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.

Wednesday, July 01, 2009

Oscar Pistorius continued

A more diverse range of "non heavy-duty" thoughts and opinions on Oscar Pistorius

Well, after what was actually an epic post on the science of Oscar Pistorius yesterday, I decided it might be good to present a less scientific argument on the same topic. As I mentioned yesterday, the in-depth, "heavy-duty" dissection of scientific method was not the purpose of this site, but it was the next step in the ongoing debate over the Pistorius advantage.

I can't stress enough that the type of analysis of the research done on Pistorius should be done for all science, and it should have been part of the CAS deliberations. It wasn't, and so instead of a fair hearing and verdict, Pistorius was, through the combined efforts of science and law, able to hijack the hearing and take the verdict without any scrutiny of the research. That kind of scrutiny, even at a basic level, would have revealed that:

  • He was compared to distance runners on many occasions, not sprinters
  • The tests performed on him were not fully explained - the methods seemed to have differed from one athlete to the next, probably because they were done over a long time period and therefore direct comparisons are invalid
  • His state of training may have been completely different to the athletes he was compared to, which would massively influence the findings and comparison
  • Those tests may also be unreliable for the purposes they were used - comparing two athletes requires a different approach than generating a database
  • They are very easy to manipulate, in the absence of independent verification
  • Comparisons were made with selected athletes, without explanation of why other athletes' data was ignored - at one point, Pistorius is compared to 1 sprinter and 2 distance runners, when in theory, 4 sprinters were available for comparison. Apart from the fact that the comparison is wrong, the lack of transparency represents a major problem in an issue this controversial.
  • It was concluded that Pistorius was "essentially the same", when in fact the statistical method used by the paper revealed a significant difference
All in all, the paper is fraught with error - above is the short summary, because I know yesterday was a monster post, and so for those who didn't have the time to plough through it, that's the summary - there's more to it, of course, but feel free to spend some time in the epic analysis!

Some other viewpoints

For today though, I thought it might be good to have a more debatable post, more as a filler than anything. So here are a couple of links that might be of interest.

The first comes from a former elite 400m runner in South Africa, and a person whose insight and expertise I really respect and value. Arnaud Malherbe is still the SA record holder at 44.59s, and his blog has described some of the issues around Pistorius and his selection for the World Championships in Berlin later this year. They are well worth reading, because his approach is less scientific, and more legal and logical. His insight is that of a former athlete, someone who understands the sport intimately, and I think the points he makes are valid, particularly regarding selection for the team. It's always interesting to get the views of an athlete, rather than the rather heavy science you had thrown at you yesterday!

Blade Runner Part I
Blade Runner Part II
Blade Runner Part III

The third article in particular deals with the possibility of selection if he reaches a qualifying standard - that currently seems a long way off - a 1.61 second improvement is required within the next two weeks and he's been well off the sort of form required. But, he runs is Oslo this Friday, and the possibility exists.

Then here are some other interesting thoughts from a PhD student, written last year, but only discovered today. Karl Zelik is a PhD student at the University of Michigan. His PhD topic? The biomechanics of locomotion, amputees and the use of prosthetic technology to aid movement. So he is one person who has insight gained from years of immersion in the subject.

He presents some quotes from Prof Herr, who was actually one of the authors one the paper that I discussed yesterday. One quote in particular stands out:

"Nevertheless, Herr has publicly stated that within 20-30 years, he predicts the Paralympics will be faster than the Olympics. To the Boston Globe, he further commented that "Even today, some people pity those with disabilities. In the future, [the disabled] will be physically more capable. And then, being physically unique will no longer inspire pity. It'll be unique. And even sexy."

The research produced by Pistorius suggests this prediction has not yet been fulfilled. But yesterday's post was all about how that research could have, and should have been dealt with in a matter befitting its quality - it should have been discarded through careful scientific scrutiny, not the whirlwind hearing the CAS gave it. Perhaps 20 to 30 years is a conservative estimate. I hope so - I would love to see a 41 second 400m, and then we'll tackle the topic once again, when Pandora's box is open.

That's it for Pistorius (for now). I am hopeful that common sense, scientific validity and correct research will prevail and the issue will again find its way into research labs or courts, this time, the right way. But until then, enough on this topic. The summer sports year is in full swing this weekend - the Tour, Oslo Golden League, Wimbledon, and the build-up to Berlin, so that's where our attention turns

Join us then!


leagz said...

excellent article but you omit a horrific scenario. you quote within 20-30 years ... the Paralympics will be faster than the Olympics ... And even sexy.
Many athletes will take extreme measures to win - if there is open competition, could that include unnecessary amputations?

louissyfer said...

I had the same dreadful thought as leagz.

I hope you and others will or have submitted your findings to the proper sectors.

