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.

Sunday, July 28, 2013

Alan Oliveira runs 10.57s. Leg length or something else? Over to the IPC and IAAF

Alan Oliveira runs 10.57s: Is it leg length, or something else? Over to the IPC/IAAF

Hot on the heels of a 20.66s World Record for double-amputees in Lyon a week ago, Alan Oliveira of Brazil, the fastest double amputee in the world, today destroyed his own 100m World Record with a performance of 10.57s in the London Olympic stadium.

I wrote about his emergence as the heir to Oscar Pistorius last week, describing the implications of his incredible improvements in 2013.  He last week won the 100m, 200m and 400m titles in the IPC World Championships, and is now a staggering 0.44s faster than the next fastest in history at 100m (Pistorius).  The improvement has come within the last two months, because prior to that, Oliveira's best 100m time was 11.33s.

The leg length - in play, but a red herring

Of course, the current debate is all about his legs, and more specifically, their length.  That is a red herring.  While partly true, there are many reasons to suggest that what Oliveira has achieved in 2013 is not the result of excessively long legs, but some other factor, which has already been proven to exist by scientific research.  Unfortunately, the IPC seem intent on pursuing length as the critical one, with new rules controlling length to be announced soon.

It's unclear what this will mean for Oliveira in the short term, but the problem is that they won't solve the larger problem, and the next athlete to come along with once again push the sport into the same dilemma.

Let's look at the leg length issue in a bit more detail.

London 2012 flashback 

When he defeated Pistorius in the London 2012 200m final last year, the accusation made by Pistorius was that he "couldn't compete with Alan's (long) stride length".  An easy explanation to test, because all it took was counting the strides, and it turned out that Oliveira's stride was not all that long.  In fact, Pistorius took fewer strides than Oliveira, and thus had the longer stride - 92 steps vs 98 steps, for a step length of 2.2 m vs 2.0 m for Pistorius and Oliveira, respectively (remember that a stride is two steps - I counted steps, but report strides later in the discussion).  And the final 100m showed the same pattern - Pistorius' average step length was 2.3 m, compared to 2.2 m for Oliveira.

So, stride length, at least at a superficial level, is not where the advantage came from back then, and it's not the sole explanation now either.

That said, it would be incomplete and false to suggest that Oliveira's leg length should not be the subject of some scrutiny.  In the week leading up to that 200m final, Oliveira revealed in an interview that he had recently increased his blade length by 4cm, taking him from a racing height of 1.77m to 1.81m, and he was clearly relatively taller than his rivals.

To understand what all that means, let's consider how the IPC set the maximum allowable leg length for double amputees.  First of all, it's not an easy task to do - there is no such thing as a "normal height", and when someone does not have legs, then trying to be specific about how tall they would have been is a complex exercise in dealing with ranges.  That's because we don't all share the same limb proportions.

There is an average ratio of say, arms to height, and a similarly average ratio of femur length to total leg length, but these averages don't often apply to elite athletes.  One example is Michael Phelps, the world's greatest swimmer, who stands 1.93m tall and remarkably, wears the same length pants as Hicham el Guerrouj, the world record holder in the mile, who stands only 1.75m tall!

That is, a difference of 18cm in height, with the same leg length.  Such are the variations between people.  One is a swimmer, one is a runner, and they are arguably born to excel in their specific events by virtue of completely different leg to total height ratios.  For pages and pages of similarly mind-blowing stats, I would highly recommended David Epstein's book, "The Sports Gene", which is due out this week.

But for now, let's leave it at the fact that the IPC cannot simply say "You should be X cm tall based on your arm length".

Instead, what they have done is establish a maximum allowable height for each double amputee.  The image below, which was released in the aftermath of the London 2012 controversy, shows the height limits for the key players in this debate.  It invites four thoughts:

Thought # 1 - taller is not necessarily better, and that has important implications

First, notice that Oliveira is allowed race at 185.4 cm, whereas Pistorius was able to go to 193.5 cm.  Presumably, that's a limit based on arm length, modified and improved by femur length to give a total height, which is the absolute maximum for someone who has exceptionally long legs relative to their body (like el Guerrouj).  We know that in London, Oliveira was 181cm, so he was 4cm short of the limit.  He was thus perfectly legal, as I'm sure he is now.  The issue is thus not cheating, but perhaps whether the limits are 'fair'.  

To address that, it's interesting to wonder about why he would stop at 181 cm?  If you're going up from 177 cm as he did, and if longer legs mean better performances (as people somewhat simply suggest), then go to the limit of 185.4 cm. More length, more speed?

The answer to that question is that at some point, going longer becomes counter-productive.  That's because the start becomes so severely compromised (as we saw with Oliveira in London), as well as balance around the bend, that the overall performance gets slower.  Top end speed may be greater, but the net result of longer limbs is less balance and therefore slower times.  Thus, there exists a "sweet spot", an optimal length for each athlete, and that's why none of the top double amputees are competing at their maximum allowable height.  Pistorius, for instance, races at 186 cm.

Thought # 2 - athletes discover the sweet spot by testing, so everyone is "optimized"

The implication of Thought # 1 is that the elite athlete have discovered their optimal sweet spot, because if they didn't, they go maximum length to find more top end speed.  We know from the PR around Pistorius that testing on blades is extensive - he traveled to Iceland often, and representatives of these carbon fiber manufacturers visit athletes for field testing regularly.

Part of the process is discovering how far below the limit the athlete should stop.  So for instance, we should be asking how Oliveira knew to stop at 181 cm prior to London, and why Pistorius was at 186 cm in the first place when he could have gone to 193 cm?  Why not 189 cm?  They had room to play with, but decided not to use it.  The answer is that they're optimized at those 'sub-max' heights.

For Oliveira, however, that may have changed since 2012.  It's conceivable that since his breakthrough (remember that he was just 19 in London last year) he has had more time and more technological support, and thus more opportunity to work out his ideal racing height.  

One source at the IPC reported to me earlier this year that Oliveira had in fact gone shorter, and thus discovered a much faster start and bend performance, driving his times down.  Others are saying he is now even longer - perhaps right up to his 185.4 cm limit.  A curve ball in this debate is that between 19 and 20, he may have grown, and so his upper limit of 184.5 cm from London may have increased, allow him more to play with.  

His 100m performance improvements suggest shorter, because his start is so much better, unless he has improved his balance and co-ordination spectacularly in the last 12 months.  However, we don't know what height he has raced at.  The IPC will, unless they have been grossly negligent in getting the blades measured, and that is surely inconceivable given the obvious focus on them.  They'll have to consider this information as they decide what to do next.  Which leads me on to point 3...

Thought # 3 - the IPC have to change the rule, but how effective will it be, and what do they base the change on?

It's clear that if the current progression continues, the IPC and the IAAF will have to reassess the situation of double-amputees racing in the able-bodied events.  Oliveira won the 400m title at the World Champs this weekend, but with a less than stellar time.  He said after that he doesn't train for the longer distance.  With an Olympic Games coming up in his own country, and with 3 years of preparation, maturity and strength to gain, it's almost inconceivable that he won't at least attempt to run in both Olympic and Paralympic competitions, emulating Pistorius.  Perhaps he will focus on the 200m - another half a second improvement on his 20.66s WR puts him into a final there, with possibilities of a medal should three years produce similar improvements to 2013.  Given that he is only 20, and clearly still in a period of rapid improvement, such an improvement is well within the realms of possibility. 

If he does jump up to the 400m, his chances are even better - the time lost at the start can be recovered over the final 350m, and his top-end speed, which must surely be comparable to Usain Bolt's, as well as remarkable sustained speed in the second half, should see him go considerably faster still.

So, what are the IPC and IAAF to do?  Refer again to the table above, and remember that those maximum height allowances are based on data collected from hundreds if not thousands of people to establish a range of human "norms".  If the IAAF and IPC decide to change them to make Oliveira race at a shorter height, they would have to justify it by saying something along the lines of "We are now adopting mean or average height rather than allowing for extreme individuals within the normal population".  That is, they would have to pretend that extremes like Phelps or el Guerrouj don't exist, and I can't see how that is legally or scientifically defendable.  You have to allow for cases at the extreme end of "normal", which is why the answer to that apparently simple question "What is a normal height?" is so very complex.

Alternatively, they could modify the guidelines slightly, perhaps to 1SD above means, and reduce Alan Oliveira's maximum allowable height subtly.  If it meant he was forced to drop to say, 181cm, he'd be at the same height he was at in London, and that means more of the same debate and performances. 

The point is this:  Because none of the athletes are at the maximum allowable height (see Thought #2), any change in the policy will have to be drastic, or it won't affect them anyway.  And drastic changes mean re-writing the understanding of human anthropometry, possibly discriminating against individuals who are normal but 'extreme', and may thus be impossible to implement.  All in all, very sticky for the IPC.

Thought 4: Leg length possibilities for Oliveira

And then finally, as I return to where I began, this discussion of length may be something of a red herring.  Again, there is no doubt that as the legs get longer relative to total height, the person is more likely to be a successful runner.

