Temporal and Spatial Characteristics of Pacing Strategy in Elite Women’s 400 Meters Hurdles Athletes

ain objective of the study was to assess the pacing strategy of running 400 m hurdles of the world-level female athletes over the past 40 years based on the functional asymmetry -temporal and spatial characteristics. The data were collected from 1983 to 2019 using the review of scienti c literature. Over the 35 years of the study, 37 top-level competitions with 283 nalists-competitors were included. The analysis of the 400 m hurdle covered mainly spatial and temporal factors of the run, related to those technical skills, the level of motor skills, and somatic structure. In addition to the basic statistics, the ANOVA analysis of variance, regression analysis, Pearson correlation, the principal component analysis (PCA), and Kaiser's criterion was used for

with 283 nalists-competitors were included. The analysis of the 400 m hurdle covered mainly spatial and temporal factors of the run, related to those technical skills, the level of motor skills, and somatic structure. In addition to the basic statistics, the ANOVA analysis of variance, regression analysis, Pearson correlation, the principal component analysis (PCA), and Kaiser's criterion was used for the multivariate analysis. The nal result in the 400 mH run is determined not by the simple sum of the individual temporal and/or spatial characteristics of the run (the number of steps, the type of attacking leg, but their interaction in the area of functional asymmetry. The decisive factor in the 400 mH run strategy is the second curve, where the emphasis is on the optimal setting of the stride pattern in the context of minimizing the loss of running speed. Additionally, the application of multidimensional statistical methods is a valuable tool that allows to signi cantly deepen the interpretation of the obtained results, and thus optimize a strategy for a 400 mH run. Keywords: 400 mH run; pacing strategy; asymmetry; temporal and spatial variables; multidimensional statistics 1. Introduction The analysis of asymmetry in the area of biomechanics is useful in assessing the effectiveness and optimizing the performance of a speci c motor structure, thus reducing the risk of injury [1–3]. This mainly applies to the so-called functional asymmetry between the limbs, e.g., differences in the performance of one limb's movement structure about the other [4,5]. Therefore, asymmetry is part of contemporary sports training, implemented in running competitions, and it concerns mainly the time-space characteristics of a single running step, repeated many times [6–8], e.g., in a 400 mH run. Asymmetry (morphological, functional, gait, and fundamental movement patterns) is also related to the health of well- trained runners. Linkage with injuries is well-documented and asymmetry is one of the main intrinsic factors of acute and recurrence injuries [3,8–10]. The 400 m hurdles race is one of the most dif cult athletic competition events [11]. The nal result is determined both by the level of motor skills such as speed, anaerobic

related to the health of well- trained runners. Linkage with injuries is well-documented and asymmetry is one of the main intrinsic factors of acute and recurrence injuries [3,8–10]. The 400 m hurdles race is one of the most dif cult athletic competition events [11]. The nal result is determined both by the level of motor skills such as speed, anaerobic endurance strength [12–15], and the technical ability to overcome the hurdles [16,17]. A Int. J. Environ. Res. Public Health2022,19, 3432.

Int. J. Environ. Res. Public Health2022,19, 3432 2 of 12 high level of motor skills and technical skills directly affects a speci c skill called “hurdle rhythm”. This rhythm is de ned as running with the minimum loss of speed, regardless of the fatigue and pattern of clearing the subsequent hurdles [18–20]. In most cases, it determines success in the hurdle race. Additionally, this rhythm is related to the so-called 400 mH pacing strategy. Attempts to explain this strategy as an effort optimization began in the 1960s, mainly on the analysis of the most important international competitions [20–22]. It includes the application of a certain number of steps between hurdles (15–17) and time variables to cover the distance between these hurdles [23,24]. The 400 m hurdles pacing strategy should include the symmetry aspects, regardless of the sport level at which it is implemented. In this case, symmetry is taking into account two interdependent factors. One concerns the temporal aspects-time divisions (splits) in particular sections of the race: the rst turn, the rst straight line, the second turn, the second straight line. The second aspect refers to the spatial elements de ned as the number of steps between successive hurdles (from 1 to 10 hurdle). These primary factors can be extended to additional factors including the increase in the number of steps(e.g., from 13 to 14),the multiple of these changes in a single run, the ratio of the odd (13,15) to even (14) steps, and the use of the attacking leg-left or right [23,25,26]. Detailed and in-depth analysis of the sports movement observations and the resulting conclusions, requires the application of more and more advanced statistical analyzes. In particular, it concerns the analysis of the time and spatial series of sports movement to extract the features of the movement, that will optimize the result of the sport. In our case, the nal time in the 400 mH race. The multidimensional relationships between measured variables determining the nal result require the usage of multivariate statistical analyses. The article proposes a multidirectional statistical analysis (e.g., the principal component analysis (PCA), ANOVA, regression analysis,