I respect Pistorius greatly but he has an obvious advantage with his locomotion legs. Even my non running friends can see it. And he has no worry of calf and shin injuries.

Ray said...

...not to mention PF...

Responding to "leagz" and "lucifer", Oscar was amputated as a baby, and didn't have to re-learn to walk and run without feet. The horrific scenario would be parents doing the same to their children, in the hopes of producing a world-class athlete.

Ross Tucker and Jonathan Dugas said...

Hi all

Yes, that would be a potential concern, perhaps in the future.

Ray is right - it might require amputation at birth. However, I wouldn't rule out that one of the most obvious technological advances that will be made is to improve the ease of use of prosthetics. They're obviously very difficult to balance on, and so what Pistorius does do very well is make use of his equipment better than anyone else.

That is largely a function of his childhood and the fact that he learned to walk and run on prosthetics. It wouldn't surprise me if they soon develop blades that offer the same performance, but with easier use and more comfort, bringing the scenario more into play, if people would go that far! It's almost inconceivable, but unfortunately, not completely so!

Just to add, I know that the IAAF do read the posts, and so they know what my take is on the research. Whether they choose to do anything, that's up to them, I don't think I'd push it anymore than just trying to get my opinion out.

But they know of this latest development, they know what the research problems are, and perhaps they'll take it up in the future.


markie said...

People seem to support Pistorius either out of pity or what they feel as a sense of fairness. If people said his legs were bionic, that might change the perception, especially if a bionic runner eventually crushes the world record by more than a couple of seconds (which would clearly demonstrate an unfair advantage). Not sure if this future scenario of amputation at birth for bionic limb replacement is very likely though as I doubt anybody is willing to place all their kid's legs in one basket (so to speak) on being the world's number one 400m (or 800m) runner.

Ken Jakalski said...

Hi Dr. Tucker!
I don’t necessarily disagree with your conclusion that Pistorius may have a competitive advantage. In fact, I think you are probably correct, but I can’t say, from a scientific standpoint, that I understand why you feel as you do.

Rather than presenting scientific and data-based arguments, many of your comments were unsubstantiated claims and allegations that struck me as things that a non-scientist, like me, might say had I lacked insight on the corroborative studies presented in the paper.
You indicated that the fatigue test data are “very easy to manipulate, in the absence of independent verification.” However, the literature indicates these tests are remarkably reproducible, and my experience on the track indicates precisely the same thing.

I am very familiar with that formula in Bundle, Hoyt, Weyand (2003) and Weyand/Bundle ‘05, since we have purchased the rights the patented algorithm from Rice University, and have been using it with our athletes for the past four years.

You went on to note this in your original blog: “Pistorius could control the result of this testing by stopping early, given that he knew the theory is that he fatigues less quickly than able-bodied runners.”

This conjecture suggests that you are probably not aware of the testing procedures relative to the ASR protocol. To manipulate the trials, as you noted in your first blog, means he’d have to do that so well that he could conform to the exact number of seconds to run without knowing what that number was. That allegation, given how tight the actual and predicted values in figure 3 in the just released JAP paper are, is simply not supported by the consistency of the data.

You also point out that the “state of training may have been completely different to the athletes he was compared to, which would massively influence the findings and comparison.”

It’s hard for me to understand how you could indicate that state of training is an issue that would “massively influence” the findings and comparison if you had read the ASR papers, or the equation given in the Pistorius paper, which shows that the predictions are based very directly on the athlete’s: 1) aerobic fitness (speed at VO2max) and 2) sprinting fitness (top speed). If you do recognize that speed at VO2max and top speed can change with training, and you have read the equation in the Pistorius paper that includes this term, I don’t know why you would offer that conditioning is not properly taken into account when it clearly was by the two fitness related terms above in the equation.

From my experience using the ASR formula with over 300 athletes over the past four years, I can clearly attest that it’s accurate for sprint and endurance athletes, and sensitive to the athlete’s state of training precisely because it takes their sprint and aerobic conditioning fully into account.
I don’t understand what leads you to believe that this paper does “nothing to dispel old arguments or introduce new points to the debate.” I found the data on the speed-duration relationship and sprinting mechanics added different and important data compared to the informative results published by Bruggeman et al. Pistorius’s very short minimum swing time at top speed was certainly new. That data was highly compelling for those who, like me, believe that there indeed may be a mechanics advantage to Pistorius’s carbon fiber blades.

What is the basis for your conclusion that Pistorius has an advantage? I have no doubt about your opinion on this issue, but from your posts I cannot tell what is the scientific basis for the position you have taken.


Ken Jakalski
Lisle, IL USA

Ken Jakalski said...

Hi again, Dr. Tucker!

