Recommendations for marathon runners: on the application of recommender systems and machine learning to support recreational marathon runners.

Every year millions of people, from all walks of life, spend months training to run a traditional marathon. For some it is about becoming fit enough to complete the gruelling 26.2 mile (42.2 km) distance. For others, it is about improving their fitness, to achieve a new personal-best finish-time. In this paper, we argue that the complexities of training for a marathon, combined with the availability of real-time activity data, provide a unique and worthwhile opportunity for machine learning and for recommender systems techniques to support runners as they train, race, and recover. We present a number of case studies-a mix of original research plus some recent results-to highlight what can be achieved using the type of activity data that is routinely collected by the current generation of mobile fitness apps, smart watches, and wearable sensors.

Author multidisciplinarity and disciplinary roles in field of study networks.

When studying large research corpora, "distant reading" methods are vital to understand the topics and trends in the corresponding research space. In particular, given the recognised benefits of multidisciplinary research, it may be important to map schools or communities of diverse research topics, and to understand the multidisciplinary role that topics play within and between these communities. This work proposes Field of Study (FoS) networks as a novel network representation for use in scientometric analysis. We describe the formation of FoS networks, which relate research topics according to the authors who publish in them, from corpora of articles in which fields of study can be identified. FoS networks are particularly useful for the distant reading of large datasets of research papers when analysed through the lens of exploring multidisciplinary science. In an evolving scientific landscape, modular communities in FoS networks offer an alternative categorisation strategy for research topics and sub-disciplines, when compared to traditional prescribed discipline classification schemes. Furthermore, structural role analysis of FoS networks can highlight important characteristics of topics in such communities. To support this, we present two case studies which explore multidisciplinary research in corpora of varying size and scope; namely, 6323 articles relating to network science research and 4,184,011 articles relating to research on the COVID-19-pandemic.

The fading impact of lockdowns: A data analysis of the effectiveness of Covid-19 travel restrictions during different pandemic phases.

As countries struggled with SARS-COV2 outbreaks at the beginning of 2021, many citizens found themselves in yet another period of increasing travel restrictions, if not a strict lockdown. At the same time there was concern that further restrictions would prove to be less effective due to a range of reasons including increasing pandemic fatigue or the lack of appropriate supports. In this study we investigate whether restrictions remained effective as a way to limit non-essential travel in order to curb virus transmission. We do this by analysing adherence during periods of increasing and decreasing restrictions in 125 countries during three different 4-month phases, early (March-June 2020), middle (July-October 2020), and late (November 2020-February 2021) over the course of the first year of the pandemic, and prior to significant population-wide vaccination. We use the strength of the relationship between restriction levels and the level of personal mobility associated with non-essential travel in order to determine the degree of adherence to the restrictions imposed. We show that there is evidence of a significant decrease in adherence to restrictions during the middle and late phases of the pandemic, compared with the early phase. Our analysis further suggests that this decrease in adherence is due to changes in mobility rather than changes in restrictions. We conclude, therefore, that restrictions have become less effective at curbing non-essential travel, which may alter the cost-benefit analysis of restrictions and lockdowns, thus highlighting the need for governments to reconsider large-scale restrictions as a containment strategy in the future, in favour of more focused or flexible mitigation approaches.

Decoupling of Internal and External Workload During a Marathon: An Analysis of Durability in 82,303 Recreational Runners.

AIM: This study characterised the decoupling of internal-to-external workload in marathon running and investigated whether decoupling magnitude and onset could improve predictions of marathon performance. METHODS: The decoupling of internal-to-external workload was calculated in 82,303 marathon runners (13,125 female). Internal workload was determined as a percentage of maximum heart rate, and external workload as speed relative to estimated critical speed (CS). Decoupling magnitude (i.e., decoupling in the 35-40 km segment relative to the 5-10 km segment) was classified as low (< 1.1), moderate (≥ 1.1 but < 1.2) or high (≥ 1.2). Decoupling onset was calculated when decoupling exceeded 1.025. RESULTS: The overall internal-to-external workload decoupling experienced was 1.16 ± 0.22, first detected 25.2 ± 9.9 km into marathon running. The low decoupling group (34.5% of runners) completed the marathon at a faster relative speed (88 ± 6% CS), had better marathon performance (217.3 ± 33.1 min), and first experienced decoupling later in the marathon (33.4 ± 9.0 km) compared to those in the moderate (32.7% of runners, 86 ± 6% CS, 224.9 ± 31.7 min, and 22.6 ± 7.7 km), and high decoupling groups (32.8% runners, 82 ± 7% CS, 238.5 ± 30.7 min, and 19.1 ± 6.8 km; all p < 0.01). Compared to females, males' decoupling magnitude was greater (1.17 ± 0.22 vs. 1.12 ± 0.16; p < 0.01) and occurred earlier (25.0 ± 9.8 vs. 26.3 ± 10.6 km; p < 0.01). Marathon performance was associated with the magnitude and onset of decoupling, and when included in marathon performance models utilising CS and the curvature constant, prediction error was reduced from 6.45 to 5.16%. CONCLUSION: Durability characteristics, assessed as internal-to-external workload ratio, show considerable inter-individual variability, and both its magnitude and onset are associated with marathon performance.

