Relationship between starting and finishing position in short track speed skating races

Enhancing Short Track Speed Skating Performance through Improved DDQN Tactical Decision Model

This paper studies the tactical decision-making model of short track speed skating based on deep reinforcement learning, so as to improve the competitive performance of corresponding short track speed skaters. Short track speed skating, a traditional discipline in the Winter Olympics since its establishment in 1988, has consistently garnered attention. As artificial intelligence continues to advance, the utilization of deep learning methods to enhance athletes' tactical decision-making capabilities has become increasingly prevalent. Traditional tactical decision techniques often rely on the experience and knowledge of coaches and video analysis methods that require a lot of time and effort. Consequently, this study proposes a scientific simulation environment for short track speed skating, that accurately simulates the physical attributes of the venue, the physiological fitness of the athletes, and the rules of the competition. The Double Deep Q-Network (DDQN) model is enhanced and utilized, with improvements to the reward function and the distinct description of four tactics. This enables agents to learn optimal tactical decisions in various competitive states with a simulation environment. Experimental results demonstrate that this approach effectively enhances the competition performance and physiological fitness allocation of short track speed skaters.

Tactical positioning behaviours in short-track speed skating: A static and dynamic sequence analysis

Tactical positioning is essential for success in short-track speed skating as the race format (direct, head-to-head competition over multiple laps) prioritises finishing position over finishing time. Despite this, current research into tactical positioning treats the race's laps as discrete, independent events. Accordingly, the aggregate metrics used to summarise each lap's tactical positioning behaviour do not allow us to explore the sequential nature of the data, e.g., Lap 2 occurs after Lap 1 and before Lap 3. Here, we capture the sequential relationships between laps to investigate tactical positioning behaviours in short-track speed skating. Using intermediate and final rankings from 500 m, 1,000 m, and 1,500 m elite short-track races, we analyse whole-race and sub-race race sequences of group and winner tactical positioning behaviours. This approach, combined with a large dataset of races collected over eight seasons of competition (n  = 4,135), provides the most rigorous and comprehensive description of tactical positioning behaviours in short-track speed skating to date. Our results quantify the time-evolving complexity of tactical positioning, offer new thoughts on race strategy, and can help practitioners design more representative learning tasks to enhance skill transfer.

Tactical positioning in short-track speed skating: The utility of race-specific athlete-opponent interactions

In short-track speed skating, tactical positioning is essential for success as the race format (head-to-head) prioritises finishing position over finishing time. At present, our understanding of this phenomenon is based on measuring the similarity between athletes' intermediate and final rankings. However, as this approach groups athlete performances across races, each lap's estimate of tactical importance ignores the athlete-opponent interactions specific to each race. Here, we examine the utility of race-specific athlete-opponent interactions for investigating tactical positioning. Using intermediate and final rankings of elite 1,000 m short-track speed skating competitors collected from 2010/11 to 2017/18 (n = 6,196, races = 1,549), we compared the current method to a novel approach that accounted for race-specific athlete-opponent interactions. This approach first applied the current method to each race independently before using these values to form (1) discrete, empirical distributions of each lap's tactical importance and (2) race-specific tactical positioning sequences. Our results showed that accounting for race-specific athlete-opponent interactions provided a higher measurement granularity (i.e., level of detail) for investigating tactical positioning in short-track speed skating, which better captured the complexity of the phenomenon. We observed 61 different tactical positioning behaviours and 1,269 unique tactical positioning sequences compared to the current approach's nine-point estimates of tactical positioning importance. For this reason, we recommend that researchers and practitioners account for race-specific athlete-opponent interactions in the future as it offers a deeper understanding of tactical positioning that will enhance both strategic and tactical decisions.

Champion Position Analysis in Short Track Speed Skating Competitions From 2007 to 2019

Purpose: Short track speed skating is a racing sport with racing tactics are equally crucial to speed and technical skating skills. Therefore, to investigate the relationship between starting and finishing positions for elite skaters and subsequently, explore pacing patterns for champions are necessary.

Methods: To investigate a pattern of effective tactical positioning strategy, Kendall’s tau-b correlations between starting and finishing position were calculated, with 500 m races having the most positive correlation (0.347, P < 0.05).

Results: Furthermore, starting position distributions of winners in each round, as well as the fluctuations in champion starting positions across rounds were analyzed. Our findings showed that skaters on the first track were inclined to win the rounds in 500, 1,000, and 1,500 m (28, 28, and 22%, respectively), and the differences between starting and finishing positions for champions were minimized in semi-finals. Meanwhile, the pacing patterns were gaining more fluctuations by the increase of race distances for champions, as the average standard deviation of lap rankings equaled 0.90, 1.15, and 2.21 for 500, 1,000, and 1,500 m races, respectively.