A little about me:

I am a high school track coach in Lisle, Illinois, a suburb 40 miles west of Chicago. In 1997, two paralympians, Tony Volpentest and Marlon Shirley, came to compete here in Lisle as a result of my involvement with their coach at the time, Bryan Hoddle. Marlon Shirley’s focus was the high jump. Tony Volpentest raced both the 100 and 200 meters. I assembled an elite Masters field to race against Volpentest. In the 100 meter dash, he finished 4th in 11.66. However, in the 200, he dominated the field, posting a 22.94, which at that time was an ‘unofficial’ paralympic record. I wanted to know how this guy could run as fast as he did considering he had neither feet nor lower arms (in fact, he rests his stumps on padded paint cans in order to start).

My questions and concerns led me to the locomotion labs at Harvard and then at Rice. I was fortunate to spend a good amount of time with the researchers, who understood that my focus was on determining what might be the implications of their work for able bodied sprinters. After all, I spent a career training for the mechanics coaches believed were important for speed at that time - active push-off in the later phase of stance and correct arm swing to improve force contribution. And I was looking at a young man who didn’t have feet for push off or lowers arms to swing effectively, yet I witnessed him running faster than 96% of every able bodied sprinter I had coached up to that time.

Trying to figure out the mechanics implications for able bodied sprinters has become one of my passions in the latter third of my thirty-five year high school coaching career. What was it that the Flex Foot was really doing for Volpentest and Shirley? Did the Flex Foot give them some kind of advantage that far outweighed their disabilities? What is the Flex Foot actually ‘doing’? If it is actually acting as a spring, what is it in the mechanics of able bodied sprinters it is trying to mimic—or improve? By today’s standards, the keel bars both these athletes used would be considered quite crude. Yet even though the current Cheetah “blade” appears far more elegant, the damn thing still attaches to the back of a prosthetics sleeve the same way Volpentest’s and Shirley’s did: with bolts.

Although this previous comment might suggest that I am biased in that I’m basing my views on the contribution of prosthetics due to their structure and design, I assure you that I’ve approached the research findings with an open mind, and even though I marveled at the performance of the paralympians I observed competing on my home track, I had to know whether the efforts of Pistorius were indeed, as Marlon Shirley described them in a Sports Illustrated article, “an act--not sports."

Thanks again,

Ken Jakalski
Lisle High School
Lisle, IL USA

Ross Tucker and Jonathan Dugas said...

Hi Ken

Thanks for the emails and questions, it's great to hear from you! Thank you for presenting them in a balanced way, that's also very appreciated!

To respond:

The first point I was making about the potential to manipulate the tests was mostly to raise the issue that no independent observer was every present, and so there is potential to control the outcome. I hear what you're saying about Figure 3 though, and it's a fair point. I did actually read about the tests, last year in fact, because a few journalists contacted me for opinion once the CAS cleared Pistorius. Obviously, at that point, the exact findings were not known, but I did know that they'd used the ASR protocol.

To answer the second point, the state of training is important. I realise that state of training affects both the top speed and the speed at VO2 Peak (Note that it is a peak, and not a max, as was referred to in the paper).

You'll know from your own coaching and work with sprinters that top speed and speed at VO2 max do not necessarily respond to training in the same way. I know that sprinters who I have worked with often start the season with close to the same anaerobic capacity or speed, but their fitness is well below par.

So what I am saying is that unless Pistorius is in the same condition as those he is being compared to, one cannot infer "similar" fatiguability, particularly when that comparison is being made with different types of runners. A fit and well-trained Pistorius would thus be able to maintain the slower speeds for longer, and the effect would be the "flatten" out the exponential decay. Fig 3 would then look totally different, as one "anchor point" would have shifted.

I would suggest that any individual athlete, with a training intervention, would see the shape of their curve change - the comparison that must be made is not between Athlete A and B, but between Athlete A before and after training, and I'm not sure that has been done.

More to the point, though, the post yesterday (this was just a summary) was an analysis of the paper, which is very poor. The methods are not explained properly for PIstorius, subjects are matched to find results (why not show the data of 4 other sprinters? If, as you say, the ASR test is so reliable and they have data for 300 people, where is it? It should have been presented)

My opinion is therefore not necessarily to be found in this post - this post was never intended as a means to put across why he has the advantage. Rather, it was a discussion around the flaws in the research that cleared him, and they are numerous. Remember that this is a debate that has been going on for 18 months, and I've stated many times before why I believe he has an advantage. I think maybe you're looking for this single, isolated post to give that answer, but there is a broader context. THis post was just an insight into the research, and I believe that research to be very flawed and manipulated through subject selection and study design.


Ross Tucker and Jonathan Dugas said...

Hi again Ken

Been thinking more abut the concept of fatigue and those tests.