However, Oliveira has achieved almost a second of improvement in the 100m within one year.  His 200m performance trajectory is similar.  That invites three possibilities:

  1. He is still racing on the same length blades as London (height 181cm).  In this case, his improvement is solely due to training, co-ordination and normal development.  One can still say he has an advantage, but his improvement is distinct from it.

  2. He has changed up, and gone to longer legs, then I find it hard to believe that 3 to 4cm (he only has this to play with, it's not as though he can race at 195cm - see table above) can contribute to that kind of performance.  This is particularly true given that any increased length must surely compromise the start and bend, and so the effect is even larger.  Simply put, it cannot be solely due to running "taller" in 2013

  3. He has changed down, and found a better "sweetspot" that gives him a better start and bend performance, faster overall, with some accepted reduction in top speed.  If this is the case, and he's running at say 179cm, then it's even more of a problem for the IPC and IAAF because whatever they plan to change in their guidelines would need to be even more drastic.
What if stride length is not the factor at all?

But what if it is not in the length at all?  What if that simple exercise of counting his strides, and comparing them to Oscar Pistorius' in London 2012 actually hints at the solution?

Remember, that night, Pistorius took 49 steps on the bend and 43 steps in the home straight.  That is a step length of 2.0 and 2.3 m respectively.  Oliveira, on the other hand, had average step lengths of 1.92 m on the bend and 2.2 m on the straight.  In a race of absolute stride lengths, Oliveira is second-best.

The key, however, is the stride length relative to body height - someone who has excessively and disproportionately long legs will have a longer stride relative to their height, so you can partially test this 'accusation' by comparing stride length to total length.

So, running that logic for 2012, if Pistorius, at 186 cm, takes 230 cm steps, his step length to height ratio is 1.24.  Oliveira, at 181 cm with 220 cm steps, is at 1.22, and so in fact, Pistorius' steps are actually longer, not only in absolute terms, but also relative to his height.  This is the primary reason that I wasn't convinced that Oliveira's advantage was stride length back in 2012, and I'm not convinced now (though I will allow for the possibility that he has since increased his length - see Thought # 4).

The answer is more likely stride speed, not length. And that closes the loop

The ratio is however significant, because it says that Oliveira's advantage, which is now even greater than it was in London 2012, comes not from stride length, but the other important factor - stride speed.  It is the turnover of his limbs that separates Oliveira from the rest of the world, and which has made him a realistic medal chance in able-bodied competitions.

And why is that important?  Well, it closes the loop, bringing us full circle, because the scientific research on Oscar Pistorius showed that the advantage of double-amputee athletes, as a result of super-lightweight carbon fiber blades, is that his limb repositioning speed was "off the biological charts".  Those words were written by Prof Peter Weyand, who tested Pistorius and suggested a 10-12 seconds advantage because of limb reposition times he had never seen before, even in 100m Olympic champions.

Simply, the double-amputee was able to move his limbs so fast that he could then afford to spend more time on the ground, and generate significantly lower forces than able-bodied runners who were going the same speed as him.  The "athletic limit" to sprinting, according to Weyand, a world leader in sprint mechanics, is the ability to apply huge force to the ground.  Pistorius was able to run world class speeds without that limit existing, because his ultra-lightweight limbs allowed him be break another limit - the speed with which the legs could be moved.

Alan Oliveira has taken that to a new level.  When a man is running a 200m race and his strides are about 10% shorter than his rival's, then the only way to run faster than the rival is to have stride speeds that are 10% or more faster.  That's the Oliveira advantage - extra-ordinary speed of leg movement.  He is able to capitalize on the technology more effectively than any runner before him, and may also be able to generate force more rapidly than his predecessors.  The result is less time on the ground, less time in the air to reposition the limbs, and 20.66s and 10.57s performances.

Oliveria - validating the theory, vindicating the research, with no end in sight

Oliveira is, simply put, the validation of scientific theory, and he vindicates the predictions made about what would happen when the pool of athletes with access to carbon fiber blades expanded to include superior athletes.  This was inevitable - it's a rapidly growing sports category, and this is a great thing.  If only they'd kept them separate based on objective evidence, rather than the emotion of the Pistorius case.

Oliveira will one day be beaten by the next generation of double amputee, who will be even faster, and will then re-ignite the same debate.  The problem for the IPC and IAAF now is that they will have to reassess their guidelines in order to slow the runner down.  In other words, Oliveira is too fast, so we have to rewrite the rules.

Effectively, what they would be doing is setting a bar, at say 20.50 s for a 200m and 45 s for a 400m, and saying 'We welcome your participation, but just don't be too fast, or we'll have to change our rules to make you slower'.  It is analogous to putting weights on the bicycles of the top men of the Tour de France, to make the race more competitive, or make Djokovic and Murray play with wooden rackets to slow their dominance of tennis.

Well, the end is not in sight, because just as Pistorius was not going to be the pinnacle of athleticism on prosthetics, why should Oliveira be?  This is progress.  It's human progress.  A normal progression of ability as better athletes emerge.  The IPC and IAAF are looking at the technology, when they should be looking at how the heck they managed to duff the case against carbon fiber blades in the first place. 

Final word - a lot of the above discussion revolves around the very basic analysis of the London 2012 200m final, where I counted the strides for Pistorius and Oliveira.  What should happen is a similar discussion, in even more detail, now that Oliveira is rewriting the record books.

However, that won't happen, because the powers that be don't seem to recognize the importance of gathering the data to inform this kind of discussion.  I can appreciate that they have bigger issues, and may not have the resources to do it themselves.  Certainly, they are custodians over more than just one category and three of its events.  

However, in the build up to the Lyon IPC World Championships, I tried to approach the IPC for permission to analyse Oliveira's 100m and 200m races.  I wanted split times at 10m intervals, so that we could discover just how much time he lost at the start, when he hit top speed and how that top speed compared to Usain Bolt's (I suspect it is the same, or faster).  However, the IPC were not as enthusiastic, and so the study concept was never approved.

What a pity, because now we have to guess - we don't know his leg length, or the limit, or just how he put those world records together.  In time, maybe this debate will force those facts into the open, and that will be a good thing.  But until the evidence emerges, we talk about average stride lengths and stride speeds, which suggest that it's not solely about longer legs (as the public and even some rivals are still proclaiming), but about stride speed.

The evidence is out there, waiting to be found.  For Pistorius, the evidence was found - it showed clearly what was happening.  Why is it paid no attention?

In time, perhaps it will be.  Today, in 10.57s, Oliveira guaranteed that it would be.  There's much still to learn.


Friday, July 26, 2013

On performance analysis: Common sense, guided

Performance analysis in sport: The guidance of common sense

With the Tour de France now disappearing in the rear-view mirror, I've been weighing up a post on the value of performance analysis as a predictive tool in sport, particularly given the criticism of our (Doc at the Veloclinic, Ammatti, Fred Grappe and Antoine Vayer) recent analysis of the Tour de France.

Throughout the process, I encouraged the use of insight and circumspection when looking at performance metrics, but however strongly the message was emphasized that performance does not constitute proof of doping, there is always a convenient and 'lazy' way to dismiss it as 'pseudoscience' (which is the new "never failed a test" defence, incidentally).  That particular stick (pseudoscience) has been wielded in the context of Pistorius, Armstrong, hydration, barefoot running, running technique and fatigue, but not previously with the extremism seen during the Tour.

So here are some thoughts on the method, and an attempt to create some context around the value of trying to understand the world with imperfect methods (which we all admit they are).

Acceptable uncertainty - performance analysis is never exact

In my work with the SA Sevens Rugby team, we analyse performances.  We analyse opposition patterns, we study their options and tendencies in various phases of the game.  The purpose is to better understand, and thus predict, what they are likely to do.  We can tell our players to expect Samoa to play a certain pattern, whereas Fiji will do the opposite.  The players run onto the field knowing with a reasonable degree of certainty where a lineout throw will go, how the opponent will defend rucks, what they'll attempt on kick-offs and how they are likely to run at us from broken play.

This is the same concept applied to sports the world over.  In the NFL, it exists at perhaps its highest level, where expert analysts break down seemingly random patterns and discover 'tells' and methods to pre-empt opposition plays.

However, everyone involved recognizes that it is not formulaic.  There is uncertainty, and this is accepted.  Performance analysis, regardless of the sport, is always an exercise in probability because it happens in uncontrollable conditions, and so it adds value by adding insight rather than by conclusively and accurately predicting what will happen.  I can't give you the error bars on this kind of analysis, because sport is fluid and contextual, and so even the 'safest' bets are vulnerable to unique and specific situations.  The smaller the data set, the greater the error, but it's hard to assign a value to it.

