2025. Download PDF Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following Data driven shoe design improves running economy beyond state-of-the-art Advanced Footwear Technology running shoes Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share Data driven shoe design improves running economy beyond state-of-the-art Advanced Footwear Technology running shoes John Kuzmeski , Montgomery Bertschy , Laura Healey , Zach Barrons , Wouter Hoogkamer bioRxiv 2025.04.13.648601; doi: https://doi.org/10.1101/2025.04.13.648601 Share This Article: Copy Citation Tools Data driven shoe design improves running economy beyond state-of-the-art Advanced Footwear Technology running shoes John Kuzmeski , Montgomery Bertschy , Laura Healey , Zach Barrons , Wouter Hoogkamer bioRxiv 2025.04.13.648601; doi: https://doi.org/10.1101/2025.04.13.648601 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Physiology Subject Areas All Articles Animal Behavior and Cognition (7635) Biochemistry (17697) Bioengineering (13894) Bioinformatics (41951) Biophysics (21455) Cancer Biology (18593) Cell Biology (25509) Clinical Trials (138) Developmental Biology (13380) Ecology (19903) Epidemiology (2067) Evolutionary Biology (24322) Genetics (15611) Genomics (22509) Immunology (17737) Microbiology (40398) Molecular Biology (17183) Neuroscience (88619) Paleontology (667) Pathology (2833) Pharmacology and Toxicology (4825) Physiology (7644) Plant Biology (15158) Scientific Communication and Education (2046) Synthetic Biology (4296) Systems Biology (9825) Zoology (2271)","source_license":"CC-BY-4.0","license_restricted":false}