One of the most important differences between elite athletes and "also-rans" (like me) is not top speed, but the ability to sustain a sub-maximal speeds/intensities for longer. This ability is almost entirely a function of training status.

I know from work with cyclists that the ability to ride at a sustained, sub-maximal power output is the first thing to improve with training. Sprint(top) speed does not change, and may even come down slightly depending on the type of training being performed, but without question, the state of training impacts on the decay in time to exhaustion at a range of speeds. It has to, and conversely, any training intervention that improves "fatigue tolerance" (and a 400m race is pretty much all about fatigue tolerance) will change the shape of the curve for a given subject. Because of the nature of this study, and its desire to compare one athlete to the "norm", this becomes very important. Note that it's quite different when you want to characterize an athlete, or describe a population - this was a study with no purpose other than to find whether OP was similar to others. Therefore, it's essential that this variable be controlled for.

So given a fitter, and better trained Pistorius, I would expect the following results from the Weyand study to change:

First, his speed at VO2 peak would be higher. As it is, it's greater than that of the able-bodied runners, even though the authors somehow conclude it's "essentially the same", when the difference is equal to 5 SDs. This would imply an advantage from the outset.

Second, his running economy would either remain 17% better than the able bodied runners, or may even improve (since economy can improve with training). This would again suggest an advantage because the cost of running would be reduced, meaning less reliance on anaerobic energy supplies.

Third, his VO2 "peak" would be higher than it is in the paper, which means his "engine" would be comparable to those of the able-bodied runners. Weyand has hung much of his argument on the fact that Pistorius may use less O2, but that his "engine" is smaller to start with - the state of training would change this finding completely. All of a sudden, PIstorius would have a similar VO2peak and still sprint at a lower VO2.

Finally, a fit and trained Pistorius would fatigue less rapidly - his curve would be flatter as a result of greater fatigue tolerance as he'd run for longer at the slower speeds. His top speed would however not change. Remember that his current top speed is 10.8 m/s, which he held for only 2 seconds. With training, would that change? Maybe a tiny amount - maybe 11m/s, but not the same magnitude of change as for the sustained running ability.

This change would produce a different shaped curve to what you currently see in Fig.3

So, point is, the state of training affects everything, and it wasn't controlled.

Note that this doesn't even bring into the debate the fact that he's being compared to distance runners, which is really only so that his economy falls within 2SD of the mean. The study design was flawed, the methods poorly explained, and it cannot be used to judge that he is "similar" to anyone.

Hopefully that makes it a little clearer. I must stress again that this post is the most recent of many discussing the issue, and the purpose was not to state why I thought an advantage existed. I'd done that before. This post was more to point out that the CAS ruling was made with some very dubious data.


Ken Jakalski said...

Hi Dr. Tucker!

I appreciate your quick response to a controversial subject. You have developed reasons why you believe that not controlling for state of training is a critical flaw in the fatigue tests presented in the Weyand et al Pistorius study. I still disagree. The data in several publications show precisely the opposite of what you are claiming.

You note: “Unless Pistorius is in the same condition as those he is being compared to, one cannot infer "similar" fatiguability, particularly when that comparison is being made with different types of runners.”

But the conclusion of the 2005 Weyand & Bundle AJP paper (Energetics of high speed running: integrating classical theory and contemporary observations) indicates sprinters, mid-distance runners and endurance runners all fatigue in the same way:

“The similar fractional utilization of the anaerobic and aerobic power available during high-speed running……….explains why the high-speed running performances of different event specialists can be accurately predicted (R2 = 0.97; n = 254) from two direct measurements and the same exponential time constant.”

The data in the 2003 Bundle, Hoyt and Weyand paper show exactly the same thing. Unless you find flaws with those papers, I don’t understand your point.

“A fit and well-trained Pistorius would thus be able to maintain the slower speeds for longer, and the effect would be the "flatten" out the exponential decay. Fig 3 would then look totally different, as one "anchor point" would have shifted.”

Again, I’ll disagree. It would look exactly the same because equation 1 (and Fig 3B) expresses performance relative to top speed and speed at VO2max and not in absolute terms.

Both the 2003 and 2005 paper show that athletes who differ markedly in sprinting and aerobic fitness fall on the same exact curve (i.e. same flatness as described by the ASAR exponent and equation 1).

I do agree that the absolute speeds of Pistorius or any other athlete will change with their conditioning, but these changes are fully predicted by the two variables presented in eq 1: top speed and speed at VO2max. This is why the ASR is such a valuable training tool for me, precisely because it is accurate regardless of the athlete’s specialization and state of training. Working with high school athletes, this is extremely important for the very reason that training levels—and ages—can be so different.
You also note the following:
“So, point is, the state of training affects everything, and it wasn't controlled.”

