Prediction and Analysis of Tokyo Olympic Games Swimming Results: Impact of the COVID-19 Pandemic on Swimmers' Performance

to the COVID-19 pandemic in the 2019–2020 season, swimming competitions and training have been limited leading to a setback in performances. The study analyzed if, during the subsequent season, swimmers' have been able to regain the lost performance. Swimming time trends were analyzed comparing Tokyo with Rio Olympics and with mathematically predicted results. The gap between the gold medalist and the last nalist, and the differences between men and women have also been considered. Swimming competition results of females and males, in 100 m and 200 m Freestyle and Backstroke, were collected from the Olympics' of cial website. Results showed that at Tokyo Olympics almost all swimmers' times improved as compared to Rio's. Analysis of performance trends highlighted that performance progression does not proceed in a linear fashion and that is best predicted by more recent results. Women's progression was higher than men's and the gap between the rst and last nalist constantly decreased, except for the Tokyo Olympics. In conclusion, the unprecedented Tokyo Olympic Games and quali cation year seems not to have disrupted all

trends highlighted that performance progression does not proceed in a linear fashion and that is best predicted by more recent results. Women's progression was higher than men's and the gap between the rst and last nalist constantly decreased, except for the Tokyo Olympics. In conclusion, the unprecedented Tokyo Olympic Games and quali cation year seems not to have disrupted all Olympic swimmers' performance, suggesting that stakeholders support and athlete's coping ability might safeguard the subsistence of performance. Keywords:swimming; performance analysis; performance prediction; COVID-19 pandemic 1. Introduction Periodization training is considered one of the main issues for coaches and athletes, it underlies the process of athlete preparation and is fundamental for setting realistic per- formance goals in the plan for major competition [1,2]. Traditional swimmer's training is characterized by detailed annual plans that comprises performance peeks for at least three competitive events each year. Even more complex is preparing for the Olympic Games which foresees at least 4 years of training periodization, conceived as the purposeful sequencing of different training units, so that athletes could attain the desired state and planned results at the right time [3]. Monitoring athletes using variables that best correlate to actual sports performance is much more important than following theoretical concepts; it is therefore necessary to develop effective and applied methods to train and progress high level athletes for long-term success, according to their speci c requirements [4–7]. Conducting studies to predict future sports results can be helpful for developing appropri- ate training plans and strategies. Observing results from previous years and monitoring the progress and trends in swimming along time makes it possible to estimate and forecast future results and allows to predict the direction in which this discipline is heading [1]. Regarding this, performance analysis, as the investigation of races in competitions, provides an essential role to support the development of athletes from a scienti c perspective [8]. Int. J. Environ. Res. Public Health2022,19, 2110.

analysis, as the investigation of races in competitions, provides an essential role to support the development of athletes from a scienti c perspective [8]. Int. J. Environ. Res. Public Health2022,19, 2110.

Int. J. Environ. Res. Public Health2022,19, 2110 2 of 17 Accordingly, a constant progression in swimmers' performance, particularly in Olympic medalists, has been reported between subsequent Olympic Games and within the pre- Olympic year [2,9–11]. Lately, performance progression of world-ranked swimmers qual- i ed for the Tokyo 2020 Olympics was analyzed during the ve preceding consecutive seasons. A performance improvement of 2–4%, dependent on the stroke and race distance, was reported from 2015 until the 2018–2019 season. On the contrary, a setback in perfor- mance of 1–2% appeared in the 2019–2020 season, likely ascribed to the consequence of the COVID-19 lockdown, and supposedly affecting how the swimmers were preparing for Tokyo 2020 [12]. Multiple features in swimming vary completely from other sports, among them the horizontal body position, the higher energy cost due to the resistive force of the water and the forward propulsion with both arms and legs at the same time, which implies that athletes, to achieve high level in competition, need to promote upper limbs muscle functioning and sensibility to water pressures to a very high level [13–15]. Indeed, it is generally recognized that swimming technique and coordination make the greatest contribution to performance so that the swimmers' skill in reducing water resistance, as well as in applying propulsive forces effectively, may be more important than race duration in dictating the physiological and energetic demands of swimming [13,16–19]. Therefore, the absence of training and competition in the water is a major problem to overcome for swimmers. As none of the training strategies available during con nement would be suitable to replace the in-water training gains, the postponement of the Tokyo event could appear a fair and reasonable decision, at least for swimmers [12]. On the other hand, social distancing precautions enacted to slow the spread of COVID- 19 have affected not only the Tokyo Olympic and Paralympic games, but all other games as well, including the cancelation of quali cation tournaments. These changes have raised a sense of uncertainty, confusion, and frustration, and made it dif cult to set a series of concrete goals [20,21]. A study