movement to extract the features of the movement, that will optimize the result of the sport. In our case, the nal time in the 400 mH race. The multidimensional relationships between measured variables determining the nal result require the usage of multivariate statistical analyses. The article proposes a multidirectional statistical analysis (e.g., the principal component analysis (PCA), ANOVA, regression analysis, correlation) of hurdles observation, which relates to the analysis of changes in spatial attributes (e.g., number of steps, distance traveled), and non-spatial attributes (e.g., speed and time of crossing each hurdle). This happens in terms of functional symmetry. Therefore, the main aim of this study was to investigate the pacing strategy in Women's 400 m hurdles, studying functional symmetry. Speci cally, we aimed to: (1) study the basic temporal and spatial characteristics of pacing strategy in female 400 m hurdles athletes, (2) overview the relationship between temporal and spatial characteristics, (3) investigate variables of the greatest importance for the nal result, (4) identify main components of strategy by reducing the dimensionality of data set, (5) assess differences between female groups varying sports level. 2. Materials and Methods 2.1. Material Two hundred and eighty-three female 400 mH runners belonging to 23 national teams around the world were analyzed (age: 26.05 3.8 years; body height: 172.57 5.09 cm; body mass: 59.74 4.33.2 kg). One of the statistical analyzes included the division of runners into three groups. Differences between the groups were con rmed by comparing mean personal best times in the of cial 400 m H race: Group A, 53.38 0.42; Group B, 54.51 0.29; and Group C, 55.83 0.67. The study was approved by the Human Ethics Committee of the Opole University of Technology, Opole, Poland (U-KO-212/2021/PO and 25 March 2021). 2.2. Methods Data acquisition was performed through computer searches of scienti c literature for the years 1978–2019 using several primary sources of information: Medline (PubMed), Web of Science, Evidence Database (PEDro), EMBASE, Scopus, and Google Scholar. The search used the following keywords: “400 mH race”. Moreover, this descriptive review was performed by the Preferred Reporting Items for the review statement

2.2. Methods Data acquisition was performed through computer searches of scienti c literature for the years 1978–2019 using several primary sources of information: Medline (PubMed), Web of Science, Evidence Database (PEDro), EMBASE, Scopus, and Google Scholar. The search used the following keywords: “400 mH race”. Moreover, this descriptive review was performed by the Preferred Reporting Items for the review statement (http://www.prisma- statement.org, accessed on 15 September 2021). The literature review made it possible to obtain titles and summaries of all publications included in the keyword. This allowed

Int. J. Environ. Res. Public Health2022,19, 3432 3 of 12 for the isolation of publications for inclusion so that the allocation criteria, the women's 400 mH run in the articles, were met. In this way, the full texts of the manuscripts for the current review were obtained. Only studies focused on 400 mH strategies or pacing strategies in championship competitions are included in the manuscripts. Articles that analyzed data obtained from local competitions or national championships were excluded. The nal analysis included 283 individual runs at championship competitions (Table). All the races concerned were event nals, the competition was held on synthetic tracks, and the time was of cially measured with an accuracy of 0.01 s. The data (time and space) were obtained from materials published over the last 40 years. Table 1. Types of competitions, years of the competition, and the number of competitors participating in the nal of the 400 mH race.Competitions Years Athletes Olympic Games (9) 1984, 1988, 1992, 1996, 2000, 2004, 2008, 2012, 2016 70 World Championships (15) 1983, 1987, 1991, 1993, 1995, 1997, 1999, 2001, 2003, 2005, 2007, 2009, 2011, 2015, 2017112 European Championships (11) 1978, 1982, 1986, 1990, 1994, 1998, 2002, 2006, 2019, 2012, 2014, 2016 85 Friendship Games (1) 1984 8 Olympic Trials (1) 2008 8 Total 1978–2017 283 Variables from three areas were presented in a detailed statistical analysis: 1. Variables characterizing the body structure (BH, BW, BMI) and the level of motor skills (PB 400 m) and technical skills (Technical Index = PB 400 H-PB 400). Additionally, the age of female runners was taken into consideration. 2. Time variables (“split times”)—times of individual sections between hurdles (t1–2,t2–3,. . .t9–10) and times of three characteristic parts of the run (t1–4,t4–7,t7–10). Additionally, the run-up time to the rst hurdle (t0–1) and the nish section time (t10) were taken into account. 3. Spatial variables (“stride pattern”, i.e., the number of steps between successive hurdles)—as in the case of time parameters, successive sections (n1–2,n2–3,. . .n9–10) and selected parts were taken under consideration (n1–4,n4–7,n7–10). 2.3. Statistical Analysis The normality of distributions was assessed using the

