International Journal of Sports Physiology and Performance, 2015, 10, 1055 -1057 http://dx.doi.org/10.1123/ijspp.2014-0415 © 2015 Human Kinetics, Inc.
brief report
Correction Factors for Photocell Sprint Timing With Flying Start Thomas Haugen, Espen Tønnessen, and Stephen Seiler Purpose: A review of published studies monitoring sprint performance reveals considerable variation in start distance behind the initial timing gate. The aim of the current study was to generate correction factors across varying flying-start distances used in sprint testing with photocells. Methods: Forty-four well-trained junior soccer players (age 18.2 ± 1.0 y, height 175 ± 8 cm, body mass 68.4 ± 8.9 kg) performed sprint testing on an indoor sprint track. They were allocated to 3 groups based on sprintperformance level. Times for 10- and 200-m sprint with foot placement ranging from 0.5 to 15 m back from the initial timing gate were recorded twice for each athlete. Results: Correction-factor equation coefficients were generated for each of the 3 analyzed groups derived from the phase-decay equation y = (y0 – PL) × exp(–k × x) + PL, where y = time difference (0.5-m flying start as reference), x = flying-start distance, y0 is the y value when time is zero, PL (plateau) is the y value at infinite times, and k is the rate constant, expressed in reciprocal of the x-axis time units; if x is in seconds, then k is expressed in inverse seconds. R2 was ≥.998 across all athlete groups and sprint distances, demonstrating excellent goodness of fit. Within-group time differences were significant (P < .05) across all flying-start distance checkpoints for all groups. Between-groups time-saving differences up to 0.04 s were observed between the fastest and the slowest groups (P < .05). Conclusions: Small changes in flying-start distances can cause time differences larger than the typical gains made from specific training, or even the difference between the fastest and slowest elite team-sport athletes. The presented correction factors should facilitate more meaningful comparisons of published sprint-performance results. Keywords: sprint-performance assessment, dual-beamed timing, timing methodology, start momentum, sprint-start procedures, accelerated sprinting Valid and reliable timing is critical for effective monitoring of sprinting performance. In photocell timing, the athlete must start a certain distance back from the initial timing gate to avoid premature triggering caused by a typical starting posture with a forward lean of the upper body. Thus, the athlete’s center of mass is moving or having some momentum before the time initiation; this is called a flying start.1 A review of published studies monitoring speed performance reveals considerable variation in start distance behind the initial timing gate.1–5 Clearly, recorded sprint time decreases as a function of flying-start distance up to a certain point, as a typical sprint-velocity curve follows a hyperbolic relationship,6 but correction factors across varying flying starts have so far not been generated. Therefore, the aim of this study was to generate correction factors across a range of flying-start distances used in sprint testing with photocells. Such information should facilitate more meaningful comparisons of published sprint-performance results.
Methods Subjects Forty-four well-trained junior soccer players (14 female and 30 male, age 18.2 ± 1.0 y, height 175 ± 8 cm, body mass 68.4 ± 8.9 kg) volunteered to participate in the current study. They were all familiarized with the sprint-testing procedures through biannual
Haugen and Tønnessen are with the Norwegian Olympic Federation, Oslo, Norway. Seiler is with the Faculty of Health and Sport Sciences, University of Agder, Kristiansand, Norway. Address author correspondence to Thomas Haugen at [email protected].
or annual testing that their high school or academy performed for training purposes. Written informed consent and parental consent where necessary were obtained from all participants. The study was in accordance to the Helsinki Declaration and approved by the local ethics committee at the University of Agder, Faculty of Health and Sports Science.