Conclusion: In conclusion, elite skaters should adopt flexible tactics at the lowest cost of energy consumption. The overall variability of lap ranking in long-distance races were distinctly higher than it in short distance.

Winning by hiding behind others: An analysis of speed skating data

In some athletic races, such as cycling and types of speed skating races, athletes have to complete a relatively long distance at a high speed in the presence of direct opponents. To win such a race, athletes are motivated to hide behind others to suppress energy consumption before a final moment of the race. This situation seems to produce a social dilemma: players want to hide behind others, whereas if a group of players attempts to do so, they may all lose to other players that overtake them. To support that speed skaters are involved in such a social dilemma, we analyzed video footage data for 14 mass start skating races to find that skaters that hid behind others to avoid air resistance for a long time before the final lap tended to win. Furthermore, the finish rank of the skaters in mass start races was independent of the record of the same skaters in time-trial races measured in the absence of direct opponents. The results suggest that how to strategically cope with a skater's dilemma may be a key determinant for winning long-distance and high-speed races with direct opponents.

Factors Affecting Cyclists' Chances of Success in Match-Sprint Tournaments

PURPOSE

To further the understanding of elite athlete performance in complex race environments by examining the changes in cyclists' performance between solo time trials and head-to-head racing in match-sprint tournaments.

METHODS

Analyses were derived from official results of cyclists in 61 elite international sprint tournaments (2000-2016), incorporating the results of 2060 male and 1969 female head-to-head match races. Linear mixed modeling of log-transformed qualification and finish ranks was used to determine estimates of performance predictability as intraclass correlation coefficients. Correlations between qualifying performance and final tournament rank were also calculated. Chances of winning head-to-head races were estimated adjusting for the difference in the cyclists' qualifying times. All effects were evaluated using magnitude-based inference.

RESULTS

Minor differences in predictability between qualification time trial and final tournament rank were suggestive of more competitiveness among men in the overall tournament. Performance in the qualification time trial was strongly correlated with, but not fully indicative of, performance in the overall tournament. Correspondingly, being the faster qualifier had a large positive effect on the chances of winning a head-to-head race, but small substantial differences between riders remained after adjustment for time-trial differentials.

CONCLUSIONS

The present study provides further insight into how real-world competition data can be used to investigate elite athlete performance in sports where athletes must directly interact with their opponents. For elite match-sprint cyclists, qualifying time-trial performance largely determines success in the overall tournament, but there is evidence of a consistent match-race ability that modifies the chances of winning head-to-head races.

The Impact of Different Competitive Environments on Pacing and Performance

PURPOSE

In real-life competitive situations, athletes are required to continuously make decisions about how and when to invest their available energy resources. This study attempted to identify how different competitive environments invite elite short-track speed skaters to modify their pacing behavior during head-to-head competition.

METHODS

Lap times of elite 500-, 1000- and 1500-m short-track speed skating competitions between 2011 and 2016 (N = 34,095 races) were collected. Log-transformed lap and finishing times were analyzed with mixed linear models. The fixed effects in the model were sex, season, stage of competition, start position, competition importance, event number per tournament, number of competitors per race, altitude, and time qualification. The random effects of the model were athlete identity and the residual (within-athlete race-to-race variation). Separate analyses were performed for each event.

RESULTS

Several competitive environments, such as the number of competitors in a race (a higher number of competitors evoked most likely a faster initial pace; coefficient of variation [CV] = 1.9-9.3%), the stage of competition (likely to most likely, a slower initial pace was demonstrated in finals; CV = -1.4% to 2.0%), the possibility of time qualification (most likely a faster initial pace; CV = 2.6-5.0%), and competition importance (most likely faster races at the Olympics; CV = 1.3-3.5%), altered the pacing decisions of elite skaters in 1000- and 1500-m events. Stage of competition and start position affected 500-m pacing behavior.

CONCLUSIONS

As demonstrated in this study, different competitive environments evoked modifications in pacing behavior, in particular in the initial phase of the race, emphasizing the importance of athlete-environment interactions, especially during head-to-head competitions.

