Issues

Journal & Issues

Volume 21 (2022): Issue 1 (March 2022)

Volume 20 (2021): Issue 2 (December 2021)

Volume 20 (2021): Issue 1 (July 2021)

Volume 19 (2020): Issue 2 (December 2020)

Volume 19 (2020): Issue 1 (July 2020)

Volume 18 (2019): Issue 3 (December 2019)

Volume 18 (2019): Issue 2 (September 2019)
Special Issue: Selected papers presented at the 12th Symposium of the Section Computer Science in Sport of the German Association of Sport Science (September 4.-7., 2018)

Volume 18 (2019): Issue 1 (July 2019)

Volume 17 (2018): Issue 2 (December 2018)

Volume 17 (2018): Issue 1 (July 2018)

Volume 16 (2017): Issue 3 (December 2017)

Volume 16 (2017): Issue 2 (November 2017)

Volume 16 (2017): Issue 1 (July 2017)

Volume 15 (2016): Issue 2 (December 2016)

Volume 15 (2016): Issue 1 (July 2016)

Journal Details
Format
Journal
eISSN
1684-4769
First Published
16 Apr 2016
Publication timeframe
2 times per year
Languages
English

Search

Volume 16 (2017): Issue 2 (November 2017)

Journal Details
Format
Journal
eISSN
1684-4769
First Published
16 Apr 2016
Publication timeframe
2 times per year
Languages
English

Search

6 Articles

Special Issue on Modeling in Endurance Sports Movement and Sport Science, Editors: Chris R. Abbiss, Dietmar Saupe

Open Access

Editorial: Special Issue on Modeling in Endurance Sports

Published Online: 30 Nov 2017
Page range: 76 - 77

Abstract

Open Access

Linear and Nonlinear Prediction Models Show Comparable Precision for Maximal Mean Speed in a 4x1000 m Field Test

Published Online: 30 Nov 2017
Page range: 78 - 87

Abstract

Abstract

Maximal oxygen uptake (VO2max) is one of the most distinguished parameters in endurance sports and plays an important role, for instance, in predicting endurance performance. Different models have been used to estimate VO2max or performance based on VO2max. These models can use linear or nonlinear approaches for modeling endurance performance. The aim of this study was to estimate VO2max in healthy adults based on the Queens College Step Test (QCST) as well as the Shuttle Run Test (SRT) and to use these values for linear and nonlinear models in order to predict the performance in a maximal 1000 m run (i.e. the speed in an incremental 4×1000 m Field Test (FT)). 53 female subjects participated in these three tests (QCST, SRT, FT). Maximal oxygen uptake values from QCST and SRT were used as (a) predictor variables in a multiple linear regression (MLR) model and as (b) input variables in a multilayer perceptron (MLP) after scaling in preprocessing. Model output was speed [km·h−1] in a maximal 1000 m run. Maximal oxygen uptake values estimated from QCST (40.8 ± 3.5 ml·kg−1·min−1) and SRT (46.7 ± 4.5 ml·kg−1·min−1) were significantly correlated (r = 0.38, p < 0.01) and maximal mean speed in the FT was 12.8 ± 1.6 km·h−1. Root mean squared error (RMSE) of the cross validated MLR model was 0.89 km·h−1 while it was 0.95 km·h−1 for MLP. Results showed that the accuracy of the applied MLP was comparable to the MLR, but did not outperform the linear approach.

