Welcome to the Science of Sport, where we bring you the second, third, and fourth level of analysis you will not find anywhere else.

Be it doping in sport, hot topics like Caster Semenya or Oscar Pistorius, or the dehydration myth, we try to translate the science behind sports and sports performance.

Consider a donation if you like what you see here!

Did you know?
We published The Runner's Body in May 2009. With an average 4.4/5 stars on Amazon.com, it has been receiving positive reviews from runners and non-runners alike.

Available for the Kindle and also in the traditional paper back. It will make a great gift for the runners you know, and helps support our work here on The Science of Sport.

Friday, December 07, 2007

Running Economy Part I

The Zersenay Tadese study - the most economical runner ever?

Yesterday we introduced a new series, Running Economy, inspired by this study, which was published just last week in the British Journal of Sports Medicine. It makes a great start to a series that we'll build on next week, looking at the importance of running economy to performance.

We must of course make the disclaimer that just as VO2max does not by itself guarantee performance, so too running economy is one of many possible variables, both physiological and psychological, that come into play when one tries to understand just what is goes into determining performance. But it certainly seems to be an important one...we spoke yesterday of the growing awareness of running economy as a crucial determinant of performance, and of how little is known about it.

So it is perhaps fitting to begin our series with the very latest work, and then we'll deconstruct the theories as we progress.

The Zersenay Tadese Study

The title of the research article that was published last week was "The key to top-level endurance running performance: A unique example". The researchers, based in Madrid, Spain, had access to Tadese, the World Cross Country champion who put an end to the reign of Kenenisa Bekele earlier this year. He then followed this up with a win in the World Half-Marathon championships, where he beat world record holder Sammy Wanjiru, running 58:59, in a race where some time may have been lost due to tactical manoevuring in the middle part of the race!

Zersenay Tadese is a class act. Regular readers will know that I (Ross) am a particular fan of his, and recently picked him to be the next world record holder in the marathon, when he eventually makes the move up from the track. I think he might even have a more successful marathon career than Bekele. And if he gets the preparation right, Beijing 2008 might also see him challenge Bekele on the track. So it was with some personal interest that we discovered this article, courtesy George.

Looking at the study, it is always remarkable to have access to such a uniquely talented athlete, clearly at the peak of his 'game'. It's always been one of the big barriers to understanding the 'eternal mystery' of the dominance of the Kenyans that they've never really been studied in their prime. Most of the published work done on the East Africans comes from the labs of Bengt Saltin, and while those runners were certainly talented, they were by no means the best Kenyans in their best condition. And of course, there's a potentially crucial difference between the Kenyan who can run 27:30 for 10k compared to the one who runs under 27 minutes! So the fact that Tadese was tested during the month leading up to the World Cross Country Championships, and that he allowed the results to be published is a great coup for the researchers.

So let's take a look at the results...

Blood hematocrit

The first finding of interest is the hematocrit, measured one week before the Championships. It was "only" 43.7%. I say "only" because in another endurance sport, cycling, I suspect that many riders would not even get out of bed on pull on their jerseys if it was this low! In fact, the value of 44% is pretty much in the normal range, considering that he was doing a moderate to high volume of training at the time of testing (about 150km/week).

The authors suggested that this measurement, together with other findings we won't go into here, indicates "no manipulations to artificially increase blood oxygen transport capacity", and "no artificial stimulation of erythropoiesis". I'm sure sceptics are out there (they'd hardly report it if his hematocrit was 58%, would they!), but we'll go along with that one. Let's get onto running economy.

Running economy

We introduced the concept of running economy yesterday, but it bears repeating. Running economy is a measure of how much (or little, as the case may be) oxygen the runner uses for a given, sub-maximal speed. In theory, two runners can have the same maximal capacity for oxygen use (called VO2max), and then the one who is more economical at the sub-maximal speeds is likely to be the better runner.

So there are two measurements we need to look at here. The first is the MAXIMUM - the VO2max. The second is the oxygen use at sub-max, which tells us how economical Tadese is.


Zersenay Tadese's VO2max was reported as 83 ml/kg/min. This is a very high measurement, but that's expected of a world class long distance runner. When compared to other world class athletes, Tadese's value 'disappears' into the crowd. In otherwords, if we lined up a group of athlete's VO2max values and asked you to pick the the two-time World Champion based on VO2max alone, there's a good chance you'd be wrong! Just as you would have been completely wrong when you had to guess that Derek Clayton could run a 2:08 marathon with his VO2max of "only" 69 ml/kg/min.

Of course, when it comes to VO2max, there's a lot of hype around the measurements. It has become something of an urban legend, much like the size of the fish you caught on your last fishing expedition which gets larger every time you re-tell the story! So depending on who you believe, Greg Lemond had a VO2 max of 92.5ml/kg/min, the Cross-Country skier Bjorn Daehlie was at 96 ml/kg/min in the off-season, and the physiologist who tested him predicted that he'd be above 100 ml/kg/min when he was "fit"! We're pretty sure you can relate your own stories of this super-human measurement!