run-up time to the rst hurdle (t0–1) and the nish section time (t10) were taken into account. 3. Spatial variables (“stride pattern”, i.e., the number of steps between successive hurdles)—as in the case of time parameters, successive sections (n1–2,n2–3,. . .n9–10) and selected parts were taken under consideration (n1–4,n4–7,n7–10). 2.3. Statistical Analysis The normality of distributions was assessed using the Shapiro–Wilk test. In the pre- sentation of results, basic statistical measures were used, i.e., arithmetic means, standard deviation, minimum and maximum variables, kurtosis, and skewness. The relationships between temporal/spatial parameters and nal results were analyzed using the Pearson cor- relation. To assess the in uence of the tested parameters on the result of the 400 m hurdles, a stepwise forward regression was used. The procedure begins with an equation that contains only a free expression. The rst variable in the equation is the one that has the highest correlation with theYvariable. The variable remains in the equation if the coef cient of regression of the variable differs signi cantly from zero. The next variable introduced into the equation is the one that has the highest correlation withY. (Yhas been adjusted for the effect of the rst variable). If the regression coef cient is signi cant, adding the next variable is implemented in the same way. In studies carried out to reduce the number of variables characterizing the results in 400 m hurdles, the principal component analysis (PCA) was used. The PCA method involving matrix operations is used for multidimensional data exploration, projection, and visualization. PCA analysis results in new principal components which are linear combi- nations of vectors subjected to analysis regarding maximization of variance description. Reduction of data dimensionality is carried out by studying the acquired eigenvalues of principal components to describe the percentage of described data variance. In this study, Kaiser's criterion is used, which eliminates principal components of eigenvalues of less than 1 from further analysis [4].

components to describe the percentage of described data variance. In this study, Kaiser's criterion is used, which eliminates principal components of eigenvalues of less than 1 from further analysis [4].

Int. J. Environ. Res. Public Health2022,19, 3432 4 of 12 The ANOVA test was used to assess the differences of selected variables between groups with different sport levels (high, medium, low). If statistically signi cant differ- ences were identi ed, then detailed posthoc comparisons were calculated. In all analyses, statistical signi cance was set onp= 0.05 (calculatedp< 0.05 was recognized as statistically signi cant). All analyses were carried out with the use of the R programming language with additional packages (R Core Team, 2018). 3. Results Studied female athletes were heterogeneous terms of age. It suggested mean of anthro- pometric measurements suggested a normal range of body height to body mass proportions. However, the minimum–maximum range of BMI indicated underweighted participants including in studies, whereas wide ranges of min-max of sport results suggested differences in sports level of participants. It allowed separating several groups of athletes on different sports levels for her part of the analysis. The most variation of results presented time index (42.31% of variability) compared with 400 mH times and PB 400 m (Table). Table 2. Descriptive statistics of demographic, anthropometric, and basic sports achievements in the entire group of athletes (n= 283). Variables x SD Min–Max Median Skewness Kurtosis r Age (years) 26.05 3.38 18–37 26.05 0.54 0.45 0.7 ** Body height (cm) 172.57 5.09 157–185 172 0.30 0.62 0.19 *** Body mass (kg) 59.74 4.33 45–69 59 0.08 0.15 0.03 BMI 20.06 1.10 16.46–22.72 20.07 0.19 0.07 0.16 ** 400 m H time (s) 54.53 0.86 52.16–56.90 54.6 0.56 0.33 - PB 400 m (s) 51.93 1.22 49.24–56.91 51.90 0.36 0.43 0.36 *** Time Index (s) 2.60 1.0 1.71–5.20 2.62 0.17 0.7 0.10 Time Index, the difference between the 400 m H time and the time of 400 m at; **p 0.01 forr= 0.16,***p 0.001 forr= 0.19. Temporal and spatial characteristics of the pacing strategy are presented in Tables–5. Changes in basic time variables (t1,2,3 . . .) indicate a systematic reduction in running speed, assessed by the time of covering nine 35 m sections (Table). Each of the above races is important for the