The result is that a player cannot run onto the field with a text-book in their mind and then fail to use common sense, as well as all their other senses, to assess a given situation.  Playing off memory, stats and data is a disaster, when you have eyes, ears and insight.  Just because the pre-match analysis said that the opponent would do X does not mean options Y and Z are off the table, and so a 'smart' player is needed to discern the actual event from the performance analysis predictions.  On that note, we've had players who can't seem to grasp this, and who take to the field with only one option in their minds.  They prove to be inflexible and are probably better off with less information.  For most, however, the guidance is beneficial, if interpreted sensibly.

If I tell Novak Djokovic that Andy Murray is likely to serve down the T on points where he is leading but out wide when behind on the scoreboard (for instance, this may happen on 75% of points, with an error), Djokovic would be foolish to leap wide during the ball-toss, but he'd also be foolish to discard the information because there's "uncertainty".  Inch, don't leap.

In cycling, there seems to have been far tooo much "leaping", in the sense that people seem to have either blindly embraced or discarded the concept of performance analysis, whether it be for the metric time up a mountain, estimated power output, or the physiological implications of that performance, without recognizing the necessary nuance.

Keep the other senses

So performance analysis in rugby, football, tennis, American football and basketball may be quite different from performance analysis is cycling and running in many respects, but it is similar in one important aspect - it does not replace common sense or give permission to disengage every other sense in order to rigidly accept a black and white version of the world of sport, which is clearly nuanced and everyone recognizes this.  For this reason, it never constitutes proof.

Why people would want to accept such extremism is beyond me.  That inflexibility fosters blindness and prevents insight.  In effect, people who dismiss performance metrics and their implications as "worthless" because we are estimating power output are analogous to a blind man, offered 40% vision, but who refuses it because he only wants 100% sight.  It's 20/20 or nothing, and I believe that's a flawed and narrow understanding of the world.  It would be the same as a head coach saying to his analysts "If you can't guarantee with 100% certainty what my opposition are going to do, I don't want to know it at all".

Before I'm seen to be proclaiming that say, 40% is "good enough", I will say once again that we all recognize that it's not.   We want 90%, 100%.  That's why the process began with a call for the data, biological and performance.  That's because the "blurred" image offered by 40% vision, similar to the "performance pixellation" I wrote about after many stages in the 2013 Tour de France, may well lead a person into many blind alleys and unseen obstacles.

However, what seems overlooked is that people still have other senses - they have small, hearing and touch.  And they should also have some common sense.  So 40% vision added to other senses makes anyone better off than they were in total blindness.  Unless, of course, said blind man decides that with his new-found 40%, he is going to ignore every other sense.  This would be equally foolish in the opposite direction.

So if we can gain any insight at all, and combine it with our other senses, then just like our SA 7s rugby players, or the NFL footballers or a basketball player, we can run onto the field or court with more confidence, provided we retain the ability to interpret every situation as it develops for what it is.

As applied to the 2013 Tour de France

Therefore, when Chris Froome rides away from a field on the first week of the Tour at a power output that is higher than benchmarked, and produces a time that puts him in the company of known dopers, we should ask questions of that performance.  But we cannot conclusively use it to prove that he is doping.  That would be extremism, and it would be wrong.  It's for this reason that I have written, and still believe, that Vayer is too far to the extreme when he declares performances 'mutant'.  They are not - they're still within the realms of physiological plausibility, though on the high side.  That's what Fred Grappe concluded when provided access to Froome's data, and it is the conservative and correct approach.

Similarly, Rodriguez or Quintana should be regarded with some 'wonder' for getting progressively better, and eventually exceeding historical benchmarks during the race.  Quintana, incidentally, produced the best performance of the entire Tour on its very final climb of Semnoz, benchmarked against historical norms using the pVAM method (pVAM, while I'm discussing it, is just that - a benchmarking method that for the first time, makes Semnoz comparable to Alp d'Huez even though it had never been done in the Tour.  And that's progress - for all the criticism of the pVAM method, it opens that possibility.  Now it needs evolution)

A process, and evolution with uncertainty

Without the analysis of pVAM and the estimates of power output, plotting those power outputs against duration, and historical reference points, these performances have no context.  Therefore, performance analysis asks questions, it does not answer them.  In time, those answers may emerge, and performance analysis can help us to evaluate those answers as realistic or false.  But with an eyes tightly shut approach, we are guessing.  For instance, is it normal to show higher power outputs in week 3?  We don't know.  Had we gathered data for 20 years, we'd have a pretty good idea.

The point is, this is all a process.  It's never going to be perfect, but if we halt evolution based on imperfect then we'll never move forward.  If Henry Ford and others had decided not to proceed with mass produced cars because he had in mind the perfect luxury vehicle, then we'd still be sitting on horse-drawn carriages.  I think that Doc, Ammatti and even Vayer (despite some differences in the interpretation) are doing cycling a favour by calling for transparency and starting a discussion.  I honestly believe that performance analysis is making progress in cycling as a result of their efforts.  Again, it's not perfect.  There is uncertainty. But then, progress dies of boredom when there is certainty.

It doesn't deserve outright dismissal, and it doesn't warrant embracing as conclusive proof of anything (nor does it ever ask to be seen this way).  So I'd thank all those for participating in the discussion.  I hope it advances insight and enjoyment of the sport (it certainly does for me).  I applaud people for wanting accuracy, I think that is always good.  And if I have ever drawn a conclusion that goes beyond what the error of the estimates does not allow, I'd expect to be called out on it.  Run me out of town i I say that a performance is proof of doping without recognizing its context or explaining that kind of extreme statement.

But equally, I'd hope that people read the articles (yeah, I know, they're long) and then consider my interpretation of the numbers, and the explanation, the nuances, and then avoid the extremism reaction.  We all have other senses, after all, so we can accept uncertainty and navigate with only partial vision, provided we engage those senses, most of all common sense.

Two final parting thoughts

That said, two final thoughts on the 2013 Tour, from a performance analysis perspective:

1.  These are the SRM data for five Tour performances, compared to our estimates using two methods.  One is the method of Ferrari, using pVAM, the other is the CPL method, which pretty closely approximates Vayer's method.  Once again, they're from Ammatti Pyoraily's vast bank of performances:

Perfect? No.  But not nearly as unusable as some have suggested.  You can decide for yourself whether those estimates are worthless or not.  I'd point out that the estimations are just as likely to underestimate power than overestimate it (3 vs 2 for CPL), so sometimes estimation gives benefits to the cyclist.  Collectively, if we wish to avoid performance pixellation, the average error for these five performances is 0.6%.

The timing of climbs was criticized because it's based on TV observations.  An error of a few seconds, far smaller than any of the other errors, so it's a bizarre criticism that emerged in the Tour.  If we were hand-timing Usain Bolt in a 10-sec sprint, it'd be different, but not over 30 minutes or more.  Wind?  Of course.  When you're going in a straight line with tailwind for 90% of the race route, like Boston Marathon runners experienced in 2010, wind blows the whole thing out the water.  I dare say it's a lot less significant, based on SRM data, in the variable Tour geography.

Is it acceptable to be within 1-3% per climb?  That depends what one wishes to conclude.  If the pursuit is proof of doping, then of course it isn't, but even exact SRM data wouldn't "prove" anything anyway.  We can always do better - you'll notice that Ferrari's method tends to overestimate power, and that's because for lighter riders and faster speeds, the formula he developed doesn't account fully for drag.  That's why Doc at the Veloclinic has been trying to develop a better method for performance prediction, taking into account the components - all part of the process of getting better, and that's the evolution that will make 2014 better than 2013.

But I think on the whole (and again, the above table shows only five 2013 comparisons, but we've pooled this with other data from Sorensen and the 2010 Tour de France where Horner provided his SRM data) the error is consistently in that 1 - 3% range, and I'd argue that the method is not as poor as some have suggested.

More to the point, the exact data wouldn't allow conclusive proof of doping or non-doping anyway, so given that the interpretation is subject to nuance and context, the error in the estimate can be accepted at this level.  It's an acceptable level of uncertainty.  That's the performance uncertainty aspect I wrote of earlier - we may not have 100% vision, but we have senses.  We just need to use them fully as we add to them, uncertainty and all.

2.  This whole performance analysis concept became heated and controversial because it seems that many in the English speaking world had a vested interest in the success of specific riders.  Much like the most aggressive defence during the Lance Armstrong era was coming from the USA, some of the most vocal defense of Froome was, understandably, coming from the UK.  And even British media.  There is nothing inherently wrong with this.

By the end of the Tour, Quintana and Rodriguez were at the same level as Froome on Ax-3-Domaines, and Quintana even surpassed it on Semnoz to produce the best climb of the race, statistically speaking.  Even accepting the constraints of "performance pixellation", the trend was quite clear.

And in some parallel universe, hypothetically speaking, I wonder how the numbers and estimates would have been perceived if Quintana and Rodriguez started the Tour in that condition?  What if Contador had joined them at speeds as fast as Armstrong and Ullrich a decade before?  I dare say that had it been Quintana and Rodriguez accelerating clear on Ventoux and Ax-3-Domaines, the whole process of our performance analysis would have found a different story from the very first mountain.  A new script, certainly one in which we'd be doing less defensive explanation than we've done.  Would those performances have attracted a similar skepticism from the UK press as Froome did from the French media (by all accounts)?  Certainly, and that's the key - the performance warrants questions, not accusation, and measuring it, imperfection and all, can only be helpful.