The above claim is directly contradicted by several papers and hundreds of ASR trials that have now been in the literature for 4-6 years. You indicate you’re familiar with ASR and these papers, but unless I’m way off base here, you seem to be missing the point.

I’m also at odds with your insight on the importance of Pistorius’s running speed at VO2max:

“OP hits his O2 peak at a speed of 5.0 m/s, whereas the control sprinters hit it at 4.9 m/s……. the difference of 0.1 m/s translates into about 2 seconds in a 400m race at the speeds reported.”

How did you arrive at this? To determine the 400 m performance difference that a 5.0 vs. 4.9 m/s speed at VO2max makes for a runner with Pistorius’s top speed of 10.8 m/s, all you have to do is plug the values into equation 1. This difference is not the 2 seconds you present. It’s roughly 1/10th of your figure (less three tenths of a second).

I get the impression that you are suggesting that a sprinter's 400 m race pace is their VO2max speed. Is this accurate?

Ken Jakalski
Lisle High School

Ross Tucker and Jonathan Dugas said...


You're taking this post, and my subsequent comments, completely out of context and developing a straw man which you are now criticizing.

So I'll attempt to explain one more time, and then just refer you to the last 18 months for the physiological arguments.

First of all, it's a very different scenario to say that the state of training affects fatigue rate than it is to say that distance specialization affects it. SO when you throw out a quote that suggests that the distances all fatigue at the same rate, what does that have to do with state of training?

The 2003 and 2005 papers are two studies of many, and I have years of experience in working with athletes in training research and I can assure you that the fatigue profile is altered by state of training when looked at in absolute terms (and this is very important, as I'll get to in a moment). And that is the point - not necessarily distance, but training status.

The fact that you use this equation in your training of athletes shows that performance alters as a result of training, and the reason performance alters is because you can improve an athlete's fatigue resistance through training. Bearing in mind that this is the question being answered (does Pistorius have equal fatigue resistance to able-bodied runners), state of training is vital.

You still have not picked up on the fact that the paper displays data selectively. If, as you say, the test is so accurate for distance runners and sprinters, then why have we been given the comparison between two distance runners and one sprinter, rather than the 4 sprinters they claim to have.

If, as you suggest, they have data on hundreds of sprinters, then why has this not been shown instead?

My point in all this is to raise these very serious and important questions relating to study design. You have turned it into a criticism of the 2003 and 2005 papers, which you obviously put a great deal of faith in, but which are not the focus of my dismissal of these findings.

Because what we have here is an untrained, unfit sprinter, who is being compared to distance runners with regards to fatigue profiles of potentially (note that this is not reported) fit and highly trained distance runners. That is, from a physiological point of view, a ridiculous comparison. And can I just emphasize that the main reason it's flawed is because of its impact on the aerobic assumptions drawn, and not necessarily on the fatigue tests.

The whole point behind that equation, as you have so eagerly pointed out, is that if you have 2 direct measurements, then you can predict high speed running performance. Therefore, if one changes independent of the other, then the exponential path between must also change. You make the case that because they're expressed relative to the top speed and top aerobic speed, the equation (Fig 3B) would not change. But what happens if you change only one point?

Ross Tucker and Jonathan Dugas said...

Then something that I've just realised now is that when you start expressing speeds and fatigue RELATIVE to the same individual, then you are heading into very murky waters when it comes to comparing different athletes. You can't do this when trying to compare one athlete to the rest of the population. This is actually a huge problem, and this discussion has thankfully contributed to my clearer realization of that.

A 400m race is basically an all-out effort of 45 seconds. Of course, it is paced, but the fact that athletes are slowing at the end means we should treat it as close to all out as possible.

Now, if an athlete can maintain a speed of 9m/s, they would finish a 400m race in 44.44 seconds. If they could improve this by even 0.1 m/s, then their 400m time drops to 43.9 seconds. I don't need to point out that this is a massive difference, and will probably be what separates gold from nothing in a world champs.

Why is this important? Because it shows you why a "flattening" of that curve has an impact on performance. If you anchor top speed, but improve the athlete's ability to sustain submaximal speeds, as happens through training, the result is that their sprint performance at anything less than top speed will improve in ABSOLUTE TERMS. I'm sure you know all this, given that you use the equations, probably for this very purpose.

But what I am trying to point out is that state of training affects the degree to which performance falls off, and it's completely irrelevant that it happens at the same RELATIVE rate. Let me ask it this way: Under what circumstances would be ever expect to see the relative rate of fatigue change? I believe it would change if one anchor point is shifted and not the other, as would happen if training status is altered. You don't, so what then is the answer? Never? Because if that's true, then you're working off a circular argument.