time of 400 m at; **p 0.01 forr= 0.16,***p 0.001 forr= 0.19. Temporal and spatial characteristics of the pacing strategy are presented in Tables–5. Changes in basic time variables (t1,2,3 . . .) indicate a systematic reduction in running speed, assessed by the time of covering nine 35 m sections (Table). Each of the above races is important for the nal results. The most signi cant temporal parts of the run are the so-called “entrance in the bend” and “leaving the bend”. Their signi cance is con rmed by bivariate correlations (t5–6,r= 0.73 andt8–9,r= 0.75; Table), whereas the most signi cant spatial characteristics arent5–6andnt6–7(r= 0.36 andr= 0.37, respectively) (Table). Table 3.Descriptive statistics of basic temporal variables (IHU) in the entire group of athletes. Variable x SD Min–Max Median Skewness Kurtosis r t 0–1 6.49 0.16 6.00–7.01 6.50 0.11 0.31 0.34 t 1–2 4.18 0.13 3.86–4.60 4.20 0.22 0.07 0.48 t 2–3 4.27 0.13 4.00–4.70 4.26 0.48 0.07 0.57 t 3–4 4.36 0.13 4.00–4.86 4.35 0.49 0.72 0.55 t 4–5 4.48 0.13 4.20–4.88 4.48 0.54 0.05 0.65 t 5–6 4.62 0.14 4.40–5.10 4.60 0.65 0.07 0.73 t 6–7 4.75 0.16 4.40–5.40 4.74 0.42 0.61 0.71 t 7–8 4.94 0.18 4.50–5.60 4.95 0.48 0.94 0.71 t 8–9 5.11 0.16 4.70–5.80 5.10 0.49 0.61 0.75 t 9–10 5.25 0.18 4.80–6.00 5.27 0.38 0.54 0.68 t 10–F 6.08 0.35 5.21–8.10 6.08 1.23 4.82 0.50 t, the time between particular hurdles;F, nal time in 400 mH;p 0.05; signi cance is marked in bold.

Int. J. Environ. Res. Public Health2022,19, 3432 5 of 12 Table 4. Descriptive statistics of basic and addition spatial variables (IHU) in the entire group of athletes. Variable x SD Min–Max Median Skewness Kurtosis r n 1–2 15.09 0.57 13–17 15 1.18 4.42 0.21 * n 2–3 15.08 0.56 14–17 15 1.35 4.15 0.18 * n 3–4 15.11 0.56 14–17 15 1.27 3.61 0.26 * n 4–5 15.15 0.57 14–17 15 1.23 2.92 0.30 * n 5–6 15.38 0.65 14–17 15 1.30 0.74 0.36 * n 6–7 15.60 0.71 14–17 15 0.71 –0.64 0.37 * n 7–8 16.00 0.78 14–18 16 0.10 –0.86 0.32 * n 8–9 16.38 0.75 15–18 16 0.20 –0.22 0.32 * n 9–10 16.65 0.74 15–19 17 0.01 0.31 0.29 * n, means numbers of steps between particular hurdles;p 0.05, signi cance is marked in bold and *. Table 5.Descriptive statistics of hurdle variable addition and differences (temporal and space). Variable x SD Min–Max Median Skewness Kurtosis r t 1–4 12.81 0.35 12.10–14.00 12.80 0.70 0.60 0.63 *** t 4–7 1.85 0.37 13.10–15.00 13.81 0.46 0.14 0.82 *** t 7–10 15.30 0.46 14.40–17.40 15.29 0.52 0.87 0.81 *** n 1–4 45.28 1.63 41–51 45 1.34 4.17 0.22 ** n 4–7 46.13 1.76 42–51 45 1.21 0.91 0.38 *** n 7–10 49.02 2.03 44–55 49 0.14 0.23 0.34 *** t 4–7–t 1–4 1.04 0.31 0.20–1.80 1.00 0.03 0.10 0.26 *** t 7–10–t 4–7 1.45 0.39 0.60–2.90 1.44 0.32 0.49 0.19 ** t 7–10–t 1–4 2.50 0.51 0.90–3.90 2.50 0.32 0.03 0.30 *** n 4–7–n 1–4 0.84 1.03 0–4 0 0.84 0.45 0.31 *** n 7–10–n 1–4 3.74 1.69 0–10 4 0.37 0.45 0.19 ** n 7–10–n 4–7 2.90 1.41 0–7 3 0.22 0.10 0.01 ***p< 0.001; **p< 0.01; signi cance is marked in bold. The changes in the stride pattern (stride number) in this group were found in the second turn (n4–7). Running speed losses were signi cant in its nal partr(t7–10–t1–4) = 0.92; p< 0.001). Hurdlers who performed well in terms of the fast initial section of the race (=“speed” hurdlers) applied a 15-step rhythm at

This study analyzes pacing strategies in elite women's 400 m hurdles over 40 years.