Ultimately, though, the final word on the 2013 Tour is that we shouldn't be accusing, just wondering.  And given cycling's history, and the fact that those entrusted with running the sport have shown themselves to be unable to clean it up, we cannot simply believe (blindly) in miracles this time around.  So we wonder, reasonably, and use some kind of performance metric to gain some insight.  Proof, no.  But equally, not worthless.

On that note, crank punk nails the problem facing cycling's appeals for trust.  There've been scandals in the past, and as sure as anything, there will be doping scandals in the future.  Paul Kimmage also recently gave this interview - he's over on the cynical extreme, but his thoughts are eloquent and with some caution, well worth listening to. Jump to the 10:00 mark.

On that note, thanks for reading the Tour coverage, let's do it again in a year.  Next up, the IAAF World Championships in Moscow.


Monday, July 22, 2013

Oliveira 20.66s: A double amputee will soon medal in able-bodied Olympics

Double amputee sprinters version 2.0: Alan Oliveria runs 20.66s 200m 

The next generation of double-amputee sprinters have arrived, sooner than expected. Alan Oliveira of Brazil today won the 200m world title in 20.66s, improving the previous world record for double-amputees by 0.64s.

Oliveira, who shot to prominence after defeating Pistorius amidst blade controversies at the London's Paralympics, added the 200m WR to his 100m WR set a month ago (10.77s, 0.16s faster than previous).

So, we now see the successor to Pistorius, running 0.14s and 0.64s faster than Pistorius over 100m and 200m, respectively.  How much faster will he be over 400m?  The answer is "substantially".  Even if he does not continue increasing that advantage over the previous benchmark as distance increases, he will run ± 44.3 s.  Chances are, it grows even larger.  If you saw the race today, and the final 100m, you'll appreciate why.

So, a prediction, and you heard it here first (actually, if you've been following this debate about carbon fibre blades, you'll have heard it here years ago):

If he desires to run against able-bodied athletes, Alan Oliveira will win a medal in the 400m at the 2016 Olympic Games.  He is still only 20, with much strength to gain, but his recent improvements are staggering - 0.56s in the 100m and 0.80s in the 200m since the London Paralympics.  That suggests much more to come, and it suggests a medal in the able-bodied Olympics in 2016.  Of course, he may not wish to, which would be interesting.  If it is not him, it may be the next athlete, but it will happen.

This is not a criticism of Alan Oliveira, who deserves credit for his performances in the Paralympic events.  He will become the best amputee sprinter in history, if he is not already.  This is however an issue of the cross-over of amputee and able-bodied competitions, and the implications of this kind of progress for that decision.

The decision to allow Pistorius his wish to compete in able-bodied events was always going to have predictable repercussions in the future.  These were unfortunately obscured by marketing, emotion and the incomplete (dishonest?) presentation of science.  Allied to this was the media's almost total inability and lack of will to challenge the PR campaigns and to ask the difficult questions while they fell over themselves to tell the heart-warming, popular story.  (I hope that this section of the media will tell the same story for Oliveira or his successor when they run sub-43 seconds for 400m.)

The repercussions come more clearly into view today, because the evolution of a young sport means that better athletes will emerge, and times will continue to plummet. No doubt the authorities will scramble to look at the technology, but they'd be wasting their time - this is not an issue of technology changing - it's too soon for that to have happened significantly, and I'd hazard that Oliveira has had less R&D support than Pistorius so is likely still on technologically inferior equipment - he's just better at using them.  It's the athletes improving, because that's how a young sport evolves.

By allowing the cross-over to happen, the authorities and media, who were so starstruck by Pistorius that they couldn't see this, made the incorrect starting assumption that Pistorius belonged, physiologically, at the highest level of able-bodied sprinting.  The science was however clear - this assumption was wrong and Pistorius was getting between 5 and 10 seconds advantage in a 400m race, but the decision was made regardless.

It was thus inevitable that Pistorius' performance would soon be beaten, even with the same technology.  Oliveira today revealed just how much further we may still have to go - a 0.64s improvement in a single 200m race is astonishing, and it puts into context the true capacity of Pistorius and his performances.  And of course, on top of this, we know that technology progresses.  The dual combination of technology and a better athlete pool creates a situation almost impossible for authorities to control - they can be grateful we are only talking about one athlete at a time.  For now - in time, depth will be added to the dilemma.

So, do we now see a reassessment of the decision because an athlete may suddenly be "too fast"?  How can a decision be reversed based on the weak reason of "technology" when that technology was not even understood to begin with?  How far backwards can the IAAF bend on this?  And what happens when the next generation emerges, even faster than Oliveira?  Do we repeat the back-tracking again, in which case the entry requirement for participation in able-bodied events becomes that "you're just not quite fast enough?"  A theoretically farcical situation, but not all that far off, judging from 20.66s of sprinting today.

The issues around the performance advantages of carbon fiber blades was always going to be confirmed by the clock.  It's ticking a little more quickly than many might have thought.  Is anyone listening?


Wednesday, July 17, 2013

Tour de France 2013: Alp d'Huez preview

Preview of Alp d'Huez - history, performances and predictions

Alp d'Huez is the Tour de France most iconic climb, and its frequency and prestige make it the most informative barometer for the sport.  Unfortunately, since 2008 when the biological passport was introduced, it's only featured twice, so the more recent data is thin.  2013 adds two more ascents to the history of Alp d'Huez in the Tour, and in time, this decade will make an interesting comparison to the 1990s and 2000s.

Looking back, because it has been a regular feature in the Tour, performance history is so deep that even bad years (too hot, too cold, too windy, too tactical) can be dealt with and the emerging picture of cycling's "performance barometer" is clearest when looking at this mountain.

Below is some historical context to the climb, and my thoughts ahead of its double-inclusion in tomorrow's 2013 Tour de France.

The rankings

The figure below shows the top 140 performances, colour coded by eras.  Those eras are 1991-1997 (pre-Festina, in red), 1999-2003 (post-Festina/early Armstrong, in orange), 2004 (the ITT, purple), 2006 (pre-biological passport/post-Armstrong, yellow), and 2008-2011 (two ascents in the biological passport era, green)  (data courtesy ammattipyoraily's comprehensive list)

That's a very busy image, so to streamline it a little, I've cut out all but the Top 3 times per year in the figure below

Generally speaking, red shades represent the 1990s, and it's clear that they're heavily weighted to the bottom (higher ranking) and right (faster times) of the graph.

The fastest performances on the mountain belong, not surprisingly, to Marco Pantani, who has broken 37:00 twice.  He also occupies third place, and only then do the performances of Armstrong, Ullrich, Indurain, Zulle, Riis and Virenque feature.  The implication of this is that Pantani was finishing 200km stages with faster ascents than the ITT of 2004, which is quite astonishing (the 2004 ITT was hot, but not exceptionally.  I can't vouch for the conditions during those Pantani ascents).

Also indicated on the graphs above are the pVAM and dpVAM predicted times.  Recall that pVAM is the predicted performance based on the collection of top-3 finishers in the GC in Grand Tours from 2008-2013 (post-biological passport).  The dpVAM is the same concept, but based on the climbing times from 2002 to 2007 (pre-passport).  (Based on what you see above, it would be interesting to have a similar prediction based on the times in the 1990s)

pVAM for Alp d'Huez is 1606 m/hour, giving a predicted time of 41:48.  That actually lies to the left of the data set in the above graphs, as it is slower than the 140th best time in history, which provides some perspective on what the doping eras of the 1990s and 2000s provided.  The dpVAM is indicated as 40:22 (VAM = 1664 m/hour), which would be good for 64th on the all-time list.

The figures below show the relationship between power output and performance on the climb, once again indicating where the historical prediction lies, as well as the performance of noteworthy former champions of the Tour.

Alp d'Huez's depth & history provide a dose of perspective

The mountain-top finishes of the 2013 Tour have been dominated by Chris Froome.  When using the pVAM and dpVAM method of comparison, Chris Froome is the only cyclist who has finished faster than the predictions on both climbs.  Remember that this prediction is nothing more than a standardized, objective method of evaluating climb performance using historical benchmarks - the implications of being faster or slower than the prediction can be discussed separately.  All that matters for now is a relative comparison between different riders of the 2013 race against this benchmark.

On Ax-3-Domaines, Richie Porte joined Froome under the historical "undoped" benchmark, whereas Mont Ventoux saw ten men go faster than pVAM (that suggests favorable conditions on the day).  Nobody other than Froome has been faster than the dpVAM prediction in either finishing climb this year, and he did it on both climbs.

