Tuesday, July 09, 2013

Leaps and Bounds of a Watery Kind

Something a bit different today.


Just a chart showing the progression in the men's 1500 metre swimming world record since records began for the event over a century ago for long course, i.e. 50 metre pools.

The data is from this Wikipedia link

Nothing really specific, other than to post up an example of the way progression in performance has occurred over a century has not been linear nor predictable in other ways but rather occurs with quite different rates of improvement. It was prompted by a discussion about, you guessed it, using past performances (i.e. cycling mountain ascent times) or the ubiquitous estimations of power to body mass ratios as a means to calibrate a modern day "dopeometer", which is of course a path fraught with problems.

Lasting of the order of 15 to 20 minutes, the elite 1500m swim is an event dominated by an athlete's aerobic metabolic capabilities, their morphology and water drag characteristics and with quite a deal of technique/form involved, e.g. making best use of turns, and there are far more degrees of freedom of movement than say pedalling bicycle cranks permits.

Looking at the chart we can see rapid improvement occurred in the 1920s, then only gradual change until the late 1950s when there was a consistent and significant rate of improvement for 20 years through until the late 1970s. I'm guessing improved access to suitable facilities enabling more athletes to compete played a big role in helping to drive this rapid change, along with presumably improvements in training, technique and so on.

Since then the improvement has been far more incremental despite the 1990s and 2000s being the EPO era and the 2000s the era of the swimsuit technology wars. What I haven't done though is to consider the change in power demand for these more incremental performance improvements, i.e. does a small change in speed require a significant change in power? I am presently not well versed in how linear or curvilinear the speed versus power relationship is for swimmers. No doubt there has been plenty of research into this.

Of the 46 new world records plotted, 28 (61%) were by American or Australian swimmers. The only "eastern bloc" athlete in this list is Vladimir Salnikov of the former Soviet Union with 3 records set in the early 1980s. Current world record holder is Sun Yang of China with his swim at the 2012 London Olympic Games.

Edit to add:
After Charles' comment about looking at other shorter swimming events - I plotted the progression with the 400 metre world record as well, and overlayed the two - and adjusted the time scales so the relative progression can be directly compared.

A broadly similar pattern, which is not surprising as you'd expect similar means of performance improvement, but the progression with the 400m event is more consistent than for the 1500m event.

The 400m swim, from a energy demand perspective, is similar to cycling's individual pursuit, and I'd expect roughly one-quarter to one-third of the energy demand is met by anaerobic metabolism (compared with say 10% for the 1500m), the balance of course supplied by aerobic glycolosis.


Andrew R. Coggan, Ph.D. said...

1. Nice graph!

2. The relationship between metabolic demand and swimming speed is, perhaps a bit surprisingly, close to linear. The explanation given is that while the metabolic cost of forward motion increases non-linearly with speed, the metabolic cost of horizontal support simultaneously decreases.

Charles (SolarEnergy) Couturier said...


This is obviously very interesting for a swim person (which I am).

Just for sake of discussion, and with no intent to divert too much from the topic, I know that Brent Rushall as expressed a few concerns on the relatively slow rate of improvement for this particular event, for what it's worth.

Could be interesting to see how improvement over say 200free or even better, the 400im compares to that of the 1500. But I guess that'd be my job to do it, as it's probably not your intent.


Rod Havriluk from Swim Technology Research has what comes as close as you can get to a power meter for swimmers. A device that measures force applied, or required to swim. Can be used in a pool (as opposed to a flume). He is very nice to speak with, should you ever have any request for literature etc.

Thanks to both of you Alex, and Andy for your note on this surprising linear relationship.

Charles G. Couturier

Alex Simmons said...

Andy - interesting. Briefly looking at published papers (Toussaint et al) I see a cubic relationship between power and swimmer's velocity, but I haven't studied it closely (and hadn't intended to).

Charles - see extra chart - shows a very similar pattern but less dramatic changes in rate of performance progression compared with 1500m.

I picked the 1500m as the duration of the event means it's dominantly an aerobic effort (as a sensible comparison to road cycling).

The 400m is as well, but with presumably an anaerobic metabolic energy demand of roughly a quarter to a third, very similar to cycling's individual pursuit.

Charles (SolarEnergy) Couturier said...

Hi Alex, thanks you didn't have to..

Were you doing this analysis with road cycling in mind? (since most noise these days is about TdF), or did you have Track cycling in mind.

I'm asking this because I'm (doing my best) helping a master level track rider, with no previous experience or knowledge of what this sport actually is.

In fact, and you don't need to, but I'm wondering how the improvement over a ~4min long event on track compares to ~4min long event in the pool.

Both sports have probably been strongly improved over time, in accordance with our understanding of underlying physics.

Improvements in economy (swim technique, cycling cda) have often resulted into mini breakthrough. Like this time trial taking place between Lemond and the French guy, where the former showed up with a new bike triathlon steering.

That kind of stuff beats dope, in fact, I teach my athletes that it is actually doping, positive and good doping. Any way to improve your performances cheating the right way, being more aero, buying an entertaining trainer to work harder in winter, working with power, working with a coach, knowing an outstanding physiotherapist (triathlon), all that stuff adds up you know...

So yeah, there's some domestic know how, AUS, US for the 1500 over time, and the young Yang is trained in Australia, not in China. And that knowledge in itself is probably superior, at least in swimming, to illegal dope.

Hey thanks!

Charles (SolarEnergy) Couturier said...

As for road cycling, the only way for maybe getting some clues by reading power files, would be to get the smartest folks on earth to have access to all the files, training files and everything, and be allowed to maybe include this in a greater investigation, matching testimonies with PMC values at that time of year, drawing ratios between TSB values and CP points, etc... Only then can one (smart enough) go like Hmmmm, story doesn't match my analysis.

And that ideal scenario doesn't make sense at all, because if the law would impose this, then rider would be disadvantaged training and racing with power.

So this path really turns me mad.

Alex Simmons said...

I can't find a source for 4km IP world record history. The 4km IP event goes back to the 1964 Olympic Games, I'm not sure about before then at world champs.

It used to be a 5km IP. There are many confounding factors, as the discipline has been held on tracks that have varied in technology a lot and the distance and how the race is run (qualifying and finals) as well as different amateur and professional events.

The analysis was just an interesting means to see if progression for a WR was consistent or came in steps. It seems to be a bit of both for swimming.

Swimming is interesting given the consistent use of a 50 metre pool. Of course the pools have improved, but not nearly as much as the impact of the differences in velodromes over that time.

Charles (SolarEnergy) Couturier said...

I get it. Thank you for this nice exposé.

For your information, since I believe you're based in Australia.

The human being having achieved what could be considered as one of the most suspicious swim career/performances has to be Shelley Taylor-Smith. She's a long distance marathon swimmer. To the best of my knowledge, the swimming world at the time did not consider her achievements as being suspicious. I don't personally think it was.

Yet, in 1991, she won the overall FINA World Cup Championship, finishing in top spot, in front of all the males.

Never have I heard of a single occurrence of a female beating all the males in a noble cyclic endurance sport.

So things can get quite weird in swimming due to the highly technical component. We now understand, a few decades after, why she dominated the scene to this extent.

Anyway, made in Australia :)