Estimating the cost of training disruptions on marathon performance.

Completing a marathon usually requires at least 12-16 weeks of consistent training, but busy lifestyles, illness or injury, and motivational issues can all conspire to disrupt training. This study aims to investigate the frequency and performance cost of training disruptions, especially among recreational runners. Using more than 15 million activities, from 300,000 recreational runners who completed marathons during 2014-2017, we identified periods of varying durations up to 16 weeks before the marathon where runners experienced a complete cessation of training (so-called training disruptions). We identified runners who had completed multiple marathons including: (i) at least one disrupted marathon with a long training disruption of ≥ 7 days; and (ii) at least one undisrupted marathon with no training disruptions. Next, we calculated the performance cost of long training disruptions as the percentage difference between these disrupted and undisrupted marathon times, comparing the frequency and cost of training disruptions according to the sex, age, and ability of runner, and whether the disruptions occurred early or late in training. Over 50% of runners experienced short training disruptions up to and including 6 days, but longer disruptions were found to be increasingly less frequent among those who made it to race-day. Runners who experience longer training disruptions ( ≥ 7 days) suffer a finish-time cost of 5-8% compared to when the same runners experienced only short training disruptions (<7 days). While we found little difference (<5%) in the likelihood of disruptions-when comparing runners based on sex, age, ability, and the timing of a disruption-we did find significant differences in the the cost of disruptions (10-15%) among these groups. Two sample t -tests indicate that long training disruptions lead to a greater finish-time cost for males (5%) than females (3.5%). Faster runners also experience a greater finish-time cost (5.4%) than slower runners (2.6%). And, when disruptions occur late in training (close to race-day), they are associated with a greater finish-time cost (5.2%) than similar disruptions occurring earlier in training (4.4%). By parameterising and quantifying the cost of training disruptions, this work can help runners and coaches to better understand the relationship between training consistency and marathon performance. This has the potential to help them to better evaluate disruption risk during training and to plan for race-day more appropriately when disruptions do occur.

Correction: Collaboration in the time of COVID: a scientometric analysis of multidisciplinary SARS-CoV-2 research.

[This corrects the article DOI: 10.1057/s41599-021-00922-7.].

Longer Disciplined Tapers Improve Marathon Performance for Recreational Runners.

For marathoners the taper refers to a period of reduced training load in the weeks before race-day. It helps runners to recover from the stresses of weeks of high-volume, high-intensity training to enhance race-day performance. The aim of this study was to analyse the taper strategies of recreational runners to determine whether particular forms of taper were more or less favorable to race-day performance. Methods: We analyzed the training activities of more than 158,000 recreational marathon runners to define tapers based on a decrease in training volume (weekly distance). We identified different types of taper based on a combination of duration (1-4 weeks of decreasing training) and discipline (strict tapers progressively decrease training in the weeks before the marathon, relaxed tapers do not) and we grouped runners based on their taper type to determine the popularity of different types of taper and their associated performance characteristics. Results: Kruskal-Wallis tests (H(7)≥ 521.11, p < 0.001), followed by posthoc Dunns tests with a Bonferroni correction, confirmed that strict tapers were associated with better marathon performance than relaxed tapers (p < 0.001) and that longer tapers of up to 3 weeks were associated with better performance than shorter tapers (p < 0.001). Results indicated that strict 3-week tapers were associated with superior marathon finish-time benefits (a median finish-time saving of 5 min 32.4 s or 2.6%) compared with a minimal taper (p < 0.001). We further found that female runners were associated with greater finish-time benefits than men, for a given taper type ( ≤ 3-weeks in duration), based on Mann Whitney U tests of significance with p < 0.001. Conclusion: The findings of this study for recreational runners are consistent with related studies on highly-trained athletes, where disciplined tapers were associated with comparable performance benefits. The findings also highlight how most recreational runners (64%) adopt less disciplined (2-week and 3-week) tapers and suggest that shifting to a more disciplined taper strategy could improve performance relative to the benefits of a less disciplined taper.

