The effect of current race position on change-over time in swimming relay races

Abstract

Short change-over times (COT) in swimming relays can play an important role in winning a race. Therefore, experts recommend to specifically train change-overs between swimmers to minimize COT (Maglischo, 2003; Siders, 2010; Saavedra et al., 2014). But, change-overs in swimming constitute a quite risky situation, since a false start and hence a disqualification is a possible outcome. Therefore, we assumed that also cognitive factors can play a crucial role in performing a change-over in a competition. We applied the assumptions of the New Prospect Theory of Kahneman & Tversky (1979) to such situations. And this theory states, that the pain of suffering a loss is much higher than the pleasure that comes from an equal-sized gain and thus loss aversion leads to risk-averse decisions. Our work group recently found out, that the team members of a relay team that won a medal in an important competition showed significantly longer COTs than non-medal-winners. A possible explanation of this finding is that the risk-taking of medal winners is reduced compared to non-medal winners because they have something to lose. In this study outlined here, we were interested in the question, what happens with the risk-taking in change-overs within the race when temporal lag to the team behind or in front is short versus long. Therefore, we analyzed top-class international 100m freestyle relay competitions over a 10 years period including three Olympic games and 5 World Championships. We considered the current race situation on all three change-overs within a relay race. We performed two linear mixed models for longitudinal data analyses with the fixed effects current race position at change-over (medal rank vs. no medal rank) and temporal lag to the athlete behind/in front (short vs. long) and the random effects start block model (OSB 9 vs. OSB 11), type of competition (OG vs. WC), and relay position (2nd, 3rd, 4th). The main findings were, that the temporal lap to the relay team behind affects COT more strongly than the temporal lap to the relay team in front. In line with the predictions of NPT, swimmers currently placed within the medal ranks exhibit significantly shorter COTs when the relay team directly behind exerted great pressure due to a short temporal lag and thus increasing the probability of a loss.

Full text

The effect of current race position on change-over time in
swimming relay races
Claudia Braun, Sebastian Fischer & Armin Kibele
University of Kassel
II. Theoretical Background
Short change-over times (COT) in swimming relays can play an important
role in winning a race. Therefore, experts recommend to specifically train
change-overs between team members to minimize COT (Maglischo, 2003;
Siders, 2010; Saavedra et al., 2014). However, in a real competition, COT
could be influenced by other factors such as the race position of a relay
team or the order of fast and slower team members. Derived from the New
Prospect Theory (NPT) of Kahneman and Tversky (1979) a relay race in
swimming is comparable to a mixed lottery: wins and losses are possible
outcomes. In such cases, loss aversion leads to risk-averse decisions.
According to change-overs in relay races, Fischer, Braun & Kibele (2019)
showed significantly longer COTs for relay teams currently placed within the
medal ranks (to avoid the risk of a false start, thus losing the race) instead of
outside the medal ranks (willing to take the chance of reaching the medal
ranks). In case of bad options (definite loss compared to a possible higher
loss) diminishing sensitivity leads to higher risk-taking propensity
(Kahnemann, 2015).
IV. Method
Top-class international 4x100m freestyle races were analyzed across a 10-
year period including three Olympic Games (OG; 2008, 2012, 2016) and five
World Championships (WC; 2007, 2009, 2011, 2013, 2015). The examined
data was downloaded from an open access website and verified by a second
website (http://www.swimrankings.net; http://www.omegatiming.com). A
total of 220 swimmers (116 female, 104 male) were included in this study
with an average participation of 1.7 ± 1.2 races. We performed two linear
mixed models (LMM) for longitudinal data analyses with the fixed effects:
current race position at change-over (medal rank vs.no medal rank) and
temporal lagto the swimmer behind/in front (short vs. long); and the
random effects: start block model (OSB 9 vs. OSB 11), type of competition
(OG vs. WC), and relay position (2nd, 3rd, 4th).
I. Research Question
Does the temporal lag (short vs. long) to the relay team behind or in front
influences risk behavior regarding change-over timein swimming relays?
III. Main Findings
The temporal lag to the relay team behind affects COT more strongly than
the temporal lag to the relay team in front. In line with the predictions of
NPT, swimmers currently placed within the medal ranks exhibit significantly
shorter COTs when the relay team directly behind exerted great pressure
due to a short temporal lag and thus increasing the probability of a loss.
VI. Discussion
This study shows that COT of relay teams currently placed on a medal rank was
significantly affected by the temporal lagto the relay team behind. When the
team behind was close, COTs were significantly shorter compared to a longer
temporal lagto the team behind. In such situations, the probability of a loss is
increased which led to risk-seeking behavior (higher probability of a false start)
on the start block (Kahneman & Tversky, 1979). In case of greater certainty to
keep the current medal rank, relay teams exhibit on average longer COTs
compared to swimmers of a relay team currently not positioned in the medal
ranks (Fischer, Braun & Kibele, 2019). A short temporal lagto the team in front
did not affect COTs of teams currently placed on a medal rank which reflects
stronger weighting of losses compared to wins as predicted by the NPT. COTs of
relay teams currently not placed on a medal rank were neither affected by the
temporal lagto the team behind nor in front. They showed a steady stronger
risk-seeking behavior –likely based on diminishing sensitivity - irrespective of
short or long temporal lags to the front or the back.
0,00
0,05
0,10
0,15
0,20
0,25
0,30
0,35
0,40
medal no medal medal no medal medal no medal medal no medal
short long short long
temporal lag to the team behind temporal lag to the team in front
change-over time [s]
V. Results
Fig.1 Mean COTs and standard errors as a function of current race position at change over and temporal lag to the team behind or in front
** ****
VII. Literature
Fischer, S., Braun, C., & Kibele, A. (2019). Jason Lezak again and again - linear mixed
modelling analysis of change-over times in relay swimming races. Journal of
Sports Sciences, 1–8.
Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under
risk. Econometrica, 47(2), 263–292.
Maglischo, E. W. (2003). Swimming fastest: The essential reference on technique,
training, and program design. Champaign and Ill: Human Kinetics.
Saavedra, J. M., García-Hermoso, A., Escalante, Y., Dominguez, A. M., Arellano, R., &
Navarro, F. (2014). Relationship between exchange block time in swim
starts and final performance in relay races in international championships.
Journal of Sports Sciences, 32(19), 1783–1789.
Siders, W. (2010). Competitive swimming relay exchange times: A descriptive study.
International Journal of Sports Science and Coaching, 5(3), 381–387.
Correspondence address: cbraun@uni-kassel.de
The LMM revealed a significant interaction for the current race position at change-over and the temporal lag to the team behind,F(3, 2115) = 12.5, p< .001.
Furthermore, the interaction for the current race position at change-over and the temporal lag to the team in front was significant, F(3, 1637) = 6.2, p< .001. In
both analyses random effects (start block model, type of competition, relay position) were taken into account.