In a previous post, we learned that finals swims were faster, on average, than preliminaries swims (link).
Our analysis left open the question whether individual athletes who make finals swim faster in finals than they did in preliminaries. In this post, we’ll confirm that athletes are in fact more likely to swim faster in finals than they did in preliminaries. We’ll also present evidence that athletes who make finals try hard in both the preliminaries and the finals.
The following plot shows that the athletes who make finals do in fact swim faster in finals than they did in the corresponding preliminaries 68% of the time.
The likelihood of swimming faster in finals increases to age 14, after which it declines. At all ages for which we have meaningful data, women are slightly more likely to swim faster in finals than men. In our next post, we’ll see that it’s more difficult for women to make finals.
These results raise further questions. Are finals athletes swimming faster in finals or merely slower in preliminaries? It’s a classic chicken-and-egg dilemma. A priori, we’d expect athletes to swim their fastest in the preliminaries (to make finals), and then swim their fastest in finals (to receive awards and approbation). However, athletes confident of making the finals may hold back in the preliminaries and then swim their best in the finals. Or athletes eager to make finals may try their hardest in preliminaries and then be too exhausted to swim their best in finals.
To address these questions, we’ll compare the likelihood of dropping time in preliminaries for athletes who make finals versus those who don’t. If finals athletes are holding back in preliminaries, they should have a lesser likelihood of dropping time in the preliminaries than non-finals athletes.
This graph plots the likelihood that a preliminaries swim is faster than its seed time, by age. The “Finals” line is for preliminaries swims that qualify for the finals; the “Non-Finals” line is for preliminaries swims that don’t qualify for the finals. This data shows that finals swimmers are significantly more likely to drop time in the preliminaries than non-finals swimmers, which refutes the hypothesis that finals athletes are holding back in the prelims.
We already know that finals swims are faster than non-finals swims, both by construction and by statistics (see above). Now we learn that athletes who make finals are more likely to drop time in the preliminaries than athletes who don’t make finals, which indicates that faster athletes are more likely to improve than slower athletes.
This result subtly challenges a widely-held concept of youth athlete performance. On the one hand, we can rank athletes by their current performance, from fast to slow. We talk about “fast” athletes and “slow” athletes. Under this view, an athlete’s performance is intrinsic to them. Fast athletes stay fast and slow athletes stay slow. On the other hand, we can rank athletes by their rate of improvement, from rapidly improving to stagnant or even declining. In a competitive youth sport, the improving athletes become fast while the stagnant athletes become slow. This result supports the latter view, namely, that improving athletes are the ones who become fast, and that the relevant dimension of youth athlete performance is rate of improvement rather than current performance relative to peers.
We’ve seen that athletes who make finals are more likely to swim faster in finals than they did in preliminaries, and that finals athletes are more likely to drop time than non-finals athletes. We’ve also seen a age-related decline in swimmer improvement after age 13. We’ll examine this issue more closely in a subsequent post (link).