Estimating Exposure Risk to Guide Behaviour During the SARS-COV2 Pandemic.

The end of 2020 and the beginning of 2021 was a challenging time for many countries in Europe, as the combination of colder weather, holiday celebrations, and the emergence of more transmissible virus variants conspired to create a perfect storm for virus transmission across the continent. At the same time lockdowns appeared to be less effective than they were earlier in the pandemic. In this paper we argue that one contributing factor is that existing ways of communicating risk-case numbers, test positivity rates, hospitalisations etc.-are difficult for individuals to translate into a level of personal risk, thereby limiting the ability of individuals to properly calibrate their own behaviour. We propose an new more direct measure of personal risk, exposure risk, to estimate the likelihood that an individual will come into contact with an infected person, and we argue that it can play an important role, alongside more conventional statistics, to help translate complex epidemiological data into a simple measure to guide pandemic behaviour. We describe how exposure risk can be calculated using existing data and infection prediction models, and use it to evaluate and compare the exposure risk associated with 39 European countries.

How recreational marathon runners hit the wall: A large-scale data analysis of late-race pacing collapse in the marathon.

INTRODUCTION: In the marathon, how runners pace and fuel their race can have a major impact on race outcome. The phenomenon known as hitting the wall (HTW) refers to the iconic hazard of the marathon distance, in which runners experience a significant slowing of pace late in the race, typically after the 20-mile mark, and usually because of a depletion of the body's energy stores. AIM: This work investigates the occurrence of significant late-race slowing among recreational marathoners, as a proxy for runners hitting the wall, to better understand the likelihood and nature of such slowdowns, and their effect on race performance. METHODS: Using pacing data from more than 4 million race records, we develop a pacing-based definition of hitting the wall, by identifying runners who experience a sustained period of slowing during the latter stages of the marathon. We calculate the cost of these slowdowns relative to estimates of the recent personal-best times of runners and compare slowdowns according to runner sex, age, and ability. RESULTS: We find male runners more likely to slow significantly (hit the wall) than female runners; 28% of male runners hit the wall compared with 17% of female runners, χ2(1, N = 1, 928, 813) = 27, 693.35, p < 0.01, OR = 1.43. Such slowdowns are more frequent in the 3 years immediately before and after a recent personal-best (PB) time; for example, 36% of all runners hit the wall in the 3 years before a recent PB compared with just 23% in earlier years, χ2(1, N = 509, 444) = 8, 120.74, p < 0.01, OR = 1.31. When runners hit the wall, males slow more than females: a relative slowdown of 0.40 vs. 0.37 is noted, for male and female runners, when comparing their pace when they hit the wall to their earlier race (5km-20km) pace, with t(475, 199) = 60.19, p < 0.01, d = 0.15. And male runners slow over longer distances than female runners: 10.7km vs. 9.6km, respectively, t(475, 199) = 68.44, p < 0.01, d = 0.17. Although, notably the effect size of these differences is small. We also find the finish-time costs of hitting the wall (lost minutes) to increase with ability; r2(7) = 0.91, p < 0.01 r2(7) = 0.81, p < 0.01 for male and female runners, respectively. CONCLUSIONS: While the findings from this study are consistent with qualitative results from earlier single-race or smaller-scale studies, the new insights into the risk and nature of slowdowns, based on the runner sex, age, and ability, have the potential to help runners and coaches to better understand and calibrate the risk/reward trade-offs that exist as they plan for future races.

Calculation of Critical Speed from Raw Training Data in Recreational Marathon Runners.