So, this introduces a few possible scenarios for this year:

  1. If the entire peloton reproduces what we have seen in the previous two stages (that is, ranging between pVAM and 3% faster than pVAM), then we will see a performance in the range of 40:40 to 42:00. That means that the fastest time this year will not break into the top 70 of all-time.
  2. If anyone is able to produce a performance like Froome's on Ax-3-Domaines or Mont Ventoux, then expect a time slightly faster than the dpVAM, which is 40:22. They may just crack 40 minutes, which would sneak them into the top 50.
  3. In order to get into the top 20 of all-time, a performance faster than 39:15 would be required. That would be 2.8% faster than the dpVAM, the prediction made based on times from the known doped era. So far, Chris Froome has been 1.9% and 0.35% faster than dpVAM on Ax-3-Domaines and Mont Ventoux, respectively.

    The figure below shows where Froome's performances on those two climbs would be expected to feature were he to produce them again on Alp d'Huez. The red diamond corresponds to Froome's Ax-3-Domaines performance - it is 1.9% faster than the prediction from the doped era.  The pink triangle is a performance corresponding to Froome on Mont Ventoux - 0.35% faster than dpVAM. Those same performances on Alp d'Huez would be good enough for 26th and 54th on the all-time list.

Clearly, Froome's performances on those climbs, impressive as they were, are placed into context when held up against the Tour's deepest and most extensive benchmark.  That's been part of the difficulty to date - I called it "performance pixellation", when isolated performances are viewed, well, in isolation.  The deeper the data goes, the better the insight gained, and that applies to the mountain as well.  That's why this entire exercise, and that of the Doc at the veloclinic, are about the process, not proof.   This is a point I've tried to emphasize many times, but the message sometimes doesn't get through.  Hopefully Alp d'Huez adds some more perspective to help this argument along.

That said, the tactical situation of the race may confound the Alp d'Huez ascent this year.  Much like the situation on Mont Ventoux, Chris Froome can afford to be relatively conservative, and pick a moment later in the climb to attack, if at all.  Last Sunday on Ventoux, he attacked with about 7km to go, producing a furious acceleration off what was a relatively cautious first half of the climb.  The same on Alp d'Huez would skew the eventual time - David Brailsford alluded in an interview to the fact that their data suggest Froome could have been even faster, something that was obvious just looking at the pacing strategy on the climb (the other riders provide the barometer for the pacing).  Given that he does not need the time, Froome may do the same tomorrow, so we will wait to see what the performance means in the grander, historical context.

Also of interest will be the double ascent, to see just how much more slowly they ride the climb the first time around.  I'd be surprised if the main peloton breaks 46 minutes the first time (that projects a power output of about 5.1 W/kg using Ferrari's method).  An early break, which is inevitable, could well gain 3 to 4 mins within the 21 hairpins of the climb.

And finally, the weather forecast, in keeping with the individuals time-trial I'm watching as I write this, is cold and wet.  That introduces an interesting variable, because different riders tolerate changes in temperature, and also colder temperatures, far worse than others, so the ascent of Alp d'Huez may throw up a few surprises.

The analysis and thoughts to come, particularly on Facebook and Twitter, so do follow me and then join me tomorrow for what should be a tremendous day on cycling's greatest stage.


Sunday, July 14, 2013

A house of cards? Sprinting crisis as Gay, Powell and more Jamaicans fail controls

A house of cards? Sprinting tumbles down as Gay, then Powell, then Simpson, Carter named as failing drug controls

A frantic two hour period today saw first Tyson Gay, and then Asafa Powell named as having failed doping controls in separate incidents.  More is to come, because Nesta Carter's and Sherone Simpson's names were soon added to a list of four Jamaicans who failed tests at the national championships, and that still may not be the last of the big name Jamaican Olympians implicated in this latest scandal.  As I write this, I've also just read that the coach of the afore-mentioned Caster and Simpson has been detained in Italy.  It was he, allegedly, who provided them with the supplements that were contaminated with a banned stimulant.

Regardless, the hornet's nest may well and truly have been poked.  Usain Bolt is of course the biggest of them all, and his agent was quick to dismiss claims that he is among the four Jamaicans caught in the latest swoop.

Powell's response, issued in the form of a statement on Twitter, was to deny knowingly taking a banned substance.  He did reveal which substance it was - methylsyneprhine.  It is a stimulant that is found in many over-the-counter supplements, which already indicates which way this case will go.  It has become relatively common for supplements to be implicated as the source of banned substances, though of course the strict liability rules negate this as a valid excuse.

The Tyson Gay positive is a different matter.  We don't know yet what substance it was, but David Epstein is following the story, and his contacts and nous in this area of doping make him a Twitter must-follow if you want to stay on top of the both the Jamaican sprinters, and the Tyson Gay situation.

I would be lying if I said I was not personally disappointed to read Tyson Gay's name alongside the words "tested positive" today.  He was the sport's much needed foil to Usain Bolt.  Calm, quiet and respectful to the brash, loud and showboating Jamaican.  The head-to-head between them would have been a highlight of Moscow.  There has been no denial from Gay, just a statement that included this:
"I don’t have a sabotage story. I put my trust in someone and was let down.  I don’t have anything to say to make this seem like it was a mistake or it was on USADA’s (United States Anti-Doping Agency) hands, someone playing games. I don’t have any of those stories.  I made a mistake. I know exactly what went on, but I can’t discuss it right now. I hope I am able to run again, but I will take whatever punishment I get like a man"

The supplement defence - convenient, sometimes truthful, but damaged by 

With regards to the Jamaican case, it is already clear that all three (Powell, Carter and Simpson) are heading down the path of the contaminated supplement defence (this seems likely, at the time of writing).  Sources suggest that a supplement, recently introduced to the training group, is the source, either via contamination, failure to include the name on the label, or carelessness on the part of the coaches and athletes to confirm the safety of the supplement.

This may become very important in the defence, because there is, in a sense, "safety in numbers".  Back in 2010, I covered the case of two South African rugby players, Bjorn Basson and Chiliboy Ralepele, who failed a test for methylhexanamine.  Their claim?  It was part of a supplement they had taken, but which had been cleared by the company and the sports federation as being "clean".  Their defense was stronger for their shared testimony, because they were able to pass off the 'liability' as belonging to another party.  That is, they argued that they had done everything possible to ensure no contamination, and were thus (correctly, in my opinion) exonerated.

A similar thing might be possible for Jamaican sprinters, IF they have all used the same supplement, with some record of its 'clearance'.  At least, that's what they will be hoping for.  As David asked, if 3 Jamaican sprinters from one club test positive for the same potential contaminant or unlisted ingredient, what would you make of it?

My impression, and I'd love for someone to validate this, is that we've seen a pretty high frequency of "unlucky" positives coming out of Jamaica in recent years.  Athletes who have failed tests but escaped with reduced sentences because of extenuating circumstances.  One such circumstance is inadvertent use, for which lenient sentences are sometimes handed down.  It was just a month ago that Veronica Campbell-Brown failed a test for a banned diuretic and was suspended by the federation.  Her defence revolves around a similar claim - in this instance, the banned substance is part of a cream being used to treat a leg injury, and pivotal to the claim is whether the substance appears on the label.  Yohan Blake faced a similar case, though his was complicated by the recent exclusion of methylhexanamine from the banned list.

Part of me feels sympathy at those explanations, because the supplement industry is, to quote David, the wild west, where pretty much anything goes and the athlete faces a daunting challenge to 'get lucky' or avoid all supplements.  We saw it in South Africa, and it was seen so often in 2010 that I somewhat satirically named methylhexanamine the drug of the year.  It was a textbook illustration of the challenge faced by athletes.

But there's also part of me that feels the excuse is too ready-made, too easy because it is in essence, almost unprovable.

The concept of strict liability, which states that the athlete is liable for every substance in their bodies, was created specifically to plug this potential gap - in a world where every doping positive could be blamed on accidental ingestion from a supplement or medicine, doping control would be impotent.  And Jamaican sprinters seem unluckier than most, it must be said, when it comes to inadvertent doping.

So once again, the skeptic in me sees statements rushed out via Twitter, and claims that the coach has been arrested because he is the one who prescribed the supplements, and sees them as more of the same deception that has wormed its way into all sport.  It's very difficult to believe these excuses, though ironically, some of them are arguably true.  However, much like the debate we had last week around cycling has shown, the excuses and the valid explanations have all been lumped together, and taken away from the honest by those who cheated with impunity, who have discredited even valid explanations.

The Jamaican story is nowhere near completion.

Changing behavior - a painful and necessary process

Another point worth considering is that the frequency of testing in track and field, particularly in Jamaica, has been very low.  It's been reported that only one single out of competition test prior to the London Olympics in 2012.  This sudden "deluge" may be the result of an increase in testing frequency, and long may it continue.  It is only with scrutiny and significant skepticism that behavior will change.  Unfortunately, part of that behavior change is the 'pain' of heroes falling.  

It may be the catalyst for education - stay away from supplements, dear athletes.  There is little evidence that they work anyway, so the upside is small, and as you can see from Asafa and co, the downside is potentially enormous.

