September 2, 2008

Why Are Athletes Barely Better Than Their Competitors?

I'll admit I know nothing about sports, so I am asking for help on this one.

In writing the post on Olympic bravado and all around anti-Americanism, I read this an article in Time Magazine about the French vs. American swim teams, the significance of which I did not appreciate until later.

The article wrote,

The French swimmers had promised to "smash" the Americans in the 4 × 100-m freestyle relay, but the U.S. men took the gold with a last-millisecond comeback by anchor Jason Lezak. The team--Michael Phelps, Garrett Weber-Gale, Cullen Jones and Lezak-- smashed the world record by nearly 4 seconds.
So that's impressive, but doesn't that mean that the French also smashed the world record by 4 seconds?

In fact, all of the medal winners beat the world record by almost 4 seconds.  Almost every country competing beat the world record.

Yet the difference between gold and silver was 0.08 seconds.

A cursory look over other events revealed this is not uncommon: everyone is suddenly a lot better than the best ever was, but they're still barely better than each other.

So how is it that from one Olympics to the next, there is such massive improvement, yet the difference between each team is still (to me) amazingly small?

It's not a case of the far right of the bell curve, where the most elite athletes are close together in ability.  In fact, the bell curve would predict that as you move further to the right, there would be fewer people with similar scores.

It doesn't happen in other areas.  The two richest people don't have anywhere near the same amount of money.  The two tallest people on earth are different by half a foot.  Etc.

My guess is that it must have to do with the competition itself, seeing the other competitors and pushing yourself accordingly.  But if that's the case, wouldn't the best thing for coaches to do is train their athletes in isolation, and tell them the other guy is six seconds faster than he actually is?


Long-time lurker, first-tim... (Below threshold)

September 2, 2008 11:33 PM | Posted by Hunter: | Reply

Long-time lurker, first-time poster: this year's massive improvement is due mostly, from what I can gather, to recent advances in swimsuit technology. Speedo's developed a suit based on the surface properties of sharkskin, so it has some absurdly small amount of drag, and pretty much every competing team was wearing it. Countries sponsored by other swimsuit companies were even allowed to let their swimmers wear the Speedo suits, because the sponsors knew they wouldn't stand a chance otherwise.
That, and I hear tell that China really wanted records set at their Olympics, so they did things like make the pool deeper, which supposedly cuts down on drag.

Vote up Vote down Report this comment Score: 5 (5 votes cast)
I think that it's just that... (Below threshold)

September 2, 2008 11:38 PM | Posted by Fargo: | Reply

I think that it's just that the same improvements in technique, gear, and training filter out to everyone, making the performance differences come down to specific implementations of these things. Athletics seems to be like technology. The differences between competing products and standards are very small, and often based on exactly the same outline, if not exactly the same protocols. I imagine, and someone willing to actually research this could probably prove it one way or the other, that the only times you see significant gaps between competitors is during times when one side had early access to something game changing. Like a lightweight bicycle, radical breathing exercises, and junk like that.

Like yourself, I really don't know much about sports, so this is mostly raw conjecture.

Vote up Vote down Report this comment Score: 2 (2 votes cast)
I would be interested to se... (Below threshold)

September 3, 2008 12:29 AM | Posted by Chris: | Reply

I would be interested to see if more world records were broken in Beijing than in Athens, this article reports 43 world records in Beijing, I couldn't find any data on Athens.

Clicking through the world record progressions on I only saw 7 world records set in Athens, so I think the increase can be mostly attributed to the pool, which was 50% deeper and had gutters on the side lanes to absorb waves. The expected 7 world records and Phelps additional 8 is still 10 short of the 25 we saw in Beijing.

Something similiar happened in 91 in Tokyo where a track was built harder than IAAF rules permitted, Carl Lewis and Mike Powell both set WR's there.

In the absence of an athlete, like Usain Bolt whose genetic gifts outclass everyone in the field, the community is small enough for the rapid propagation of new training techniques and research. I think this, along with the increasing frequency of competitors training and challenging each other throughout the year creates a competitive environment where the peak performance level rises somewhat uniformly with the top athletes in that field.

I would argue we do see a number of athletes to the far right of the curve, and their dominance extends until training methods bring the average conditioning past where their innate gifts and training took them.

Vote up Vote down Report this comment Score: 1 (1 votes cast)
But if that's the case, ... (Below threshold)

September 3, 2008 1:09 AM | Posted by Brad: | Reply

But if that's the case, wouldn't the best thing for coaches to do is train their athletes in isolation, and tell them the other guy is six seconds faster than he actually is?

Well, yeah, sort of. The psychological component in competition is so prominent that it seems to eclipse almost all of the other factors. It's amazing what a person can achieve if they haven't been told it's impossible.

For instance, running a four-minute mile -- once proclaimed impossible -- is now pretty common, at least among professional runners, because, well, once one guy does it, it isn't impossible anymore, and that psychological barrier goes away for all future runners.

And there was a weight-lifter who broke a record because his trainer told him he was actually lifting a lighter weight, one that he had lifted successfully several times before.

But I get what you mean about how disturbing the proximity of the outcomes in modern competitions is. If you can't beat your opponent decisively, then what's the point of competing? I suppose that if you can consistently beat everyone else by 0.08 seconds, that might mean something, but to me it seems rather like winning an election 51% to 49% -- if your opponent gets a recount, it could easily end up going either way.