Keywords

  • multiple linear regression
  • multilayer perceptron
  • performance prediction
  • artificial neural network
Open Access

How to Stay Ahead of the Pack: Optimal Road Cycling Strategies for two Cooperating Riders

Published Online: 30 Nov 2017
Page range: 88 - 100

Abstract

Abstract

Within road-cycling, the optimization of performance using mathematical models has primarily been performed in the individual time trial. Nevertheless, most races are 'mass-start' events in which many riders compete at the same time. In some special situations, e.g. breakaways from the peloton, the riders are forced to team up. To simulate those cooperative rides of two athletes, an extension of models and optimization approaches for individual time trials is presented. A slipstream model based on experimental data is provided to simulate the physical interaction between the two riders. In order to simulate real world behavior, a penalty for the difference in the exertion levels of the two riders is introduced. This means, that even though both riders aim to be as fast as possible as a group, neither of them should have an advantage over the other because of significantly different levels of fatigue during the ride. In our simulations, the advantage of cooperation of two equally trained athletes adds up to a time gain of about 10% compared to an individual ride.

Keywords

  • road cycling
  • slipstream
  • pacing strategy
  • optimal control
Open Access

Predicting Elite Triathlon Performance: A Comparison of Multiple Regressions and Artificial Neural Networks

Published Online: 30 Nov 2017
Page range: 101 - 116

Abstract

Abstract

Two different computational approaches were used to predict Olympic distance triathlon race time of German male elite triathletes. Anthropometric measurements and two treadmill running tests to collect physiological variables were repeatedly conducted on eleven male elite triathletes between 2008 and 2012. After race time normalization, exploratory factor analysis (EFA), as a mathematical preselection method, followed by multiple linear regression (MLR) and dominance paired comparison (DPC), as a preselection method considering professional expertise, followed by nonlinear artificial neural network (ANN) were conducted to predict overall race time. Both computational approaches yielded two prediction models. MLR provided R² = 0.41 in case of anthropometric variables (predictive: pelvis width and shoulder width) and R² = 0.67 in case of physiological variables (predictive: maximum respiratory rate, running pace at 3-mmol·L-1 blood lactate and maximum blood lactate). ANNs using the five most important variables after DPC yielded R² = 0.43 in case of anthropometric variables and R² = 0.86 in case of physiological variables. The advantage of ANNs over MLRs was the possibility to take non-linear relationships into account. Overall, race time of male elite triathletes could be well predicted without interfering with individual training programs and season calendars.

Keywords

  • professional triathletes
  • performance prediction
  • racetime normalization
  • factor analysis
  • dominance paired comparison
Open Access

Performance Estimation using the Fitness-Fatigue Model with Kalman Filter Feedback

Published Online: 30 Nov 2017
Page range: 117 - 129

Abstract

Abstract

Tracking and predicting the performance of athletes is of great interest, not only in training science but also, increasingly, for serious hobbyists. The increasing availability and use of smart watches and fitness trackers means that abundant data is becoming available, and the interest to optimally use this data for performance tracking and training optimization is great. One competitive model in this domain is the 3-time-constant fitness-fatigue model by Busso based on the model by Banister and colleagues. In the following, we will show that this model can be written equivalently as a linear, time-variant state-space model. With this understanding, it becomes clear that all methods for optimum tracking in statespace models are also directly applicable here. As an example, we show how a Kalman filter can be combined with the fitness-fatigue model in a mathematically consistent fashion. This gives us the opportunity to optimally consider measurements of performance to adapt the fitness and fatigue estimates in a datadriven manner. Results show that this approach is capable of clearly improving performance tracking and prediction over a range of different scenarios.

Keywords

  • performance modeling
  • kalman filter
  • fitness
  • fatigue
Open Access

Predicting Short-Term HR Response to Varying Training Loads Using Exponential Equations