INTRODUCTION: Critical speed (CS) represents the highest intensity at which a physiological steady state may be reached. The aim of this study was to evaluate whether estimations of CS obtained from raw training data can predict performance and pacing in marathons. METHODS: We investigated running activities logged into an online fitness platform by >25,000 recreational athletes before big-city marathons. Each activity contained time, distance, and elevation every 100 m. We computed grade-adjusted pacing and the fastest pace recorded for a set of target distances (400, 800, 1000, 1500, 3000, and 5000 m). CS was determined as the slope of the distance-time relationship using all combinations of, at least, three target distances. RESULTS: The relationship between distance and time was linear, irrespective of the target distances used (pooled mean ± SD: R = 0.9999 ± 0.0001). The estimated values of CS from all models were not different (3.74 ± 0.08 m·s), and all models correlated with marathon performance (R = 0.672 ± 0.036, error = 8.01% ± 0.51%). CS from the model including 400, 800, and 5000 m best predicted performance (R = 0.695, error = 7.67%) and was used in further analysis. Runners completed the marathon at 84.8% ± 13.6% CS, with faster runners competing at speeds closer to CS (93.0% CS for 150 min marathon times vs 78.9% CS for 360 min marathon times). Runners who completed the first half of the marathon at >94% of their CS, and particularly faster than CS, were more likely to slowdown by more than 25% in the second half of race. CONCLUSION: This study suggests that estimations of CS from raw training data can successfully predict marathon performance and provide useful pacing information.

Devising a Pace-Based Definition for "The Wall": An Observational Analysis of Marathoners' Subjective Experiences of Fatigue.

CONTEXT: Many runners report "hitting The Wall" (HTW) during a marathon (42.2 km). However, the performance manifestation of this subjectively experienced phenomenon remains unclear. OBJECTIVE: To identify a pace-based classification for HTW by integrating subjective reports of fatigue and runners' pacing profiles during a marathon. DESIGN: Cross-sectional study. SETTING: Public race event (2018 Dublin Marathon). PATIENTS OR OTHER PARTICIPANTS: Eighty-three runners (28 [34%] women, 55 [66%] men, age = 41.5 ± 9.1 years, height = 1.73 ± 0.09 m, mass = 70.2 ± 10.1 kg). MAIN OUTCOME MEASURE(S): The pacing profiles for respondents to our postrace questionnaire that concerned the phenomenon of HTW were evaluated. Receiver operating characteristic analyses were performed on discretized outcomes of the time series of marathoners' paces during the race. RESULTS: Using the receiver operating characteristic analyses, we observed that runners could be classified as having experienced HTW if they ran any 1-km segment 11% slower than the average of the remaining segments of the race (accuracy = 84.6%, sensitivity = 1, specificity = 0.6) or if the standard deviation of the normalized 1-km split times exceeded 0.0532 (accuracy = 83%, sensitivity = 0.818, specificity = 0.8). Similarly, runners could be classified as having experienced HTW if they ran any 5-km segment 7.3% slower than the average of the remaining 5-km segments of the race (accuracy = 84.6%, sensitivity = 1, specificity = 0.644) or if the standard deviation of the normalized 5-km split times exceeded 0.0346 (accuracy = 82%, sensitivity = 0.909, specificity = 0.622). CONCLUSIONS: These pace-based criteria could be valuable to researchers evaluating HTW prevalence in cohorts for whom they lack subjective questionnaire data.

An evaluation of the training determinants of marathon performance: A meta-analysis with meta-regression.

OBJECTIVES: Marathoners rely on expert-opinion and the anecdotal advice of their peers when devising their training plans for an upcoming race. The accumulation of results from multiple scientific studies has the potential to clarify the precise training requirements for the marathon. The purpose of the present study was to perform a systematic review, meta-analysis and meta-regression of available literature to determine if a dose-response relationship exists between a series of training behaviours and marathon performance. DESIGN: Systematic review, meta-analysis and meta-regression. METHODS: A systematic search of multiple literature sources was undertaken to identify observational and interventional studies of elite and recreational marathon (42.2km) runners. RESULTS: Eighty-five studies which included 137 cohorts of runners (25% female) were included in the meta-regression, with average weekly running distance, number of weekly runs, maximum running distance completed in a single week, number of runs ≥32km completed in the pre-marathon training block, average running pace during training, distance of the longest run and hours of running per week used as covariates. Separately conducted univariate random effects meta-regression models identified a negative statistical association between each of the above listed training behaviours and marathon performance (R2 0.38-0.81, p<0.001), whereby increases in a given training parameter coincided with faster marathon finish times. Meta-analysis revealed the rate of non-finishers in the marathon was 7.27% (95% CI 6.09%-8.65%). CONCLUSIONS: These data can be used by athletes and coaches to inform the development of marathon training regimes that are specific to a given target finish time.