social distancing precautions enacted to slow the spread of COVID- 19 have affected not only the Tokyo Olympic and Paralympic games, but all other games as well, including the cancelation of quali cation tournaments. These changes have raised a sense of uncertainty, confusion, and frustration, and made it dif cult to set a series of concrete goals [20,21]. A study exploring subjective perceptions on how Olympic athletes and coaches experienced the postponement of the Olympic Games reported feelings of exceeding demands in the pressurized high-performance environment of Olympic sports and overwhelming physical and mental requirements associated with the undue year of preparation [22]. Overall, due to restricted and dif cult training environments around the globe, it has been challenging for athletes to maintain their best conditions, follow special diets, and work on the individualized tasks to achieve a high level of performance [21]. The COVID-19 pandemic impacting the training schedules of athletes has affected their sleeping habits and caused unhealthy habits and coping mechanisms such as increasing their carbohydrate intake and preferring sedentary behaviors above active ones [23]. It is apparent that the effects of lockdown are more severe and multifaceted than just a scheduled absence from training activities, and they can act as a negative stressor for many athletes indicating that special considerations are needed when athletes return to sport in the event of signi cant levels of detraining [24–26]. Regarding this, decrements in performance have been extensively proven during and after training restriction due to the ongoing pandemic conditions in age-group athletes, in individual and team sports, in sprint and endurance disciplines, at amateur and elite level [24,27–34]. As far as swimming is concerned, the pandemic-induced restrictions were reported to offer performance advantages to sprinters and to be deleterious to long-distance swimming performance. Moreover, reduced training volume during short periods of COVID-19 lockdown were reported to be higher for low level athletes than at the elite national level. Therefore, it can be argued that high level sprint swimmers, training for the Tokyo Olympics, may not have suffered such high levels of adverse effect on performance due to the

deleterious to long-distance swimming performance. Moreover, reduced training volume during short periods of COVID-19 lockdown were reported to be higher for low level athletes than at the elite national level. Therefore, it can be argued that high level sprint swimmers, training for the Tokyo Olympics, may not have suffered such high levels of adverse effect on performance due to the low training volumes [35,36]. For some sportsmen, indeed, the postponement of the Tokyo Olympic merely indicated the temporal shift of timing and adjustment to optimize their peak of performance, so they could have used it as an extended chance to further improve their performance or recover from injuries they might have endured. Overall, these are athletes capable of maintaining optimistic and positive attitudes goals [21]. The nding that many athletes and

Int. J. Environ. Res. Public Health2022,19, 2110 3 of 17 coaches alike viewed the postponement as a chance for pursuing improvement and recovery demonstrated the ability of elite athletes to cope with adversity in their preparation [22]. In the speci c population of Olympic and Paralympic athletes, as well as in soccer top professional division, individual coping styles and psychological exibility, team and social environment and institutional support, made the basis to apply speci c measures allowing their sports performance not to suffer extremely negative effects due to pandemic restrictions [37–40]. Nonetheless, at the Tokyo 2020 Olympics the public was not allowed to attend any competitions. Previous studies showed that the crowd factors provide the most dominant causes of the home advantage, which means that home teams in soccer and rugby union championships competitions win over 50% of the games. Home advantage effects in soccer vary from 51 to 78% depending on the country and division, being higher for top level teams and for full stadiums, suggesting that top athlete's performance can be boosted by a larger presence of spectators [41,42]. As a result of different chances for training and con icting athletes' reactions to the Games postponement and to the closed-door competitions, if the Tokyo Olympics performance outcome of swimmers was hampered by the pandemic emergency is still unknown [1]. Therefore, the question arises, whether this extra-year in the roadmap toward the Tokyo Olympic Games represented an actual opportunity for swimmers to successfully regain the performance lost in the 2019–2020 season, or if otherwise their performance was hindered by training restriction, lack of competitions, Olympic postponement, and the absence of spectators. The aim of the study was to determine if swimmers' performance trend was main- tained at Tokyo 2020 and was not disrupted by the peculiarity of the vent, being thus predictable from mathematical modeling of previous Olympic results. To further analyze the performance trend, the secondary hypotheses were that Tokyo 2020 times were better than Rio's Olympics; that the gap between the gold medalist and the last nisher in the Olympics nals decreased; and that women's results improved