Incidentally, while we're talking super-human, the Siberian huskies who take part in Alaska's Iditarod Sled Dog Race have been reported as having VO2 max values of 240 ml/kg/min! Eat your heart out, Bjorn!

Running Economy

So now we move onto running economy.

First things first – we have to understand what they are measuring. They have measured running economy as the total volume of oxygen needed to run 1 kilometer. This is really important, because there is evidence, shown in the Figure below, that this total volume does not vary too much with running speed. In otherwords, regardless of how fast you run, you use more or less the same volume of oxygen to run 1 kilometer! Of course, there will be individuals who don’t quite follow this trend, and so some runners will become more or less efficient as they run faster, but in general, it stays similar.

Figure showing schematic values of oxygen cost to run 1 km in three
different population groups. Note that the data are schematic only,
and represent pooled values for the three groups. Note also the general
absence of a change in economy as running speed increases.
Redrawn from Foster and Lucia (2006)

Now this might be confusing at first (certainly took some figuring out from our side – it has major implications for our understanding of fatigue, actually), because you’ve no doubt heard that the faster you run, the more oxygen you use. And this is still true – the trick is to understand the units that economy is being reported in. This is illustrated in the following example:

Take a runner who runs along at 4min/km, and uses about 200 ml/kg of O2 to run 1 km. If we look at his O2 use per minute, then we work out that he is using 50 ml/kg/min (because it takes him 4 minutes to run that kilometer and use those 200ml/kg of oxygen). If our runner increases his speed to 3min/km, if we assume that his total oxygen to run 1 km remains 200ml/kg (again, we acknowledge that some people will differ, but generally, as the figure above shows, this is the case). But now, he’s only taking 3 minutes to run the kilometer (and hence use those 200ml/kg of O2). Therefore, his actual use of oxygen PER MINUTE is now 67 ml/kg/min!

So in fact, the use of oxygen PER MINUTE goes up with running speed, but the use per kilometer remains relatively constant. This is important (and has MASSIVE theoretical implications, but that is for another day). The other thing about this figure is that it creates a number of questions – why are the East Africans lower than the Caucasians, for example? But that’s all for the later posts in this series. Let’s get back to Tadese.

Tadese has one of the lowest total volumes of oxygen to run 1 kilometer ever reported. In the article, the authors report the following “normal” values for running economy, measured as ml/kg/km (that is, volume used to run 1 km)

  • Frank Shorter – 192 ml/kg/kilometer (running at 3:06/km)
  • Elite Kenyans (including Olympic medallists) – 192 ml/kg/kilometer (running 3:00/km)
  • Elite Caucasians (Spanish runners) – 211 ml/kg/kilometer (running 3:09/km)
  • Zersenay Tadese – 150 ml/kg/kilometer (running at 3:09//km and at at 2:51/km)

The size of this difference is astounding. We do have reports of runners who have been tested in this range, but we just need to get permission to publish the results (we’re working on that!). But let’s put this into context:

  • Tadese is running along at 3:09/km, which is a 2:13 marathon pace.
  • He is using 150 ml/kg of Oxygen to cover each kilometer.
  • If we now convert this to a volume per MINUTE, we work out that he is using 47.6 mlO2/kg/min. This is absolutely astonishing for how low it is!

We cannot emphasize just how significant this is – quite what it means, and the implication thereof is something we will tackle in another post, possibly even another series (we don’t want to go off on a tangent here!).

What is wrong with this picture?

But just think about something for a moment – Tadese has a VO2max of 83 ml/kg/min, and he runs at 3:09/km using only 48 ml/kg/min. Therefore, he is using only 57% of his “maximum” at this speed.

There is good evidence that athletes are able to run for 2 hours at about 80% of their VO2max. Now, if Tadese did this, he would be able run at 66 ml/kg/min. Given his Running Economy of 150 ml/kg/km (which is 'relatively' independent of speed - see Figure, and was similar for Tadese in the two speeds he was tested at), this would mean he could cover each kilometer in 2 minutes 18 seconds! Even if his running economy worsened quite a lot and he used up 180 ml/kg/km (a 20% increase), he'd still have it in him to run 2:45/km!

Quite clearly, this is not possible. So either Tadese becomes MUCH LESS efficient as running speed increases (which is somewhat out of the ordinary, as the figure demonstrates - slight changes, sure, but not so large), or something else is going on - there's something wrong with this picture. We need to address this, but as I've said, that's another can of worms!

The inevitable disclaimer...

I have no doubt that people will attack some of the assumptions we’ve made, and so we might as well declare right now – “We know, we've made some assumptions here, the key one being that economy (as measured per kilometer) doesn't change with speed, but we use this to illustrate a potentially profound point for our understanding of what limits exercise performance”. Think about it – if you’re running along at 80% of VO2max, then you have the capacity to use more oxygen. Yet you don’t. So how then, can oxygen be limiting, as you’ve probably read many times? Interesting one that…but that’s for another day…

Explaining Tadese’s extra-ordinary running economy

So now we move onto a mechanism – how do we explain this unbelievably low use of oxygen? Well, the authors put a few things out there:

  • Very efficient utilization of oxygen by working muscles
  • Training
  • Muscle fibre distribution
  • Genetic Factors
  • And then the big one, anthropometric characteristics. That is, they measured the mass and size of the calf and found that Tadese had a significantly smaller calf muscle than the Caucasian runners, who had Running economies of 211 ml/kg/km.