It may be the catalyst for more skepticism, which is only ever a good thing.  Again, this came up just last week against the backdrop of skepticism in cycling, but it is as relevant, and perhaps even more important in other sports, because they have not had their dark secrets exposed nearly as much in tell-all books, and by persistent journalists.  Sometimes, it is necessary to burn something down, or blow it up, in order to rebuild it, and as painful as it is to be denied, for instance, the Bolt vs Gay showdown in Moscow, it is necessary for the sport.  Unless of course one wishes to argue that doping should be legalized, to avoid these controversies.  That is an entirely separate debate, but one that is bound to be energized by the disillusionment all fans would have felt earlier today, when one name after the next fell in one of sprinting's darkest days.

That is a debate for another day.  These stories, both Gay's and the Jamaican's, have yet to be fully written.  Follow David Epstein and Juliet Macur for updates, they are on it!  I will weigh in in due course.

There is also the matter of the Tour to cover, and more significant debate to decipher.  That too, in good time.


Mont Ventoux Preview: Looking forward by looking back

Historical analysis of Mont Ventoux, pVAMs and predictions

Today the Tour tackles Mont Ventoux, one of its most famous mountains, though less frequently climbed than would be ideal to do detailed analysis of times.

The climb has been done three times in the Tour de France this century - 2000, 2002 and 2009.  It is also featured regularly in the Dauphine, which is where the record ascent comes from.  That belongs to Iban Mayo, who ascended the timed portion (15.65km from St Esteve) it in 45:47 during the individual time-trial of 2004.

The record for a climb during the Tour belongs to Marco Pantani, a 46:00 ascent in 1994.

In the current era, the top 30 times are shown below, by ranking.  I have indicated some noteworthy performances - the top times of 2000, 2002 and 2009 belong to Pantani & Armstrong (49:01), Armstrong (48:33) and Schleck & Contador (48:57), respectively (times courtesy vetooo on twitter, and available here).

Also indicated is the predicted performance based on Scott Richard's and Doc's pVAM method - that corresponds to a pVAM of 1617 m/hour, a calculated power output of 5.63 W/kg (Ferrari method), and a time of 50:46.  You may recall that this pVAM was produced by analyzing the performances of the top 3 in the Grand Tours since the biological passport was introduced (2008 to present), and then working out a prediction that is influenced by the length of the climb, its gradient and its altitude.  The method allows different climbs to be compared against historical 'benchmarks', as it were, for theoretically non-doped riders (given some of the names on that list, this is of course untrue, but doping is relative).

The prediction based on the dpVAM is for an ascent at 1683 m/hour, good for a time of 48:45.

You may recall that back on Stage 8, on Ax-3-Domaines, the only two riders who were faster than pVAM were Chris Froome and Richie Porte.  Froome's ascent, much debated here and elsewhere, was faster than the dpVAM on that day as well, so it will be interesting to see how the entire peloton performs today.

More detailed analysis is shown below, this time predicting the performance time based on a range of VAMs (top graph) and also calculated power outputs (bottom graph, using Ferrari-derived power output, for ease of use)

Five observations and thoughts

  1. The record ascent of Pantani is a crazy fast performance, coming as it did during a 231 km stage (the summit was not the finish line that day).  Eros Poli got himself about 20 min clear at the bottom of the climb, but held on despite Pantani's record pursuit.  The performance predicts a power output of approximately 6.4 - 6.5 W/kg, and given who it belonged to, it's fair to say that 6.4 W/kg for 46 minutes represents a doper, at the peak of his ability, and at the peak of doping assisted performance.  In the 2000s, post the EPO test, doping got more sophisticated and probably less effective, which is why the times of the 90s and early 2000s are a little faster than those from the mid-2000s (you'll see this clearly when we look at Alp d'Huez later in the race).

  2. So, 6.4 W/kg for 46 minutes?  Makes one wonder about the recent claims of 6.5 W/kg for an hour being acceptable for non-dopers, the "new rung" of normal, I believe it was called.  Last week, all estimates for Chris Froome's Ax-3-Domaines ascent ranged between 6.2 W/kg and 6.4 W/kg for 23 minutes.  He'd need the same performance for twice as long to match Pantani, and an even better performance for almost three times as long to hit that 'new rung'.  I'll give Allen the benefit of the doubt and say that something is lost in the definition and application of FTP to TDF performances.  And failing that, you can always buy a power tap and use it to train smarter, as per Allen's signature bar on his comments, and his criticism turned sales pitch.

    Point is, I think it's fair to say that we won't see this type of performance today, notwithstanding the fact that Froome doesn't need that kind of performance at all.

  3. The second, and very important thought is that Mont Ventoux is a very difficult mountain to predict because of the likelihood of wind.  It's almost always windy, because it is so exposed, particularly in the second half, and so a headwind or tailwind, as I've explained a few times during this Tour, will affect the calculations because they'll affect the performances.  We also have only three recent ascents to work from, plus that outlier of a performance by Pantani.  Alp d'Huez is a far better climb to model things like pVAM for, because the data set is so much bigger.

    Apparently, there is a headwind forecast for today - that would slow the times down, so that all estimates of power output would under-estimate the real performances.  So let's keep that in mind so that we can use some common sense in the discussions later, bearing mind there is some error and we don't get to see the actual SRM values.

    What we can do, after the stage, is try to overlay wind maps onto the Tour map and get a better idea of what the wind would have done.  I will also, I hope, get some decent finisher's SRM values, which can be used to validate any estimations.  It's not that difficult, so before everyone has a coronary about the error, just breathe, and let's remember to be sensible about understanding the context.

  4. For some thoughts on that error, incidentally, have a look at this model created by Doc recently.  It breaks the power output up into its three components - gravity, wind resistance and rolling resistance.  What he is basically saying is the following (for those who don't read the language too well!):
    1. The gravity component is far and away the largest.  This is important for error discussions, because the error due to gravity is relatively small because it's always constant.  The calculation of the absolute power output can be affected by the assumed mass of the rider plus the bike, but the relative power output is not too badly influenced by this (it cancels out - try it and see).  There's around a 1% error in the calculated absolute power output for every 1kg "error" in mass, but the relative power output (W/kg) is only 0.3% different.  All in all, as Doc points out, the error in the gravity component of power output is small.
    2. The error in the wind component is much larger, which has implications for the assumptions of cdA (drag co-efficients).  In the model, however, the relative contribution of the wind during climbing is small, and so the total error is actually not too bad.
    3. Similarly, overcoming rolling resistance contributes only 5% to the total power output on climbs, and so huge errors in assumption would be needed to push the total error way out of control

  5. Finally, some thoughts on the day.  After Froome's small time loss in the final 30km of Friday's windy stage, there is renewed interest in the yellow jersey race, mainly because other teams now have some more optimism about their chances.  Prior to that, Froome's dominance on Ax-3-Domaines and in the TT suggested that others would be hoping for a major change in his form.

    He described the time loss on Friday as a moment of carelessness (as opposed to weakness), which is feasible in those conditions.  Perceived weaknesses from Sky last Sunday (not a single second rider in support of Froome in the top 30 over the final two Category 1 climbs), and on Friday, will provoke others into fancying their chances of isolating him very early on during the Ventoux climb.  That will, at the very least, make for an interesting tactical battle.

    Those tactics and attacks (by Movistar on that occasion) split the race last Sunday, but Froome was good enough and astute enough to control the attacks that came from Movistar.  A similar thing may happen today, though whether Saxo or Belkin can offer anything more than Movistar remains to be seen.  I'd bet the Froome will be relatively comfortably able to respond to Mollema, Contador and Ten Dam, based on his form in the race, even without team support.

    In terms of our analysis of the power, last weekend in the Pyrenees suggested were were looking at two races - the peloton was going relatively slowly (much slower than the pVAM), whereas Froome was faster.   Having seen the time that he can gain in the time-trials, Froome no longer needs to gain time in the mountains, and can thus afford to follow wheels, perhaps attacking for seconds closer to the top, if he wishes.

    As a result, the climb today will, in my opinion, be done at the pace of the peloton, and so I don't expect that we'll be having discussions about "super-human" or "mutant" performances.  I think it will be slow, because its best rider will be conservative.

    However, let's hope for a really interesting day, and proactive riding from the Spanish teams, and Belkin, with some French aggression thrown in for Bastille day.

Follow me on Twitter and on Facebook for first thoughts during and after the stage, and then I'll try to do a more comprehensive analysis here on the site by tomorrow.

Enjoy the racing

Tuesday, July 09, 2013

Healthy skepticism, dealing with doping and denial

Cycling discrimination: why skepticism should be encouraged, not silenced

Yesterday I wrote a lengthy piece in which I tried to provide some balance to the discussion about Froome's Ax-3-Domaines performance on Saturday, its physiological implications and the importance of circumspection in the answers we seek.  It was a data heavy piece, so have decided to follow it up with a more opinionated one.

