I read an article1 in the Journal of Strength and Conditioning Research today that shocked me.
The first surprise was with the abstract, which implies play in the lower soccer league was faster, so checked the publication date. It was released last month, so it cannot be satire for April Fools’ Day. The full text clarifies that only a meaningless difference was observed, but there were so many logical and procedural absurdities throughout the article that I double- then triple checked the date as my jaw dropped closer and closer to the floor. Nope! This somehow slipped through the peer review process.
Let’s take a look at some of that slippery methodology…
Instead of using measures of actual peak running speed, they used rolling average velocity over 1-, 5-, and 10-minute intervals. That means they took the total distance covered in a 60 second period of play, and divided it by time (60 sec) to calculate average velocity in that minute. This same process was applied using 5- and 10-minute time frames.
It was unsurprising to the authors (and probably anyone who has watched a soccer game) that the average speed in the fastest 1-minute interval was higher than the average speed in the fastest 10-minute interval. A soccer game, like the other forms of football, is intermittent: There are frequent stoppages in play, and in the periods of action between whistles, individual players spend the majority of the time jogging, walking or standing. With a work:rest ratio between 1:10 and 1:152,3, and the work periods randomly distributed, shorter intervals will show more variability than longer ones. This means the “peak average” will be higher in 1-minute intervals compared to 10-minute ones. It also means that the slowest 1-minute period or “valley average” will be slower than that in the slowest 10-minute period. Yet the “average averages” will be the same.
This study underlines the point that these intervals are not sensitive enough to detect the actual demands in soccer or differences between levels of play. Players perform high intensity efforts only about once per minute, however their actual frequency is distributed rather randomly4. Each effort lasts from less than a second (a jump or a kick) to maybe 10 seconds (a counter-attack) and its the ability to perform these efforts well that determine the outcome of the match5,6. Consequently, even one-minute averages are too long of an interval to capture what contributes to team or player success.
So far, it is not a bad study, rather it is one that probably did not need to be conducted. It is not contributing positively to our understanding of sport science, but it is not entirely harmless either.
The authors conclusions that “Peak running speeds seem to have little difference when comparing competition levels…” is patently wrong. They did not measure peak running speed. They measured average speed over 1-, 5- and 10-minute intervals. Furthermore, prior to calculating average speed, they removed the data corresponding to the fastest sprints achieved by the players.
Without explanation they removed all data when the players were moving faster than 10 m/s. Granted that is very fast, but many decent athletes are capable of achieving that velocity in a sprint, so it seems only running, jogging, walking and standing were included in the analysis. Accelerations greater than ≥ 6m/s2 were also filtered out by the researchers. This threshold easily surpassed by athletes (measured over meters 5-10 in a sprint)7.
To summarize, they measured the pace of activities that was neither sprinting nor involved aggressive accelerations. Next, they averaged the pace of these light activities over time scales that that are orders of magnitude longer than the high-speed work intervals in a match. They then refer to this variable as “peak running speed” which they noted was different between positions and levels of play.
How did this get past peer review? It must be April Fools’.
-CG
1 Fahey, J. T., Aldred, K., Greig, M., & Rhodes, D. (2023). Peak Running Speeds in Professional Male Football: Influence of Division and Playing Position. The Journal of Strength & Conditioning Research, 10-1519.
2 Gabbett, T. J., & Mulvey, M. J. (2008). Time-motion analysis of small-sided training games and competition in elite women soccer players. The Journal of Strength & Conditioning Research, 22(2), 543-552.
3 O’Donoghue, P. G. (2002). Time-motion analysis of work-rate in English FA Premier League soccer. International Journal of Performance Analysis in Sport, 2(1), 36-43.
4 Nedelec, M., McCall, A., Carling, C., Legall, F., Berthoin, S., & Dupont, G. (2014). The influence of soccer playing actions on the recovery kinetics after a soccer match. The Journal of Strength & Conditioning Research, 28(6), 1517-1523.
5 Faude, O., Koch, T., & Meyer, T. (2012). Straight sprinting is the most frequent action in goal situations in professional football. Journal of sports sciences, 30(7), 625-631.
6 Martínez-Hernández, D., Quinn, M., & Jones, P. (2022). Linear advancing actions followed by deceleration and turn are the most common movements preceding goals in male professional soccer. Science and Medicine in Football, 1-9.
7 McBride, J. M., Blow, D., Kirby, T. J., Haines, T. L., Dayne, A. M., & Triplett, N. T. (2009). Relationship between maximal squat strength and five, ten, and forty yard sprint times. The Journal of Strength & Conditioning Research, 23(6), 1633-1636.