The Impact of Dry-Land Sprint Start Training on the Short Track Speed Skating Start

Haug, WB, Drinkwater, EJ, Cicero, NJ, Barthell, JA, and Chapman, DW. The impact of dry-land sprint start training on the short track speed skating start. J Strength Cond Res 33(2): 544-548, 2019-This investigation sought to determine the effects of dry-land sprint start training on short track speed skating (STSS) start performance. Nine highly trained short track athletes completed a control period of normal STSS training followed by a 4-week training intervention. Before and after the control and intervention periods, athletes performed 3 electronically timed dry-land and on-ice 14.43 m maximal sprint start efforts. The intervention consisted of 2 sprint sessions per week consisting of 9 electronically timed 14.43 m dry-land sprint starts in addition to normal STSS training. The control period resulted in no substantial change in on-ice start performance (mean Δ: -0.01 seconds, 95% confidence limit [CL]: -0.08 to 0.05 seconds; effect size [ES]: -0.05; trivial); however, a small change was observed in dry-land start performance (mean Δ: -0.07 seconds, 95% CL: -0.13 to -0.02 seconds; ES: -0.49). After brief specific dry-land sprint start training, a small improvement was observed in both on-ice (Mean Δ: -0.07 seconds, 95% CL: -0.13 to -0.01 seconds; ES: -0.33) and dry-land (Mean Δ: -0.04 seconds, 95% CL: -0.09 to 0.00 seconds; ES: -0.29) start performance. This investigation suggests that STSS start performance can be improved through a brief dry-land sprint start training program.

Relay exchanges in elite short track speed skating

In short track speed skating, the relay exchange provides an additional strategic component to races by allowing a team to change the skater involved in the pack race. Typically executed every 1½ laps, it is the belief of skaters and coaches that during this period of the race, time can be gained or lost due to the execution of the relay exchange. As such, the aim of this study was to examine the influence of the relay exchange on a team's progression through a 5000 m relay race. Using data collected from three World Cup relay events during the 2012-2013 season, the time taken to complete the straight for the scenarios with and without the relay exchange were compared at different skating speeds for the corner exit prior to the straight. Overall, the influence of the relay exchange was found to be dependent on this corner exit speed. At slower corner exit speeds (12.01-13.5 m/s), relay exchange straight times were significantly faster than the free skating scenario (P < 0.01). While at faster corner exit speeds (14.01-15 m/s), straight times were significantly slower (P < 0.001). The findings of this study suggest that the current norm of executing relay exchanges every 1½ laps may not be optimal. Instead, varying the frequency of relay exchange execution throughout the race could allow: (1) time to be gained relative to other teams; and (2) facilitate other race strategies by providing an improved opportunity to overtake.

Positive Pacing Strategies Are Utilized by Elite Male and Female Para-cyclists in Short Time Trials in the Velodrome

In para-cycling, competitors are classed based on functional impairment resulting in cyclists with neurological and locomotor impairments competing against each other. In Paralympic competition, classes are combined by using a factoring adjustment to race times to produce the overall medallists. Pacing in short-duration track cycling events is proposed to utilize an “all-out” strategy in able-bodied competition. However, pacing in para-cycling may vary depending on the level of impairment. Analysis of the pacing strategies employed by different classification groups may offer scope for optimal performance; therefore, this study investigated the pacing strategy adopted during the 1-km time trial (TT) and 500-m TT in elite C1 to C3 para-cyclists and able-bodied cyclists. Total times and intermediate split times (125-m intervals; measured to 0.001 s) were obtained from the C1-C3 men's 1-km TT (n = 28) and women's 500-m TT (n = 9) from the 2012 Paralympic Games and the men's 1-km TT (n = 19) and women's 500-m TT (n = 12) from the 2013 UCI World Track Championships from publically available video. Split times were expressed as actual time, factored time (for the para-cyclists) and as a percentage of total time. A two-way analysis of variance was used to investigate differences in split times between the different classifications and the able-bodied cyclists in the men's 1-km TT and between the para-cyclists and able-bodied cyclists in the women's 500-m TT. The importance of position at the first split was investigated with Kendall's Tau-b correlation. The first 125-m split time was the slowest for all cyclists, representing the acceleration phase from a standing start. C2 cyclists were slowest at this 125-m split, probably due to a combination of remaining seated in this acceleration phase and a high proportion of cyclists in this group being trans-femoral amputees. Not all cyclists used aero-bars, preferring to use drop, flat or bullhorn handlebars. Split times increased in the later stages of the race, demonstrating a positive pacing strategy. In the shorter women's 500-m TT, rank at the first split was more strongly correlated with final position than in the longer men's 1-km TT. In conclusion, a positive pacing strategy was adopted by the different para-cycling classes.