A parallel universe - "cycling is in great health"

A quick thought on the general state of cycling first, though.  Imagine a 2013 Tour without Sky on Saturday afternoon on Ax-3-Domaines.  The fastest ascent of the day would have been 24:18 by Alejandro Valverde, well over a minute slower than the best Ax-3-Domaines times of 2001.  Behind him, a trail of riders with big time gaps on a short climb, including group of top Spanish climbers who lost over two minutes on the day, finishing in 25:16.  These are times that would not even make the top 60 on this climb (even correcting for error in timing!).

Had that happened, we would all be feeling pretty good about the world of cycling on Sunday morning, because the times would have gone in the "right" direction.  We'd say "See, the sport has changed, the stricter testing of the passport and the increased scrutiny are doing the job".  We'd be particularly impressed that these "slow" performances came on the very first mountain stage, when riders should be freshest. For context, a Twitter follower (ExRoadman1) reliably informs me that when the Tour did Ax-3-Domaines in 2001 and 2003 (the fastest ascents), they'd raced 1923km and 2095km before the stages, respectively.  2013?  Only 1148 km, and so it was very slow despite better circumstances for a fast time.

Yes indeed, had it not been for two riders on the day, we'd be very pleased with the 'health' of cycling! We'd be wrong to be so certain of course, in the same way that we'd be wrong to conclude that it's terrible and overrun with dopers (is it telling that Valverde would be the patron of a clean peloton, for example!)

So, it was Froome and Porte who spun us around.  Porte, for his part, had a terrible Sunday, something that can be viewed in one of two ways - either he paid the cost of a supra-maximal effort, a lack of pacing suited to a stage race, and we'd be more optimistic.  Or his variability is suspicious.  I'd go with the former, since doping is no longer as beneficial to once-off efforts.

As for Froome, he rode well enough on Sunday, but not spectacularly fast - the estimates ranged from 5.0 to 5.2 W/kg on those climbs, and he was accompanied by 30 other riders, so you could make the case for it being a heavy day following a fast one.  The point is, the efforts of two riders have raised eyebrows, specifically around one team, but should not by themselves be viewed as guilty.

Views on the value of constructive skepticism

On that note, and lest my last two thoughts place me on the opposite extreme of a 'dope-free cycling apologist', I thought I would follow it up today with what are really just my opinions on the sport in general, and why we have these heated debates about victimization of teams and riders without, apparently, justification.

I read a tweet the other day saying that it's unfair that Usain Bolt is exempt from suspicion and interrogation even though he is running faster than Ben Johnson 25 years ago.  That it's not fair for Andy Murray to not be questioned on his victories in the way that cyclists inevitably are.  Couple points.  First, those guys are not entirely exempt.  Granted, they don't face the same level of scrutiny, but it's too far to say they face none.  Second, they should be pressed more, because there is no doubt that cycling is not alone, and may, in some regards, be even better off than other sports in the anti-doping battle (no, not you, UCI, you don't get to take any credit for that).

But more to the point, I'd want to caution against pointing outwards when the debate asks people to look inwards.  The focus that should be placed on other sports, on other teams within a sport, or on other riders within a team, is a parallel matter, one that should be absolutely be dealt with, but not as part of the answer to specific cycling questions.  In this matter, I do not agree that anyone becomes less guilty because their illegal behavior is commonplace.  If the issue is cycling, let's talk about cycling, not tennis or athletics, because we'll never move from where we are.

Cycling is where it is, in part, because too many people who might have added value early were silenced or cast aside as being problematic, unwanted because they 'spat in the soup'.  The result, to paraphrase a piece by Paul Kimmage, is that the denial of doping hurt cycling more than doping.  And the easiest form of denial is not to openly deny doping ("It doesn't happen"), it's to distract from the debate by diverting questions and pointing to others, which seems, in my opinion, to happen too often.  We all hope, even the most cynical, that the riders we watch today are clean, or at least cleaner than those of ten years ago.  The mere existence of ongoing debate is, I hope, indicative that people want change and want to believe.  Few are maliciously cynical, even if they have by now forgotten their real purpose of becoming vocally anti-doping.

And so I would hope that those who defend the sport will at least find it possible to recognize the origins of the skepticism, and why they should not be trying to silence or divert the questions and allegations, but rather encourage them and heed the solutions they may reveal.  The mistrust of cycling can be turned into constructive feedback, unless it is diverted through defensiveness.

History repeating?

We find ourselves in the 100th Tour, the first since USADA's investigation of Armstrong exposed the Tour's previous "renaissance" after the Festina scandal of 1998 as nothing more than a more sophisticated fraud.  Back then, the sport was looking for salvation, and a Texan rolled into view with promises of miracles and hope.  He would eventually use the Champs-Elysees to admonish people who didn't believe in miracles when he finally departed in 2005.

Well, we now hear much of the same.  Promises and re-assurances that don't quite jibe with actions.  I read David Walsh's tweets from within Team Sky over the weekend, where he dismissed similarities between Froome on Saturday and Armstrong in Sestrierre in 1999, by basically arguing that Froome is different from Armstrong because Froome is a nice guy who would not pull a "Lance" and intimidate an outspoken rider the way that Armstrong did with Bassons in 1999.  

I see many others are making similar points - Sky have too much to lose, that they are credible because they have the pressures of UK Sport on their back.  They are team of nice guys, just very smart in their preparation, very scientific.  Well, I recall that Tyler Hamilton was a nice guy, one of the best in the peloton.  Hincapie had the respect of everyone, as have most dopers, before they were caught.  USPS was a credible sponsor, with too much to lose, and so was Discovery.

In terms of preparation, USPS and Lance Armstrong were systematic, scientific and methodical, you may recall?  Lance was called "Mr Millimeter" for his attention to detail, and the Europeans, they didn't know how to train for the race.  Lazy, unfocused, disorganized.  USPS, we were told, got the jump because of their scientific and systematic approach.  As for Lance Armstrong, he had cancer, surely he would never risk it all, and the hopes of millions, to cheat in a bicycle race?

Since when did the sponsor's governance and credibility become an effective umbrella against doping accusations?  And when did "nice people" not cheat when placed in the worst of circumstances possible for them?  Given what we have learned about cycling and its culture, how can we continue to confuse superficial personality with doping decisions?  Tyler Hamilton did seem a nice guy, and he still does.  So do many who testified to USADA, and to other commissions.  Their doping mistakes don't change that.  Are we still stuck in some kind of eastern European, iron-curtain clad generalization or stigma that dopers are ruthlessly seeking world domination driven by cold-war sporting philosophies?  Or that dopers are sporting equivalent of Gordon Gekko, relentlessly pursuing greed as the driving force behind rampant doping (that's a Wall Street reference, for the uninitiated)?  

It has become abundantly clear that in cycling, some good people were caught up in a very bad culture, that they made bad choices, but were not necessarily bullies, evil-doers or criminals (ok, some were, granted).  So where, in the words of one Twitter follower, does the "dreamy eyed pap" by some in the (UK, mostly) media about guys being too nice to dope originate?  How short are our memories, that we consider 'nice guys' to be even close to an admissible characteristic of a non-doper?

And as for "too much to lose" - millions of hopes were not enough for Armstrong, so I'm afraid you'll have to forgive the cynics when they don't quite buy the empty words of "trust me, I'm clean, I've got a lot to lose here".

Sky's specific place in the spotlight

Related to this, some on the weekend were accusing people of going after Sky specifically, that Sky should not have to answer these questions more than any other team.  That's only partly true.

As the team who hold a yellow jersey that has been stained by 20 years of doping and cheating, it is inevitable that they must field more of the questions.  This does not exempt Astana, FDJ, Garmin, Europcar and co. from the same scrutiny, but it is obvious that the winning team in a dirty sport will attract the skepticism.  After all, winning in cycling became synonymous with doping, so where else would you find a doper, if not wearing a yellow jersey?  That's the rationale of the skeptics and some in the media, and I think even the most ardent fan has to respect that reality, particularly if they wish to change it.  

For me, the issue with Sky is not that they are behaving the same as every other team when it comes to things like their position on doping, or the release of data (biopassport and power, as we discussed the other day), it's that they are doing so off the back of a promise that they'd be different.  They rolled into the sport with a commitment to transparency, to show that the Tour can be won clean.  That was their implied mission - win bike races differently. 

Even if they are held to account only to that promise, it's clear that they are currently inviting suspicion by not doing it.  It seems to me they have a choice in every instance, A vs B.  I'd call it an opportunity, rather than a responsibility, to emphasize that it is their choice.  A choice between two options, one which might confirm a new paradigm, and one which would re-inforce skepticism from outsiders.

As Shane Stokes, a Velonation journalist who DOES ask the challenging questions, has also mentioned, under pressure to show they are clean, they could go to extra lengths and invite experts to see blood and performance data, or they can choose to ignore it.  The latter option will only ever serve to heighten suspicion because of their promise and their dominance.  When the power output data is withheld because "pseudoscientists" can't be trusted to interpret it properly, and because it would create noise, that action is diametrically opposed to a mission of transparency and "difference".  It is choosing a course of action that will grow suspicion, when there is, in my view anyway, an option that would benefit them and the sport by diminishing it.