Procedures and Apparatus After a standardized warm-up procedure, the participants performed 4 sprints, first 2 × 25 m and then 2 × 40 m, separated by 5 to 6 minutes recovery. Tiptoe front-foot placement was 0.5 m behind the start line. Times for 10- and 20-m sprint with 0.5-, 1-, 1.5-, 2-, 5-, 10-, and 15-m flying start were recorded twice for each athlete and formed the basis for correction-factor generation. The rationale for the design was to facilitate reliability analyses, minimize possible fatigue effects across sprints, and avoid day-to-day performance variability. Sprint times were recorded by 2 identical but independent dualbeamed timing systems (Biorun, Biomekanikk AS, Oslo, Norway). System 1 timing gates covered the start line and each 5 m of the running course. System 2 timing gates were placed 0.5, 10.5, and 20.5 m from the start line during the 25-m sprints and 1, 1.5, 11, 11.5, 21, and 21.5 m from the start line during the 40-m sprints. The first timing gates (0.5–5 m from the start line) were mounted on separate tripods 1.00/1.20 m above ground level, while the remaining timing gates (10–40 m from the start line) were mounted 1.30/1.50 m above ground level. The rationale for this setup was to ensure triggering at the moment the athlete’s chest reached the vertical plane of the nearer edge of the line. The trigger criterion was the first occurrence of both beams being broken for each pair of photocells. The timing system had recently been checked for accuracy.7 1055
1056 Haugen, Tønnessen, and Seiler
Downloaded by Western University on 09/16/16, Volume 10, Article Number 8
Statistics Reliability calculations were based on mean difference between trials, intraclass correlation (ICC), standard error of measurement (SEM), and coefficient of variation (CV). Since we suspected that flying-start correction factors are affected by sprint-performance level, the participants were allocated to 3 equally sized groups (fast, medium, slow) based on 0- to 20-m sprint times with 0.5-m flying start. A general linear model with repeated measures followed by Bonferroni adjustment for multiple comparisons was used to compare within-group sprint-time differences. One-way ANOVA followed by Tukey post hoc test where necessary was used to compare between-groups time differences. Significance was accepted at the P < .05 level. PASW Statistics 18.0 (SPSS, Chicago, IL, USA) was used for the previously mentioned analyses. Graf Pad Prism (version 5.04) was used to generate equations to convert obtained sprint times across flying-start distances. Sprint times with 0.5-m flying start were used as reference. Residual error estimates for the correction-factor equation were calculated and expressed as typical error.
Results Table 1 shows that test–retest reliability (expressed as SEM or CV) for 20-m sprint times was slightly lower for the 0.5- to 2-m flyingstart distances (SEM 0.03 s, CV 1.2–1.4%) than for the 5- to 20-m flying starts (SEM 0.02 s, CV 0.9–1.0%). Test–retest reliability for 10-m sprints showed a similar trend but with slightly higher CV
values across all flying-start distances (1.4–1.8%). No significant 0- to 25-m time differences were observed, eliminating performance decline caused by fatigue across the 4 sprint repetitions. Table 2 shows sprint-performance data and correction-factor equation coefficients across athlete groups. A wide range of binomial and polynomial fitted curves were tested, but the following phasedecay equation showed superior best-fit values: y = (Y0 – PL) × exp(–k × x) + PL, where y = time difference (0.5-m flying start as reference), x = flying-start distance, y0 is the y value when time is zero, PL (plateau) is the y value at infinite times, and k is the rate constant, expressed in reciprocal of the x-axis time units; if x is in seconds, then k is expressed in inverse seconds. R2 was ≥.998 across all athlete groups and sprint distances, demonstrating excellent goodness of fit. Typical prediction error for the correction-factor equations was
brief report
Correction Factors for Photocell Sprint Timing With Flying Start Thomas Haugen, Espen Tønnessen, and Stephen Seiler Purpose: A review of published studies monitoring sprint performance reveals considerable variation in start distance behind the initial timing gate. The aim of the current study was to generate correction factors across varying flying-start distances used in sprint testing with photocells. Methods: Forty-four well-trained junior soccer players (age 18.2 ± 1.0 y, height 175 ± 8 cm, body mass 68.4 ± 8.9 kg) performed sprint testing on an indoor sprint track. They were allocated to 3 groups based on sprintperformance level. Times for 10- and 200-m sprint with foot placement ranging from 0.5 to 15 m back from the initial timing gate were recorded twice for each athlete. Results: Correction-factor equation coefficients were generated for each of the 3 analyzed groups derived from the phase-decay equation y = (y0 – PL) × exp(–k × x) + PL, where y = time difference (0.5-m flying start as reference), x = flying-start distance, y0 is the y value when time is zero, PL (plateau) is the y value at infinite times, and k is the rate constant, expressed in reciprocal of the x-axis time units; if x is in seconds, then k is expressed in inverse seconds. R2 was ≥.998 across all athlete groups and sprint distances, demonstrating excellent goodness of fit. Within-group time differences were significant (P < .05) across all flying-start distance checkpoints for all groups. Between-groups time-saving differences up to 0.04 s were observed between the fastest and the slowest groups (P < .05). Conclusions: Small changes in flying-start distances can cause time differences larger than the typical gains made from specific training, or even the difference between the fastest and slowest elite team-sport athletes. The presented correction factors should facilitate more meaningful comparisons of published sprint-performance results. Keywords: sprint-performance assessment, dual-beamed timing, timing methodology, start momentum, sprint-start procedures, accelerated sprinting Valid and reliable timing is critical for effective monitoring of sprinting performance. In photocell timing, the athlete must start a certain distance back from the initial timing gate to avoid premature triggering caused by a typical starting posture with a forward lean of the upper body. Thus, the athlete’s center of mass is moving or having some momentum before the time initiation; this is called a flying start.1 A review of published studies monitoring speed performance reveals considerable variation in start distance behind the initial timing gate.1–5 Clearly, recorded sprint time decreases as a function of flying-start distance up to a certain point, as a typical sprint-velocity curve follows a hyperbolic relationship,6 but correction factors across varying flying starts have so far not been generated. Therefore, the aim of this study was to generate correction factors across a range of flying-start distances used in sprint testing with photocells. Such information should facilitate more meaningful comparisons of published sprint-performance results.