and was not disrupted by the peculiarity of the vent, being thus predictable from mathematical modeling of previous Olympic results. To further analyze the performance trend, the secondary hypotheses were that Tokyo 2020 times were better than Rio's Olympics; that the gap between the gold medalist and the last nisher in the Olympics nals decreased; and that women's results improved more swiftly than men's with smaller differences the longer the swimming distance. 2. Materials and Methods 2.1. Procedures Swimming competition males' and females' results of the 100 m and 200 m Freestyle and Backstroke were collected from the Olympics' of cial websites (www.olympic.org and) (accessed on 5 September 2021). Although some of the analyzed competitions were already included in the Olympic program at London 1908 (men's) and Stockholm 1912 (women's), only starting from Mexico City 1968 all those competitions can be found. For this reason, the Mexico City 1968 Olympic games have been selected as the beginning point of the present analysis. The times of the rst (1 ) and the last (8 ) nalists, and the mean time of all nalists, were the variables analyzed. The study is not interventionary involving animals or humans, and it does not require ethical approval. 2.2. Data Analysis To predict the swimmers' performances at Tokyo 2020, three predictive mathematical models were applied to previous Olympic competition's times: (1) a univariate linear regression analysis (Predicted linear) and (2) a univariate non-linear regression analysis (Predicted non-linear), both applied to the previous 13 Olympics; (3) a linear regression analysis applied to the Olympics undergone after polyurethane swimsuit interdiction only, namely London 2012 and Rio 2016 (Predicted last 2 Olympics). To test the hypothesis that Tokyo Olympics swimmers' performances were close to those predicted, Tokyo performances were compared to the Rio's results for all swimming strokes, to the times estimated in the present study for Freestyle and Backstroke and to the

Int. J. Environ. Res. Public Health2022,19, 2110 4 of 17 times estimated in the study by Holub and colleagues for Breaststroke and Butter y [1]. The differences between Tokyo 2020, Rio 2016 and predicted results were evaluated by a univariate analysis of variance. If necessary, the post-hoc Bonferroni/Dunn test was used to perform multiple pair-wise comparisons. Statistical signi cance was set atp< 0.05. To analyze the trend of 100 m and 200 m Freestyle and Backstroke results for men and women, the gold medalist and the last nisher times, performance changes along Olympics were assessed in second (s), percentage (%) and slope of the regression line of results over time (slope) from 1968 to 2021. The normal distribution of the time data was analyzed with the Shapiro–Wilk test. The lack of an outlier in the data and the data uniformity obtained (Shapiro–Wilk test, statistical signi cance) represented the additional advantage of applying a univariate mathematical model. A rectilinear solid relationship, as well as a high Pearson linear correlation, emerged by the linear and non-linear analysis of the regression (p< 0.001). The univariate analysis of variance con rmed the goodness-of- t between the constructed models and the empirical data. 3. Results Results of the present study analyze swimmer's results in the Freestyle, Backstroke, Breaststroke and Butter y 100 m and 200 m of all Olympic Games since the rst one in which they were all disputed i.e., Mexico City 1968. To analyze all four strokes with same procedure, Figuresa,b anda,b depict the Tokyo 2020 and the predicted results by the linear regression (Predicted linear) undergone in the present work.Int. J. Environ. Res. Public Health 2022, 19, x 5 of 18 Figure 1. Men’s and Women’s 100 m Freestyle results from Mexico City 1968 to Tokyo 2020 Olympic Games, and the values predicted for the latter. Figure 2 shows improving performance in men’s Freestyle 200 m results for the first three Olympic Games by −6.08% followed by a 0.51% worsening at Moscow 1980. Women improved their performance by −8.02%, in the first four Olympics. Subsequent editions showed merely constant progression up to