The problem with comparing individuals with this test is that I, a pedestrian jogger, would be shown to fatigue at the same rate as Kenenisa Bekele. You do realise the absolutely ridiculous implications of that, don't you? It suggests my fatigue tolerance is the same as the fittest or fastest men in the world. And sure, given their methods, I can be compared to anyone.

So what? You now know that my fatigue profile, regardless of state of training (according to you, I disagree) or distance specialization is the same. It means nothing for comparing me to another runner. Put another way, under what circumstances would this relationship ever change when you start expressing everything relative to each athlete's top speeds?

The whole point of this research was to answer whether Pistorius was similar to those he wanted to race against. They've shown that his rate of fatigue relative to himself is the same as two distance runners and one sprinter, despite the fact that they have data from hundreds of athletes. If that doesn't strike you as strange, then nothing will.

They have not explained the methods used on Pistorius either. How many sprint trials did he do? Was it 6, was it 15? That is equally important.

Ross Tucker and Jonathan Dugas said...

Just to finish up on the state of training, when Pistorius was tested for this study, the equation predicts that the speed he'd maintain for 50 seconds is 8m/s. This gives him a 400m time of 50 seconds.

If, as would be physiologically expected, his aerobic fitness improves by say 10% as a result of training, and he now reaches a speed of 5.5 m/s at VO2peak. His top speed can, for the purposes of this example, stay at 10.8m.s. His new predicted 400m time drops to 48 seconds. 2 seconds difference in performance, and a completely different athlete, who is suddenly no longer similar to the able-bodied sprinters with respects to VO2max, speed at VO2 peak, or ability to hold sub-maximal speeds, because his ability to sustain sub-maximal speeds has improved as a result of training - that is why state of training is important to the comparison being made in this paper.

Then just to answer the last point, Weyand's whole argument is that PIstorius' lower VO2 is because he has a "smaller engine". In other words, Weyand's defence of Pistorius, and the reason he is "Similar" to other sprinters is because at sprint speeds, he is at the same % of VO2max as any other runner.

Of course, this requires that the VO2max be believable and comparable between the athletes. It's not. PIstorius' state of training means his VO2 max is potentially up to 10% lower than it should be. If this were addressed, and the state of training were controlled, then Pistorius would be running at around 30% less oxygen cost than other sprinters, despite having an engine only a few % smaller. That would not suggest "similarity".

The difference in those speeds over 400m would account for a time difference of 1.6seconds, IF the athlete did run 400m at that speed. Of course I'm not suggesting that they run a 400m race at VO2 peak speeds. And yes, before I get shot down, I realise that the application of these speeds to the 400m race is flawed, but again, if you read it in the context of the post, what I am saying is that a difference of 0.1m/s is not "essentially the same". I go on to point out that it falls a massive 5 standard deviations outside the able-bodied runner values, and so is very different.

The beloved equation does predict a difference of 0.3 seconds, yes, and that's not being argued here. The point is the theoretical basis for the conclusion of the paper, which is that Pistorius is "similar" to able-bodied runners. He is not. In his untrained state, he is similar to distance runners who are presumably trained. And that's with regards to aerobic running capacity and ability to sustain sub-maximal speeds

Ken Jakalski said...

Hi Dr. Tucker!

Thank you for your informative responses. I appreciate your willingness to correspond and engage high school coaches and others like me who don’t have your scientific training.
On the specifics, your last responses explained a lot (to me and probably many others too), and I think our exchanges have clarified how the ASR equation can be used to predict performance, whether for Oscar Pistorius or other athletes. One of my biggest concerns was that the ASR formula was getting a ‘bad rap.’
As a result of our exchanges, I think we now agree that:
- Yes, the ASR expresses fatigue in RELATIVE terms (top speed and speed at VO2max).
- Oscar Pistorius’s 0.1 s difference in speed at VO2max vs. intact-limb sprinters would provide him with a performance benefit of less than 0.3 s over 400 m.
However, I think we may still differ on the training issue. You clearly believe state of training will alter the ASR curve. However, I just don’t see any evidence for this in the ASR studies (Bundle, Hoyt, Weyand 2003; Weyand & Bundle, 2005) or elsewhere. For example, if, as you suggest, state of training affects the ASR relationship, then this should show up in the form of poor performance predictions from the ASR equation due to the different states of training of the athletes tested. Yet, in those two papers, Weyand and Bundle tested 24 athletes of different specializations who completed more 300 all-out treadmill and track trials. The ASR predicted their performances to within less than 3% on average with no individual outliers.
If state of training does affect the ASR curve as you contend, how could all the performances of each of these 24 athletes be predicted with the same high degree of accuracy? Your assertion that training state affects the curve could only be true if all 24 of these research subjects were in the exact same condition at the time of the testing. Clearly, they weren’t as the authors made no attempt to control for the state of training. In fact, the authors specifically stated that they wanted individuals whose anaerobic and aerobic fitness levels differed by as much as possible. That was part of the design.
I also know from using the ASR to train hundreds of my own athletes that the equation remains equally accurate when my athletes gain or lose fitness. In fact, when my kids improve and start to outrun the ASR predictions in workouts, even my high school kids recognize that the reason is that their fitness (top speed and/or speed at VO2max) has changed. When we retest them and fix the inputs (top speed and speed at VO2max), the equation is dead on again in their new fitness state.
I’m afraid I am just not aware of any evidence to support your claim that state of training alters the ASR relationship. To the contrary, the evidence in the literature and elsewhere indicates just the opposite.

