ECSS Paris 2023: OP-AP04
INTRODUCTION: This study aims to investigate the pacing strategy and the importance of the four different stroke in men’s and women’s 200- and 400-m individual medley competitions in Olympic Games and World Swimming Championships between 2000 and 2021, excluding 2008 to 2010. METHODS: The time in each lap and overall race were retrieved from the World Aquatics website. The final data comprised a total of 1937 data points (1052 for men, 885 for women) for the 200-m event and 1192 data points (607 for men, 585 for women) for the 400-m event. The standardized time for each stroke was calculated by dividing the actual time by a reference time specific to each stroke to accommodate the inherent disparities among the four strokes and the impact of the diving start in butterfly. The reference time was derived from the respective laps in single-stroke finals in the in the 2017 World Swimming Championships. A decision tree method was applied. The binary dependent variables were qualified or non-qualified in heats and semifinals, and medalists or non-medalists in finals. The independent variables were the pace in each stroke, represented by the ratio of standardized time in the specific stroke to the sum of standardized time in all four strokes. A total of 10 decision trees with the Classification and Regression Tree algorithm were established: heats, semifinals, and finals in men’s and women’s 200-m medley; and heats and finals in men’s and women’s 400-m individual medley. The decision tree models was established. The binary dependent variables were qualified or non-qualified in heats and semifinals, and winning medals in finals. The independent variables were the ratio of standardized time in each stroke to the sum of standardized time in all four strokes. The normalized importance of each stroke in each decision tree was calculated. RESULTS: In men’s and women’s 200-m and 400-m individual medley, butterfly held the highest normalized importance in winning medals in the finals. The pace in butterfly was the first node in eight of the 10 decision trees, except men’s 200-m semifinal and 400-m final, in which backstroke was the first node. The cut-off values for pace in butterfly in these eight models indicated that a pace larger than 0.236–0.245, i.e. spending relatively longer standardized time, in butterfly was associated with a higher likelihood of being qualified or winning medals in these competitions. It is noteworthy that the pace in this study is relative to each swimmer’s performance in all four strokes. CONCLUSION: Elite swimmers who spend a higher ratio of standardized time in butterfly is associated with a higher likelihood of winning medals or qualifying for the next stage in most international men’s and women’s 200-m and 400-m individual medley. Excellence in butterfly is the most crucial determinant in success in individual medley events while proficiency in at least one other stroke enhances the likelihood of winning.
Read CV CHINKUEI YANGECSS Paris 2023: OP-AP04
INTRODUCTION: Advances in technology have led to a loads of data in many sports, making data-driven models increasingly popular for performance analysis (Silva. et.al., 2007). Different models have been applied to swimming, predicting individual event performance based on various datasets (Wu et.al.,2021). In long-distance swimming, the pace strategy (PS) follows a U-shaped curve (Lara and Del Coso, 2021). This means professional swimmers start and finish fast, maintaining a relatively consistent speed with minor fluctuations in between. This study investigates the use of a Random Forest model to predict final race times based on athletes heat times and pace strategy. METHODS: Race data from five Olympic Games (Sydney, Athens, Beijing, Rio, and Tokyo) were analyzed, containing both heats and finals data. Data were obtained from the FINA website (https://www.fina.org), providing split times for every 50m and final times for each race. A total of 174 races were analyzed. The dataset was divided into two parts: one for training the model (heat data) and the other for evaluation (final data).Relevant features like mean time, speed at different distances, and total time were selected for model training. A Random Forest model was trained with optimal hyperparameters: 200 estimators, max_depth =8, max_features=sqrt, min_samples_leaf=1, random_state=42. To evaluate the models performance, the following metrics were used: Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Error (MAE), R-squared and Explained Variance Score. RESULTS: On the training data, the model achieved a Mean Squared Error (MSE) of 5.09, Root Mean Squared Error (RMSE) of 2.25, Mean Absolute Error (MAE) of 1.708, R-squared of 0.953, and Explained Variance Score of 0.968. On the final dataset, the model achieved an MSE of 32.53, RMSE of 5.70, MAE of 4.40, R-squared of 0.94, and Explained Variance Score of 0.94. This indicates a slight decrease in model performance on the new data, with an average prediction error of around 5.7 seconds. In a race lasting approximately 14 minutes, this translates to an error of less than 0.58% of the total time. CONCLUSION: The analysis consistently revealed a U-shaped pace strategy profile employed by all athletes across all races, regardless of whether they were competing in heats or finals. This finding highlights the consistency of this approach in 1500m swimming. The chosen Random Forest model demonstrated worthy performance on the training data, explaining over 95% of the variance in total final times. This indicates the models ability to effectively learn the underlying patterns and relationships within the dataset. Still, when applied to the evaluation data (final times), the models performance exhibited a slight decrease, resulting in an average prediction error of approximately 5.7 seconds. This study showcases the promising potential of Random Forest regression for predicting swimming times. Utilizing features derived from both heat and final performance.
Read CV Tiago RussomannoECSS Paris 2023: OP-AP04
INTRODUCTION: Drafting in short-distance triathlon cycling introduces variable power demands [1,2]. Performance evaluations solely based on cycling split times may overlook the differing power requirements experienced by riders even within the same bike group. Factors like positioning within the bike group could influence power demands and subsequently initial fatigue in the following run segment of a triathlon [3]. METHODS: We analyzed power data and television-based positional information from the 2020 male sprint triathlon World Championship event held in Hamburg. Specifically, we focused on five of the eight riders from the leading bike group, examining their power profiles and power distributions. Employing hierarchical Bayesian models, we analyzed the association between positioning within the group and the power demands during accelerations following turns. Our study was preregistered and has open data and code available. RESULTS: Within the same bike group, athletes showed distinct power profiles and employed different positioning strategies. Notably, as athletes positioned themselves further back during a turn, they demonstrated higher peak power (+24.2 W [4.8; 36.7] per position; mean [95% credibility interval]) and 10 seconds mean power (+19.3 W [10.5; 27.1]) during subsequent accelerations. However, the effect of positioning was less pronounced on the 20 seconds mean power (+6.3 W [-1.4; 13.6]), and it had a negative impact on the 20 seconds mean power before the turn (-13.4 W [-20.8; -4.99]). CONCLUSION: The positioning of athletes during cycling in a triathlon can impact the power demands, potentially influencing the performance during the subsequent running leg. Our findings indicate that to reduce power demands, athletes should position themselves at the front of the group during turns and towards the back for the remainder of the cycling segment. However, employing this strategy will likely compromise cooperative group work. Therefore, athletes and coaches must develop positioning strategies based on data and experience, tailored to individual abilities. Future studies should aim to correlate power variability with running performance in races, while accounting for individual running performance levels. [1] Bernard et al. (2009). Med Sci Sports Exerc [2] Etxebarria et al. (2014). Int J Sports Physiol Perform [3] Walsh (2019). Sports (Basel)
Read CV Simon NolteECSS Paris 2023: OP-AP04