City 1968 to Tokyo 2020 Olympic Games, and the values predicted for the latter. Figure 2 shows improving performance in men’s Freestyle 200 m results for the first three Olympic Games by −6.08% followed by a 0.51% worsening at Moscow 1980. Women improved their performance by −8.02%, in the first four Olympics. Subsequent editions showed merely constant progression up to Beijing 2008, where the performance improve- ment was by −0.73% for men and −1.97% for women. Subsequent Games of London 2012 suffered worst performances in men (+0.75%) which remained stable in women (−0.08%). In Rio 2016 both men and women improved their time, by −1.05% and −0.98%, respec- tively, while in Tokyo 2020 men showed better improvement than women, by −0.58% and −0.10%, respectively. For the 100 m and 200 m Freestyle, both men and women overcame the predicted time for Tokyo 2020 by +1.8% and +2.0%, respectively. Figure 1. Men's and Women's 100 m Freestyle results from Mexico City 1968 to Tokyo 2020 Olympic Games, and the values predicted for the latter.

Int. J. Environ. Res. Public Health2022,19, 2110 5 of 17Int. J. Environ. Res. Public Health 2022, 19, x 6 of 18 Figure 2. Men’s and Women’s 200 m Freestyle results from Mexico City 1968 to Tokyo 2020 Olympic Games, and the values predicted for the latter. 3.2. Backstroke Figure 3 shows improved performance in men’s and women’s Backstroke 100 m re- sults for the first three Olympic Games by −6.01% followed by a 0.70% worsening at Mos- cow 1980. Women improved their performance by −7.63%, in the first four events. The subsequent edition (1980 in Moscow) yielded a +0.70% worst results in men. Afterwards, men started another improvement phase, while women retained a stagnation in perfor- mance in the following Olympic (1984 Los Angeles). The most substantial progress of −2.27% and −2.54% was achieved in 2008 (Beijing) by men and women, respectively. Sub- sequently men suffered a decrease in performance at London 2012 and women at Rio 2016. Tokyo 2020 results appeared worse than predicted by +1.8% and +1.7% for men and women, respectively. Figure 2. Men's and Women's 200 m Freestyle results from Mexico City 1968 to Tokyo 2020 Olympic Games, and the values predicted for the latter. 3.1. Freestyle Figure Freestyle 100 m results, with steeper progressions in the rst three Olympic Games, by 3.65% and 6.76% for men and women, respectively. The subsequent edition (1980 in Moscow) yielded worst results both in men and women, +0.06% and +0.09%. Afterwards, men started another improvement phase, while women retained a stagnation in perfor- mance in the following Olympic (1984 Los Angeles). The most substantial progress of 2.11% was achieved in 2008 (Beijing) by men. It is worth noting that in London 2012 and in Rio 2016 men suffered a decrease in performance, +0.21% and +0.19%, respectively, while at Tokyo 2020 they showed a return to the 2008 values. Women also demonstrated some improvement in performance in the 2008 edition ( 1.43%), but not bigger than in Seoul 1999 ( 1.55%), maintaining a fairly constant progression from Los Angeles 1984 to Tokyo 2020. It should be noted that both men

decrease in performance, +0.21% and +0.19%, respectively, while at Tokyo 2020 they showed a return to the 2008 values. Women also demonstrated some improvement in performance in the 2008 edition ( 1.43%), but not bigger than in Seoul 1999 ( 1.55%), maintaining a fairly constant progression from Los Angeles 1984 to Tokyo 2020. It should be noted that both men and women overcame the predicted time for Tokyo 2020 by +2.0% and +1.1%, respectively. Figure rst three Olympic Games by 6.08% followed by a 0.51% worsening at Moscow 1980. Women improved their performance by 8.02%, in the rst four Olympics. Subsequent editions showed merely constant progression up to Beijing 2008, where the performance improvement was by 0.73% for men and 1.97% for women. Subsequent Games of London 2012 suffered worst performances in men (+0.75%) which remained stable in women ( 0.08%). In Rio 2016 both men and women improved their time, by 1.05% and 0.98%, respectively, while in Tokyo 2020 men showed better improvement than women, by 0.58% and 0.10%, respectively. For the 100 m and 200 m Freestyle, both men and women overcame the predicted time for Tokyo 2020 by +1.8% and +2.0%, respectively.

The study analyzes swimming performance trends comparing Tokyo with Rio Olympics.