Published Online: 30 Nov 2017
Page range: 130 - 148

Abstract

Abstract

Aim of this study was to test whether a monoexponential formula is appropriate to analyze and predict individual responses to the change of load bouts online during training. Therefore, 234 heart rate (HR) data sets obtained from extensive interval protocols of four participants during a twelve-week training intervention on a bike ergometer were analyzed. First, HR for each interval was approximated using a monoexponential formula. HR at onset of exercise (HRstart), HR induced by load (HRsteady) and the slope of HR (c) were analyzed. Furthermore, a calculation routine incrementally predicted HRsteady using measured HR data after onset of exercise. Validity of original and approximated data sets were very high (r² =0.962, SD =0.025; Max =0.991, Min =0.702). HRstart was significantly different between all participants (one exception). HRsteady was similar in all participants. Parameter c was independent of the duration of intervention and intervals regarding one training session but was significantly different in all participants (one exception). Final HR was correctly predicted on average after 58.8 s (SD = 34.77, Max =150 s, Min =30 s) based on a difference criteria of less than 5 bpm. In 3 participants, HRsteady was predicted correctly in 142 out of 175 courses (81.1%).

Keywords

  • adaptation
  • hr responses
  • monoexponential equation
6 Articles

Special Issue on Modeling in Endurance Sports Movement and Sport Science, Editors: Chris R. Abbiss, Dietmar Saupe

Open Access

Editorial: Special Issue on Modeling in Endurance Sports

Published Online: 30 Nov 2017
Page range: 76 - 77

Abstract

Open Access

Linear and Nonlinear Prediction Models Show Comparable Precision for Maximal Mean Speed in a 4x1000 m Field Test

Published Online: 30 Nov 2017
Page range: 78 - 87

Abstract

Abstract

Maximal oxygen uptake (VO2max) is one of the most distinguished parameters in endurance sports and plays an important role, for instance, in predicting endurance performance. Different models have been used to estimate VO2max or performance based on VO2max. These models can use linear or nonlinear approaches for modeling endurance performance. The aim of this study was to estimate VO2max in healthy adults based on the Queens College Step Test (QCST) as well as the Shuttle Run Test (SRT) and to use these values for linear and nonlinear models in order to predict the performance in a maximal 1000 m run (i.e. the speed in an incremental 4×1000 m Field Test (FT)). 53 female subjects participated in these three tests (QCST, SRT, FT). Maximal oxygen uptake values from QCST and SRT were used as (a) predictor variables in a multiple linear regression (MLR) model and as (b) input variables in a multilayer perceptron (MLP) after scaling in preprocessing. Model output was speed [km·h−1] in a maximal 1000 m run. Maximal oxygen uptake values estimated from QCST (40.8 ± 3.5 ml·kg−1·min−1) and SRT (46.7 ± 4.5 ml·kg−1·min−1) were significantly correlated (r = 0.38, p < 0.01) and maximal mean speed in the FT was 12.8 ± 1.6 km·h−1. Root mean squared error (RMSE) of the cross validated MLR model was 0.89 km·h−1 while it was 0.95 km·h−1 for MLP. Results showed that the accuracy of the applied MLP was comparable to the MLR, but did not outperform the linear approach.

Keywords

  • multiple linear regression
  • multilayer perceptron
  • performance prediction
  • artificial neural network
Open Access

How to Stay Ahead of the Pack: Optimal Road Cycling Strategies for two Cooperating Riders

Published Online: 30 Nov 2017
Page range: 88 - 100

Abstract

Abstract

Within road-cycling, the optimization of performance using mathematical models has primarily been performed in the individual time trial. Nevertheless, most races are 'mass-start' events in which many riders compete at the same time. In some special situations, e.g. breakaways from the peloton, the riders are forced to team up. To simulate those cooperative rides of two athletes, an extension of models and optimization approaches for individual time trials is presented. A slipstream model based on experimental data is provided to simulate the physical interaction between the two riders. In order to simulate real world behavior, a penalty for the difference in the exertion levels of the two riders is introduced. This means, that even though both riders aim to be as fast as possible as a group, neither of them should have an advantage over the other because of significantly different levels of fatigue during the ride. In our simulations, the advantage of cooperation of two equally trained athletes adds up to a time gain of about 10% compared to an individual ride.