Methods Subjects Forty-four well-trained junior soccer players (14 female and 30 male, age 18.2 ± 1.0 y, height 175 ± 8 cm, body mass 68.4 ± 8.9 kg) volunteered to participate in the current study. They were all familiarized with the sprint-testing procedures through biannual
Haugen and Tønnessen are with the Norwegian Olympic Federation, Oslo, Norway. Seiler is with the Faculty of Health and Sport Sciences, University of Agder, Kristiansand, Norway. Address author correspondence to Thomas Haugen at [email protected].
or annual testing that their high school or academy performed for training purposes. Written informed consent and parental consent where necessary were obtained from all participants. The study was in accordance to the Helsinki Declaration and approved by the local ethics committee at the University of Agder, Faculty of Health and Sports Science.
Procedures and Apparatus After a standardized warm-up procedure, the participants performed 4 sprints, first 2 × 25 m and then 2 × 40 m, separated by 5 to 6 minutes recovery. Tiptoe front-foot placement was 0.5 m behind the start line. Times for 10- and 20-m sprint with 0.5-, 1-, 1.5-, 2-, 5-, 10-, and 15-m flying start were recorded twice for each athlete and formed the basis for correction-factor generation. The rationale for the design was to facilitate reliability analyses, minimize possible fatigue effects across sprints, and avoid day-to-day performance variability. Sprint times were recorded by 2 identical but independent dualbeamed timing systems (Biorun, Biomekanikk AS, Oslo, Norway). System 1 timing gates covered the start line and each 5 m of the running course. System 2 timing gates were placed 0.5, 10.5, and 20.5 m from the start line during the 25-m sprints and 1, 1.5, 11, 11.5, 21, and 21.5 m from the start line during the 40-m sprints. The first timing gates (0.5–5 m from the start line) were mounted on separate tripods 1.00/1.20 m above ground level, while the remaining timing gates (10–40 m from the start line) were mounted 1.30/1.50 m above ground level. The rationale for this setup was to ensure triggering at the moment the athlete’s chest reached the vertical plane of the nearer edge of the line. The trigger criterion was the first occurrence of both beams being broken for each pair of photocells. The timing system had recently been checked for accuracy.7 1055
1056 Haugen, Tønnessen, and Seiler
Downloaded by Western University on 09/16/16, Volume 10, Article Number 8
Statistics Reliability calculations were based on mean difference between trials, intraclass correlation (ICC), standard error of measurement (SEM), and coefficient of variation (CV). Since we suspected that flying-start correction factors are affected by sprint-performance level, the participants were allocated to 3 equally sized groups (fast, medium, slow) based on 0- to 20-m sprint times with 0.5-m flying start. A general linear model with repeated measures followed by Bonferroni adjustment for multiple comparisons was used to compare within-group sprint-time differences. One-way ANOVA followed by Tukey post hoc test where necessary was used to compare between-groups time differences. Significance was accepted at the P < .05 level. PASW Statistics 18.0 (SPSS, Chicago, IL, USA) was used for the previously mentioned analyses. Graf Pad Prism (version 5.04) was used to generate equations to convert obtained sprint times across flying-start distances. Sprint times with 0.5-m flying start were used as reference. Residual error estimates for the correction-factor equation were calculated and expressed as typical error.
Results Table 1 shows that test–retest reliability (expressed as SEM or CV) for 20-m sprint times was slightly lower for the 0.5- to 2-m flyingstart distances (SEM 0.03 s, CV 1.2–1.4%) than for the 5- to 20-m flying starts (SEM 0.02 s, CV 0.9–1.0%). Test–retest reliability for 10-m sprints showed a similar trend but with slightly higher CV
values across all flying-start distances (1.4–1.8%). No significant 0- to 25-m time differences were observed, eliminating performance decline caused by fatigue across the 4 sprint repetitions. Table 2 shows sprint-performance data and correction-factor equation coefficients across athlete groups. A wide range of binomial and polynomial fitted curves were tested, but the following phasedecay equation showed superior best-fit values: y = (Y0 – PL) × exp(–k × x) + PL, where y = time difference (0.5-m flying start as reference), x = flying-start distance, y0 is the y value when time is zero, PL (plateau) is the y value at infinite times, and k is the rate constant, expressed in reciprocal of the x-axis time units; if x is in seconds, then k is expressed in inverse seconds. R2 was ≥.998 across all athlete groups and sprint distances, demonstrating excellent goodness of fit. Typical prediction error for the correction-factor equations was
