A competition-based design to assess performance of a squad of elite athletes

There is need for valid and powerful research designs to assess performance effects of interventions in squads of elite athletes.

The objective of this study is to develop a design for investigating effects on competition performance using performance of athletes in other squads as a control.

We used competition swim times downloaded from USAswimming.org for a season ending in the US Open and assumed an intervention had been applied to athletes in one of the larger squads (Ford) at one competition (Santa Clara). Data were included only for swimmers who achieved >900 Hy-Tek points at the USA Swimming Nationals. Each swimmer's points were used to select their best event. Times for the resulting 368 best-event swims in seven competitions by 148 swimmers in 19 squads were analyzed to determine the uncertainty (90% confidence interval) of the effect of the hypothetical intervention. Further analyses were performed with other selection criteria. Uncertainties were compared with those in other recent studies of competitive senior swimmers.

Uncertainty in the effect of an intervention applied to Ford for Santa Clara would have been ±0.8%. Applying other data-selection criteria resulted in generally more uncertainty. Uncertainties in recent studies of competitive swimmers using conventional designs ranged between ±0.7% and ±2.2%.

For the sport of swimming, the effects with this new design are at least as precise as those of conventional research designs using performance tests, and the outcomes are likely to have higher validity. The new design should be useful for assessing the effect of an intervention representing a substantial change from a baseline of usual practice in any sport where athletes compete often against athletes of other squads.