--Last post to follow...

Ken Jakalski said...

Last point, Dr. Tucker!

I do have one other comment to make to conclude our exchanges.
I think you may have unintentionally, but fundamentally mischaracterized the JAP paper as a defense for Pistorius when this was clearly not the objective of the paper.
You mentioned: “The whole point of this research was to answer whether Pistorius was similar to those he wanted to race against.”
Unless I’m reading it incorrectly, the JAP study was not designed to evaluate Pistorius vs. his competitors per se, or to determine whether Pistorius has a competitive sprint racing advantage. In fact, the authors very conspicuously offered no conclusion on the question of competitive advantage.
They stated their purpose: “Our general objective was to evaluate whether running with lower-limb prostheses vs. intact, biological limbs is functionally similar or not.”
At one point, you referred to “Weyand's defence of Pistorius.”
I didn’t interpret the objective of the paper as Weyand or any of the other authors defending Pistorius. Furthermore, the SMU press release indicates a lot of the data in the paper were not part of Pistorius’ defense at the CAS hearing. That’s another reason why I disagreed with you about the study offering nothing new.
Additionally, I don’t see how the authors of this paper would agree to design it as a defense of Pistorius when Dr. Weyand himself indicated that the authors don’t agree on whether Pistorius has a competitive advantage or not (see the recent LA Times article).
If you believe that study is a poor piece of research, you’re certainly entitled to your opinion. I simply feel that any criticisms should be accurate in terms of the data and study design and that the motives of scientists and others are correctly presented.
Either way, thanks again for the stimulating exchanges!
I look forward to more of your creative and entertaining blogging in the future. I may not agree with all your insights, suggestions, or interpretations, but I believe it was Will Rogers who once said: "A difference of opinion is what makes horse racing--and missionaries."

Ken Jakalski
Lisle High School
Lisle, IL USA

maryka said...

In response to the first 2 comments in this post, "extreme measures" are already (kind of) happening... UK track cyclist Ed Clancy last year said he'd be willing to have his collarbones broken in the name of science and training, if you believe the press anyway, http://www.independent.co.uk/sport/olympics/cyclists-take-secret-of-suits-to-shredder-1547673.html

Anonymous said...

I think you guys have completely missed the point and may find yourselves on the wrong side of history!

I like your posts, but the accuracy of the OP science is not the issue even though the lack of good science in this case may disturb you as scientists. Even if OP has an advantage, the real question is whether it is in violation of rules e.g. USTAF Rule 143 .."Such shoes, however, must not be constructed so as to give the competitor any unfair additional assistance, including the incorporation of any technology which will give the
wearer any unfair advantage, such as a spring or similar device."
The key issue becomes the definition of UNFAIR advantage. Scientific evidence that he runs differently is NOT indication of unfair advantage. Bill Bowerman built lighter shoes to reduce the energy required in lifting the foot off the ground hundreds of time during a race. The purpose was clearly to give his runner's an advantage in a race. Was that unfair? Illegal? I think not. There was no outside force or influence helping to push the foot up each time. He simply minimized the effort required of a natural motion.
OP's Cheetah and Spira and perhaps others focus on transferring the potential energy stored up in the shoe after the body naturally falls back to the ground into greater horizontal forces. How is that any different? How is it unfair? It is clever to use the forces of natural motion to work for the runner rather than dissipate the forces or let them work against the runner. There is no outside force or locomotion involved. Perfectly fair!
Obviously my definition of unfair begins at the point where outside, unnatural forces are brought into play. Yours may be different, but if we measure "unfair" only at the point where performance changes then a lighter shoe is also unfair...heck...even a shoe versus bare-feet could be considered unfair. It is a slippery slope and, in contrast to your post, any attempts to label one innovation/advantage unfair versus any other is the real moment when you risk opening Pandora's box. I, on the other hand, welcome OP and his blades! I believe they could lead to a variation of the idea to produce a shoe useful for able bodied runners. I suspect that will ultimately come from Spira or somebody other than the Nike machine and that could be good for the sport and for runners everywhere.


Anonymous said...