Keywords

  • road cycling
  • slipstream
  • pacing strategy
  • optimal control
Open Access

Predicting Elite Triathlon Performance: A Comparison of Multiple Regressions and Artificial Neural Networks

Published Online: 30 Nov 2017
Page range: 101 - 116

Abstract

Abstract

Two different computational approaches were used to predict Olympic distance triathlon race time of German male elite triathletes. Anthropometric measurements and two treadmill running tests to collect physiological variables were repeatedly conducted on eleven male elite triathletes between 2008 and 2012. After race time normalization, exploratory factor analysis (EFA), as a mathematical preselection method, followed by multiple linear regression (MLR) and dominance paired comparison (DPC), as a preselection method considering professional expertise, followed by nonlinear artificial neural network (ANN) were conducted to predict overall race time. Both computational approaches yielded two prediction models. MLR provided R² = 0.41 in case of anthropometric variables (predictive: pelvis width and shoulder width) and R² = 0.67 in case of physiological variables (predictive: maximum respiratory rate, running pace at 3-mmol·L-1 blood lactate and maximum blood lactate). ANNs using the five most important variables after DPC yielded R² = 0.43 in case of anthropometric variables and R² = 0.86 in case of physiological variables. The advantage of ANNs over MLRs was the possibility to take non-linear relationships into account. Overall, race time of male elite triathletes could be well predicted without interfering with individual training programs and season calendars.

Keywords

  • professional triathletes
  • performance prediction
  • racetime normalization
  • factor analysis
  • dominance paired comparison
Open Access

Performance Estimation using the Fitness-Fatigue Model with Kalman Filter Feedback

Published Online: 30 Nov 2017
Page range: 117 - 129

Abstract

Abstract

Tracking and predicting the performance of athletes is of great interest, not only in training science but also, increasingly, for serious hobbyists. The increasing availability and use of smart watches and fitness trackers means that abundant data is becoming available, and the interest to optimally use this data for performance tracking and training optimization is great. One competitive model in this domain is the 3-time-constant fitness-fatigue model by Busso based on the model by Banister and colleagues. In the following, we will show that this model can be written equivalently as a linear, time-variant state-space model. With this understanding, it becomes clear that all methods for optimum tracking in statespace models are also directly applicable here. As an example, we show how a Kalman filter can be combined with the fitness-fatigue model in a mathematically consistent fashion. This gives us the opportunity to optimally consider measurements of performance to adapt the fitness and fatigue estimates in a datadriven manner. Results show that this approach is capable of clearly improving performance tracking and prediction over a range of different scenarios.

Keywords

  • performance modeling
  • kalman filter
  • fitness
  • fatigue
Open Access

Predicting Short-Term HR Response to Varying Training Loads Using Exponential Equations

Published Online: 30 Nov 2017
Page range: 130 - 148

Abstract

Abstract

Aim of this study was to test whether a monoexponential formula is appropriate to analyze and predict individual responses to the change of load bouts online during training. Therefore, 234 heart rate (HR) data sets obtained from extensive interval protocols of four participants during a twelve-week training intervention on a bike ergometer were analyzed. First, HR for each interval was approximated using a monoexponential formula. HR at onset of exercise (HRstart), HR induced by load (HRsteady) and the slope of HR (c) were analyzed. Furthermore, a calculation routine incrementally predicted HRsteady using measured HR data after onset of exercise. Validity of original and approximated data sets were very high (r² =0.962, SD =0.025; Max =0.991, Min =0.702). HRstart was significantly different between all participants (one exception). HRsteady was similar in all participants. Parameter c was independent of the duration of intervention and intervals regarding one training session but was significantly different in all participants (one exception). Final HR was correctly predicted on average after 58.8 s (SD = 34.77, Max =150 s, Min =30 s) based on a difference criteria of less than 5 bpm. In 3 participants, HRsteady was predicted correctly in 142 out of 175 courses (81.1%).

Keywords

  • adaptation
  • hr responses
  • monoexponential equation

Plan your remote conference with Sciendo