One last thought: If you pick a thousand non-runners at random and have them do a 100m dash, you'll get your bell curve, 'cause you're measuring untrained ability. But training intensively in such a specialized activity is going to flatten that curve right out, kind of the same way I couldn't tell you the quantitative difference between (say) two top classical sopranos performing the same piece. Yes, I'm sure there's a measurable difference between their A-sharps -- but not one that's meaningful to me.

Vote up Vote down Report this comment Score: 4 (6 votes cast)
Regarding the results patte... (Below threshold)

September 3, 2008 10:20 AM | Posted by MedsVsTherapy: | Reply

Regarding the results pattern- yes, part of it is kind of an arms race. They stay close in competition, training, etc. The technology - pool design, swimsuit design, etc., is just part of the deal. The limit of performance is obviously not known. What is known is that it is very difficult psychologically to perform much beyond some expectation. In the "cold pressor task," and experimental paradigm for investigating pain-realted phenomena, expectations are clearly associated with behavior/performance. In the cold-pressor task, the guinea-pig undergraduate holds his or her arm in freezing cold water, the experience of which quickly moves from "uncomfortable" to "painful." The exposure is actually not life-threatening or physiologically damaging, but sure feels like dangerous pain. The undergraduate is asked to keep the arm in the water, despite the pain, for as long as possible. So, the experimentor can do variations to explore dimensions of the psychology of pain. If told that people rarely keep the arm in the freezing water for one minute, people will generally keep their arm in the water for about one minute. If told two minutes is the norm, then people will generally keep their arm in the water for two minutes. In each case, the guinea-pig undergraduate will testify that he or she was truly at the limit of tolerating the pain. You can google or cuil or medline cold-pressor and find this body of lit. I believe it speaks to the athletic performance issue: rising to the challenge. Many of us have finished a 10K or a marathon and pondered how much farther we could have run - most will say: I could not have lasted for another mile. How can it be that out of 10,000 runners of a 10K, most can run that distance but not 11K? The ultimate implication is, like I mentioned before, and like the sub-4 minute mile issue, or circumnavigating the globe by boat or plane, that we really can't assuredly know the limits of human performance.

Vote up Vote down Report this comment Score: 2 (2 votes cast)
I thought it was the swimsu... (Below threshold)

September 3, 2008 10:31 AM | Posted by Jim: | Reply

I thought it was the swimsuits, also, but surprisingly, the answer seems to be in the design of the pool. Many of the successful swimmers didn't wear Speedo's revolutionary new suit.

Apparently, optimizing the pool's dimensions and adding high-tech features that absorb turbulence made a big difference. Here's a more detailed explanation.

I recall a similar thing happening in track and field around 20 years ago when new artificial running surfaces were developed.

Vote up Vote down Report this comment Score: 1 (1 votes cast)
The reason for the world re... (Below threshold)

September 4, 2008 5:52 PM | Posted by Dean Jackson: | Reply

The reason for the world record break being enormous - four seconds - was advanced technology in the swimsuits, not advanced training or athletics on the part of the swimmers. The pools have also changed; note that folks in the middle lanes almost always win, due to fractionally reduced drag (and top-seeded competitors being put there).

We have better science of nutrition and of doping, and you tend to watch what works, ask or steal the secret from other coaches, and go with that. We crush the records of yesteryear because "full time athlete" is now a profession.

That said, swimming might be the worst sport of them all to test this theory on right now.

Vote up Vote down Report this comment Score: 0 (0 votes cast)
1. The curve actually occur... (Below threshold)

September 5, 2008 11:35 AM | Posted by asphaltjesus: | Reply

1. The curve actually occurs at the olympics too. It's just that the slower competitors are weeded out in preliminaries, which for the most part do not get televised. To get to the final 10 or so swimmers, there are *lots* of also-rans.

To say it a different way, the curve is still there, it's just when you get to the last 10 or so of the fastest athletes in most olympic sports, you are on the far right of the curve.

2. I second the pool design and swimsuits comments. Some of the details mentioned aren't accurate, but the general idea is the Chinese built a *really* fast pool and the suits are known to be much faster.

3. National records all over the world fell in droves earlier in the year because of the suits. It's just that swimming is not followed in the media very much.

Vote up Vote down Report this comment Score: 0 (0 votes cast)
The 4 second difference was... (Below threshold)

September 13, 2008 2:05 AM | Posted by bsmooth: | Reply

The 4 second difference was due to the race being a relay, with 4 different swimmers, each of whom improved upon world record splits (each 100m leg of the relay) by approximately 1 second.

No individual swimmer in a 400m race smashed a world record by 4 seconds in Beijing.

This relay factor, along with the other performancing-enhancing factors listed above, explains the larger than expected differential.

Vote up Vote down Report this comment Score: 0 (0 votes cast)
here's your answer, there a... (Below threshold)

October 2, 2008 10:21 AM | Posted by Camas: | Reply

here's your answer, there arent enough athletes using steroids. and that is not a joke, look at how far above the norm athletes on steroids are.

Vote up Vote down Report this comment Score: 0 (0 votes cast)