Prediction Equations for Marathon Performance: A Systematic Review.

PURPOSE: Despite the volume of available literature focusing on marathon running and the prediction of performance, no single prediction equations exists that is accurate for all runners of varying experiences and abilities. Indeed the relative merits and utility of the existing equations remain unclear. Thus, the aim of this study was to collate, characterize, compare, and contrast all available marathon prediction equations. METHODS: A systematic review was conducted to identify observational research studies outlining any kind of prediction algorithm for marathon performance. RESULTS: Thirty-six studies with 114 equations were identified. Sixty-one equations were based on training and anthropometric variables, whereas 53 equations included variables that required laboratory tests and equipment. The accuracy of these equations was denoted via a variety of metrics; r2 values were provided for 68 equations (r2 = .10-.99), and an SEE was provided for 19 equations (SEE 0.27-27.4 min). CONCLUSION: Heterogeneity of the data precludes the identification of a single "best" equation. Important variables such as course gradient, sex, and expected weather conditions were often not included, and some widely used equations did not report the r2 value. Runners should therefore be wary of relying on a single equation to predict their performance.

Kinetic changes during a six-week minimal footwear and gait-retraining intervention in runners.

An evaluation of a six-week Combined minimal footwear transition and gait-retraining combination vs. gait retraining only on impact characteristics and leg stiffness. Twenty-four trained male runners were randomly assigned to either (1) Minimalist footwear transition Combined with gait-retraining over a six-week period ("Combined" group; n = 12) examined in both footwear, or (2) a gait-retraining group only with no minimalist footwear exposure ("Control"; n = 12). Participants were assessed for loading rate, impact peak, vertical, knee and ankle stiffness, and foot-strike using 3D and kinetic analysis. Loading rate was significantly higher in the Combined group in minimal shoes in pre-tests compared to a Control (P ≤ 0.001), reduced significantly in the Combined group over time (P ≤ 0.001), and was not different to the Control group in post-tests (P = 0.16). The impact peak (P = 0.056) and ankle stiffness reduced in both groups (P = 0.006). Loading rate and vertical stiffness was higher in minimalist footwear than conventional running shoes both pre (P ≤ 0.001) and post (P = 0.046) the intervention. There has a higher tendency to non-rearfoot strike in both interventions, but more acute changes in the minimalist footwear. A Combined intervention can potentially reduce impact variables. However, higher loading rate initially in minimalist footwear may increase the risk of injury in this condition.

Physiological Demands of Flat Horse Racing Jockeys.

The physiological demands of jockeys during competition remain largely unknown, thereby creating challenges when attempting to prescribe sport-specific nutrition and training guidelines. The purpose of this study was to evaluate the physiological demands and energy requirements of jockeys during flat racing. Oxygen uptake (V[Combining Dot Above]O2) and heart rate (HR) were assessed in 18 male trainee jockeys during a race simulation trial on a mechanical horse racing simulator for the typical time duration to cover a common flat race distance of 1,400 m. In addition, 8 male apprentice jockeys participated in a competitive race, over distances ranging from 1,200 to 1,600 m, during which HR and respiratory rate (RR) were assessed. All participants performed a maximal incremental cycle ergometer test. During the simulated race, peak V[Combining Dot Above]O2 was 42.74 ± 5.6 ml·kg·min (75 ± 11% of V[Combining Dot Above]O2peak) and below the mean ventilatory threshold (81 ± 5% of V[Combining Dot Above]O2peak) reported in the maximal incremental cycle test. Peak HR was 161 ± 16 b·min (86 ± 7% of HRpeak). Energy expenditure was estimated as 92.5 ± 18.8 kJ with an associated value of 9.4 metabolic equivalents. During the competitive race trial, peak HR reached 189 ± 5 b·min (103 ± 4% of HRpeak) and peak RR was 50 ± 7 breaths per minute. Results suggest that horse racing is a physically demanding sport, requiring jockeys to perform close to their physiological limit to be successful. These findings may provide a useful insight when developing sport-specific nutrition and training strategies to optimally equip and prepare jockeys physically for the physiological demands of horse racing.