The following article is a synopsis of “Material Advantage” which was published in Physics World in July 2012. It can be found here.
When I went into the field of sports engineering, people would say to me “it’s ruining sport, it’s just about who has the best equipment”. This put me in a dilemma: I wanted to defend my work by proving that it worked, but not so much that it ruined the sport. It’s easy enough in the lab to show that one pair of shoes is better than the other but what actually happens in practice? The problem I had then was that I couldn’t quantify the effect of technology – a new swim suit, say – and compare it to the improvement caused by an athlete – for example, Usain Bolt.
What to do? Well, the answer was obvious. Give it to a PhD student and ask him to work it out.
The question we asked of him (meet Leon Foster) was the following
“What interventions can you see in performance statistics for any sport and what is the size of the effect?”
Given that I’ve been working on sports engineering for more than a generation and that I can conclusively prove that things like stiffer rackets or aerodynamic helmets work, I thought that it would be pretty easy to see these things in the statistics. I should’ve known better. There are surprises and intrigues galore in the data, a few of which I will reveal here as a taster of things to come.
How do you compare the effects of a new swimsuit with the emergence of Usain Bolt?
The first thing to do is amass your data: world or Olympic records aren’t good enough as there just isn’t enough data. What you need is not only the best performance in every year, but the top-25 performances in every year. That gives you enough resolution to see how the worlds best performances change over the years.
So what we have now (thanks Leon) is the performance statistics in all men’s and women’s track and field going back to the 1890s. The only problem now is that some measures are in time (running, swimming etc.) while others are in distance (long-jump, javelin, pole-vault etc) and comparing between them is not obvious. The article in Physics World gives details. Let’s just say that what we do is estimate the energy required to do the event and see how much it has changed over time. A period of time that works really well for analysis is that between the London Olympics of 1948 and 2012, i.e. 64 years and 16 Olympiads. Given that we haven’t had the London 2012 Olympics yet, we’ll just use data to 2010 and make some comments on where we think we’ll be in couple of months time.
Comparing like with like
The figure above describes concisely what we found for 3 events: the men’s 100-m sprint, men’s javelin and women’s 100-m swimming freestyle. The way to read the figure is as follows:
- The overall length of each bar shows how much each sport has improved in percent between 1948 and 2010 using the top-25 performances in every year;
- The sections of the bar to the right of the zero line show interventions and effects that have been allowed and that have improved performance;
- The sections of the bar to the left of the zero line show interventions that were banned or worsened performance.
What can we see?
Equipment, coaching, drugs and the Olympic effect
In every sport, the majority of the performance increase is due to global improvements: nutrition, coaching, professionalism, and the increase in the number of athletes taking part in elite sport. Technology can make a significant, if smaller part in performance gains – hollow javelins, goggles, swimming hats and swimsuits. Rule changes have a significant effect too such as that in 1984 in the javelin or the 2010 ban on full body polyurethane swimsuits. The creation of the World Anti-Doping Agency in 2000 also coincides with a small yet significant fall in performance in the men’s javelin and 100-m sprint.
The Olympic Games itself has a noticeable effect on performance in most sports we have studied. Performance oscillates to reach a peak in Olympic year and dip 2 years after. There is even a noticeable step change the year Usain Bolt burst onto the scene in 2008, his incredible performances aren’t enough to explain the improvement, it’s as though every other sprinter raised their game in order to compete.
So back to the question, “which increases performance more, Usain Bolt or a new swimsuit?” The answer is very clear, the swimsuits between 2000 and 2010 had a combined effect in the women’s 100-m freestyle of around 10% compared to the 1.4% in the year Usain Bolt appeared. That’s probably one of the reasons why we still have Usain Bolt, while the swimsuits are now consigned to history.
More detail can be found in “Material Advantage” which was published in Physics World in July 2012. It can be found here.