Statistical analysis suggests possible current in the Rio Olympics swimming pool
Several news outlets, beginning with The Wall Street Journal, are reporting that the swimming pool in Rio may have had a current that biased athletes’ performances. This is based on a statistical analysis of athlete performances across the meet, conducted by Indiana University’s Joel Stager and his coworkers. According to WSJ, Stager et al. analyzed times of athletes in the preliminary, semifinal, and final races of the 50m, 800m, and 1500m events and found consistent evidence that swimmers in the higher numbered lanes swam faster when moving toward the starting block and swimmers in the lower numbered lanes swam faster when moving toward the turn end of the pool. A separate analysis by Barry Revzin at Swim Swam came to similar conclusions about the direction and magnitude of lane effect in Rio.
Past questions about lane bias
This is not the first time questions have been raised about a current-induced bias in competition pools. In fact, Stager and his colleagues published an analysis in 2014 that suggested a similar bias in the pool used for the 2013 World Championships in Barcelona. That pool was a temporary pool built specifically for the competition by Myrtha Pools and was disassembled immediately after, before Stager et al.’s analysis was published.
A more recent paper by Stager and his colleagues found that lane bias seems to be more prevalent in temporary pools than in permanent ones. The Rio Olympics pool, like the 2013 Worlds pool, is a temporary pool also built by Myrtha Pools.
Myrtha Pools responds to the criticism
Myrtha responded to both WSJ and Swim Swam by sharing videos (1, 2) of their current test, which was conducted before the competition and on Day 3 of competition. The videos show a floating object in one of the outside lanes; neither video shows any noticeable movement of the object.
Fluid dynamics and swimming pool design
Competitive swimming pools are complicated recirculating systems that can contain special structures intended to minimize interactions between competitors. Myrtha has built many special event pools in recent years, including ones where the results did not show a bias. According to their website, Myrtha has fluid dynamicists on staff and uses computational fluid dynamics (CFD) to analyze pool performance during design, although they only show examples of freeform pools – not competition pools.
In fact, I have found remarkably few CFD analyses of swimming pools in the literature. Most papers seem to focus on distribution of disinfectants in pools or in predicting evaporation rates – both practical problems but ones with limited relevance to this particular question.
So, is there a current in the Rio pool?
It’s tough to say with certainty that there is a current in Rio’s pool. The performance analyses by Stager et al. and by Revzin do show anomalies in the times of athletes in Rio based on their swim lane, and they show that those anomalies do not exist in many other recent competitions.
I also do not think Myrtha’s current test constitutes evidence of a lack of current. Their floating object is only indicative of conditions at the air-water interface. Swimmers ride lower in the water and spend significant time completely underwater. Lane markers may also damp any flow effects near the surface.
I think introducing dye underwater in the pool would do more to reveal any flow that may exist, and this would be a worthwhile test to conduct prior to the deconstruction of the Rio Olympic pool. Additionally, it would be wonderful to see a CFD analysis of the swimming pool, but this would require significant detail about the pool’s design (inlet and outlet locations, etc.) some of which is likely proprietary information.
Neither dye visualization nor CFD simulation will change the results of this competition, but it may help reveal underlying issues in temporary pool designs so that any bias can be avoided in future competitions.
(Image credit: Rio City Government)
Special thanks to @MicahJGreen for bringing this story to my attention and to Dave B. for his assistance.