Per the request of several, I will re-post some of the errors Ross Tucker made in his posts on the Pistorius case. This list was posted previously and removed from the site.

The list is by no means comprehensive; these are only some of the errors on one subject (Pistorius) and limited to what would fit in a single post. My original awareness of the problem arose when Coach Jakalaski pointed out problems with Tucker’s posts. I had followed the Pistorius case and looked into what Ross Tucker had written and found a clear and consistent pattern of falsehoods. The same careless relationship with the truth is present on other issues as well. The examples that follow are simply from one topic per your request and in response to Ross Tucker’s claim that he is not censoring.

This list was posted previously in specific response to Ross Tucker stating he stood 100% behind all of his posts on this Pistorius and his specific challenge to me to find any false statements. When I provided the list previously as he requested, it was promptly removed from the site.

Per my earlier note, I agree with Tucker on Semenya, Bolt, and Pistorius and many other controversies.

My issue is with opinions that are supported by false information on a site that is labeled “science” and put forth by a person promoting scientific credentials to legitimize the site.

The list of false statements will appear in the next post.


Jonathon Howard

Ross Tucker and Jonathan Dugas said...

Thank you for posting this here. Just to say, you've written the following:

"The list is by no means comprehensive; these are only some of the errors on one subject (Pistorius) and limited to what would fit in a single post."

And you somehow can't understand why I don't find this personal and offensive? Seriously...

It's disgraceful, playing the man, not the ball.

Anonymous said...


INCORRECT STATEMENT #1: With respect to the data collected in the US on Pistorius, Tucker wrote: “I do not believe that this research will be forthcoming”

The data collected in the US were published in the Journal of Applied Physiology online in June of 2009 and later in hard copy.

INCORRECT STATEMENT #2: Tucker’s insinuation of financial motive for his incorrect prediction that the data would not be published: “now that the dollar signs are in place and the incentive has been achieved by those who have much to gain from today's decision”

The author’s acknowledgment section indicates verbatim: “All the authors provided their time pro bono in return for the right to publish the results.”

INCORRECT STATEMENT #3: “In other words, something changes differently as Pistorius speeds up……… either his use of oxygen remains the same, or it increases only very slightly, or it goes down”

Figure 2 shows parallel lines for the rate of increase in the rate of oxygen uptake with increasing speed (the slope of the line) for Pistorius and the intact-limb 400 m runners in the paper. Clearly those slopes are almost exactly the same and would not be statistically different. Therefore, Tucker’s statement is false.

INCORRECT STATEMENT #4: “In the end, he was declared physiologically similar to elite and sub-elite distance runners, despite having a 17% efficiency advantage.”

Pistorius was NOT shown to be 17% more economical than elite and sub-elite distance runners. Rather, the paper he was 17% more economical than intact-limb sprint runners. This is stated throughout the paper.

Regarding the 17% the difference vs. sprinters, the authors concluded the opposite of what Tucker claimed above – i.e. that Pistorius was dissimilar vs. intact-limb sprinters.

INCORRECT STATEMENT #4: “What Pistorius does is not running”

All the biomechanics books and papers I’ve ever read indicate that running by definition involves the body speeding up and going into the air at the same time. Figure 1 in the JAP paper shows Pistorius and the intact-limb runners both do exactly that.

I don’t know what personal definition of running Ross Tucker may have, but if he concludes Pistorius is not running, his personal definition does not agree with the standard scientific definition all the biomechanics people seem to use.

INCORRECT STATEMENT #5: “Second, the speeds were not reported”

The speed of every trial is presented on the graph (in fact, they are provided 3 times, once on each panel) in Figure 3.

INCORRECT STATEMENT #6: “the difference of 0.1 m/s translates into about 2 seconds in a 400m race at the speeds reported”

No, it does not. The equation one in the paper can inform anyone with basic math skills, the 0.1 m/s difference in velocity at VO2max translates into less than 0.3 seconds in a 400 m race for the athletes in question. The coach who posted pointed this out to Ross Tucker.

INCORRECT STATEMENT # 7: Referring to the fatigue tests in the JAP paper: “the 400m sprinter shows similar fatigue characteristics to elite distance athletes...extra-ordinary. You'll be aware of course, that distance runners SHOULD show better fatigue resistance, because that's what their events rely on. “

Incorrect again - when standardized using the technique in the paper, distance runners and sprinters have the exact same fatigue curves.

The coach also explained this to Ross Tucker and was kind enough to point him to the scientific papers he either did not read or did not understand prior to publicly and incorrectly criticizing this result.

INCORRECT STATEMENT # 8: “Well, less work is done on the centre of mass, and his viewpoint, one which I agree with, is that vertical force generation is particularly important during acceleration”

It is mechanically impossible for vertical force to contribute to horizontal acceleration - this is high-school level physics.


you decide.