With that, which I’m sure many of you are thinking is a very generic set of reasons and mechanisms, we’ll sign off today! But that’s on purpose, because in our next post on the series, we’ll look at what factors contribute to running economy in more detail – what is training, muscle fibre distribution, genetics? And then we can all compare ourselves to Tadese and work out why we’ll never be that economical!

Bye for now!



Ryan said...

You could've put the 2:18/km into perspective. That is about a 1:37marathon! Even the 2:45/km is out of reach by todays standards- 1:56 marathon. You're right, something else is going on.

Sled Dog Action Coalition said...
This comment has been removed by a blog administrator.
Running Physiologist said...

Having done a fair amount of lab testing on international but sub truly elite runners myself I'm shocked by the results of this study (and maybe a little sceptical). I haven't had a chance to read the original article in full text yet, but wonder if faulty O2 analysis equipment is the route of their results. Early this year using an OxyCon Pro our lab found that some of our runners had running economies of 170-180ml/km, which at the time we thought was a great improvement from the 190-200ml/km they were at the same time of the year before. However, once we looked into this in a little more detail (one runner had been injured for most of the year and done no real formal training, plus during our human calibration trial I had a running economy of 195ml/km when I am only a 45min 10km runner and normally have a running economy of 220ml/km) we decided that thing was up! Turns out that the 02 analyser had a problem and although the VO2max values we’re right the running economy was way out.

I hate to be so sceptical but a running economy of 150ml/km just seems to be so much lower than anything I have ever seen before that it makes it almost unbelievable.

Ross Tucker and Jonathan Dugas said...

Hey Running PHysiologist

Thanks for the input - I agree that something is up. Like you, I hate being sceptical and I know both Carl and Alejandro quite well and I rate them very highly, so I'm leaning towards believing the equipment.

The actual value of 150ml/kg/km is incredibly low, but there are other reported values in the mid to high 150's, so that suggests that perhaps the 150 is not too far off the mark (we're trying to get hold of the numbers, and get permission to use them, for use in a later post).

However, having said that, it's the combination of the high VO2max and the low O2 use at the sub-max speeds that adds up to this very peculiar implication, which is that Tadese can run 3:00/km at only 60% of his VO2max.

That is a statistic, predicted by these results, that I really struggle to believe. For one thing, why would he not just run faster and run a 1:45 marathon! Because we know it's possible to run at 80% of VO2max for this period of time?

So I'm with you on this one. I have a feeling that much discussion will come out of this paper. Keep us posted on anything you hear!

Thanks for the comments!


Iker said...
This comment has been removed by the author.
Anonymous said...

Hi Ross

I have been following your comments and findings of running economy, and after reading the article about Zersenay Tedase , I plugged my own numbers from tests conducted at SSI - curiosity. The values for running economy based on your formula came out around the same mark as the “Elite Caucasians” that were mentioned (ie. +-211ml/kg/km) which does not really make sense. Surely there must be error here.

Shouldn't the formula for economy would be better represented as an absolute value of oxygen consumed per kilometer to incorporate weight into the equation?

This way, chunky 80kg me would consume around 16 litres of oxygen per kilometer whilst my skinny Spanish opposite number weighing in at an estimated 58kg would only consume 12 litres.

An estimated 58kg Tedasse would only be using a mere 8.7 liters per km, hence his massive advantage.

This seems to make more sense, doesn't it?

Ross Tucker and Jonathan Dugas said...

Hi Jean

Thanks for the comments. Any chance you can send me the data from that testing and I'll have a look at it as well, just to see how your running economy changes at different speeds?

You have a good point, and I'd like to check the numbers to get an idea of where the response should go...


Anonymous said...

I have just mailed you the data now.

Anonymous said...

Running 2:18/K or even 2:45/K for 2 hours would take quite a bit of muscular strength and neural drive. So possibly he has the oxygen for it, but the neuromuscular component is not there yet.
Or more likely you are dealing with faulty equipment/software bugs as other posters have noted. I do find it odd that his VO2/Max is 83 and he hits only 47 at 2:51/K. That pace has to hurt, it is his half-marathon race pace, and he cannot go that much faster in a 10 K, which means it really hurts. I know what this feels like when your endurance is so good that 10 K and the half marathon become essentially the same event. You do not cruise through a 10 K with a half-marathon effort, it is the other way around - you suffer the 10 K pain for the entire half marathon. My max HR is 174, and I'd be holding 163 the entire half marathon. I've done a VO2 test, you can see the results


Right when I hit RER of 1 and HR of 162 (consistent with half-marathon race pace), my VO2 is 64, and all I could manage on that test was 68.

So I imagine Tedase should be hitting no lower than 70 VO2 at 2:51/K if his max is 83, at least not 47 or anywhere close!

As different as athletes come, they are all human, and all in all are quite similar. When you measure something weird, check the equipment/software first.