They have hired staff with very suspect histories and then patronized people with explanations that they did not know, when just about everyone else did.  They haven't engaged with experts like Ashenden despite having nothing to hide.  Is it a lack of trust in the experts, as much as in the public?  Do they not believe in the system they would need to work within?  Perhaps, and that is their right.  But they come in for scrutiny because they leave people guessing.  That, and the fact they have twice dominated the Tour de France (and much of the seasons, when you look at Wiggins in 2012 and Froome thus far in 2013) while being, well, the same as everyone before them.

Looking ahead

So the questions will continue, as they should, because national pride and promises are not a good enough reason to believe in people given the number of lies that have preceded them.  My point is this: I think most of us want a cleaner sport.  Some wish to arrive there by interrogating everything, by examining every detail, by challenging every performance.  They can (and do) cross a line into unfair accusation from time to time.  Others want to look in a new direction, forget the past and dismiss all questions as biased, destructive, jealous, racist (yeah, I got this one the other day).  Somewhere in the middle is clean cycling, and denying and diverting the questions blocks off access to that point.

The Tour now builds to its traverse of the Alps, and some incredibly difficult stages.  Given the way Froome has looked, and the relative 'weakness' of his rivals, it is unlikely we will see another maximal effort in this Tour - he can follow wheels, and gain seconds near the summit of the climbs if he wishes.  

And so performance analysis becomes less and less powerful.  The rest of the Tour, based on Ax-3-Domianes, will continue to ride the HC climbs at the pVAM, and so will Froome.  How much he has in reserve, we'll never know, unless someone discovers another level and shakes the Tour up.  We look forward to that.


Monday, July 08, 2013

Tour rest day: Pondering the unanswerables with physiological implications

Pondering the unanswerable questions at the TDF: Performance as proof of doping?

I spent the weekend, and much of today, pondering the many questions that Chris Froome's very fast ascent of Ax-3-Domaines asked on Saturday.  They are unanswerable questions, as much today as they were on Saturday evening.  So I have no great epiphany to offer you, no sudden insights that have emerged from the fog of discussions that ensued on Twitter, no definitive conclusions about the credibility of the Tour's latest maillot jaune.

However, I have a few brief thoughts about the debate in general, more conceptual than factual, given that we have two weeks of similar discussions to look forward to.  And the more it is discussed, the more I perceive the need to try to explain (probably in vain) the concept of performance analysis.

Dogmatic answers help nobody

First, I have realized that most people (present company excluded, if you've read this far) expect a simple answer to a complex problem. In fact, their desired answer should ideally fit into 140 characters so that it can be tweeted, and should not leave any room for interpretation. Where such room is left, people will choose to interpret it in a way that either seems naive (Froome and Sky must be clean), or pseudoscientific (change the name of your website to "the speculation of sport", for example).

Earlier today, I read this article which was written by Antoine Vayer, who is now becoming known for interpreting Tour performances in largely one way - proof of "beyond mutant" physiology (those words belong to Google Translate, which did such a job on that article that gave me flashbacks to assembling a table tennis table with Chinese-English instructions).  He writes that Froome's calculated power output of 446 W (this is normalized to a 70kg rider, by the way, so it's 6.4 W/kg, not 446/66kg, please note) and describes it as "miraculous", leaving no doubt as to his conclusion that Froome is doping (apologies if something has been lost in transation).

Vayer is certain of his conclusions, that is for sure.  Too certain, in my opinion.  Because here's the thing:  While he may very well be correct, he is equally likely to be wrong, for a couple of reasons.  First, as I've really laboured at every opportunity (though the message still hasn't been received, based on the pseudoscience criticisms), the estimation of power output is likely to carry with it a large enough error that you simply cannot say things with 100% conviction.  Some confidence?  Yes.  100% confidence? No, not yet.  The error in estimation exists because the assumptions of environmental conditions and mass influence the calculated value.  A 3% over-estimation turns a calculated power output of 446W, or 6.4 W/kg, into 425W and 6.2 W/kg, which has implications for its plausibility.  On this note, I'd also point out that the number calculated is just as likely to be an under-estimate than an over-estimate, so instead of dismissing 6.4 W/kg as high in error, bear in mind it might actually be 6.7 W/kg, and low in error.

I must also add that I think people are over-playing this error, and using it to distract from the debate - 3% is large, and most of what we've seen is smaller, barring a few exceptional situations.  If the error is acknowledged for obviously extreme situations, and performances are viewed collectively over time, rather than in isolation, I believe they get small enough to not impact the conclusions in a large way.  Beware the "pixelation" that happens when you stand too close to the picture, and you still get value.  So I really do believe that the methods we've used are more accurate than some wish to acknowledge, but they're not perfect.  In this world of extremist views, however, "not perfect" is a synonym for "worthless", which the method most certainly is not.

The physiological implications - probabilities, not certainties

The second reason I feel Vayer should adopt more caution with his conclusions is that there is room for interpretation in terms of what is "mutant" as opposed to plausible, even if the power output is accurately calculated (or better yet, measured).  Vayer has interpreted his calculated power output of 446 W as almost mutant, strongly associating Froome with known dopers in the past.  

Again, maybe Vayer will be shown to be correct in time, but that Ax-3-Domaines performance in isolation is not necessarily proof of doping.  Nor is the 411W to 420W range of the other finishers in the "suspicious" range in the way he suggests. Of course, all the performances are suspicious - as the top finishers in the Tour, they are, sadly, under suspicion by their presence and the sport's history!

However, Vayer's definitive conclusions seem, at least from the translations I see, to be made without full recognition that the performance is the result of very complex physiology which is influenced by many factors, one of the most important being the duration of the climb.  I'm told that in his magazine, "Not Normal", he does adjust the expected power output based on the length, which is vital, because intensity should be higher on shorter climbs (apologies again if I have misconstrued this previously).  That correction, or explanation, is absent from the latest article (as is the 70 kg method explanation, which also causes major confusion), but I would also disagree on the model he has arrived to get there, and that's what makes his certainty misplaced, in my opinion.

On the matter of length, when a climb is short (and 24 minutes is relatively short), then a number of 6-6.1 W/kg is less suspicious than it would be over 40 minutes.  This should be obvious, I hope.  Physiologically, 6.4 W/kg for 24 minutes does not ring any alarm bells in and of itself.  Remember, the origins of this approach are basically that performance implies physiology.  Therefore, you can work backwards from power to estimate the physiology driving it.

The challenge is that there are three factors to account for - the VO2max, the efficiency and the sustainable workrate.  If only two existed, a model could be created where one is played off against another, until a power output created an impossible balance, and a given performance could be flagged as unreasonable.  However, when three factors are in play, it's impossible to do this with certainty - too many 'degrees of freedom'.  

For example, that 6.4 W/kg calculated for Froome can be used to calculate an implied VO2max, by assuming his efficiency and his sustainable workrate, two of the three variables.  Typical assumptions for both (24% and 85% of maximum workrate) produce an estimated VO2max of 88.6 ml/kg/min.  Very high, and I'd be doing a double-take if I saw that combination of relatively high efficiency and super high VO2max in a lab, but it's not 'mutant' or impossible, just very unusual.

However, what happens if he is able to ride at 90% of maximum, given that it's only a 23 minute effort off the back of a relatively easy stage?  Well, now, the prediction for VO2max drops to 83.7 ml/kg/min.  And if his gross efficiency (a measure of how thrifty he is in using oxygen to produce energy) were to increase to 25%, then it's only 80.4 ml/kg/min.  I have summarized a range of these possibilities in the table below, with blue for lower efficiency (23%), green for moderate efficiency (24%) and red for high efficiency (25%).

On the longer climbs, lasting 40 minutes or more (think Alp d'Huez or Mont Ventoux), the sustainable power has to be lower - 80 to 85% would be, in my opinion, quite aggressive assumptions for the end of a 5 hour day in a three week stage race.  Based on the pVAM method we've been discussing, it appears that the typical intensity of those longer HC climbs at the end of stages is around 75 - 80% of maximum, especially considering that many of the climbs are at altitude.  

So, in the table above, if a rider is at 6.4 W/kg for the 40 minutes of Alp d'Huez, with an efficiency of 24%, their VO2max would need to be around 94.2 ml/kg/min and they'd need to be riding at 80% of maximum, and I do believe that is unrealistically high.  That's why a realistic performance for those longer climbs is 6.0 - 6.1 W/kg, because that brings the physiology back down to very high, but plausible levels. 

The point is that when many things are free to vary, the physiological prediction you make is a reasonable GUIDE, but not a perfect solution.  Therefore, dogmatic conclusions about mutant physiology don't do anyone any favors, least of all those of us trying to steer the conversation towards greater insight that is fair and realistic.

That said, the diagnosis, judgment and conviction of doping by performance IS possible, but the certainty only comes from truly eye-popping performances.  If a guy produces 6.5 W/kg for 45 minutes, I'll be calling shenanigans with confidence, because that's a rider, however you play with the assumptions, who breaks the 'rules', as the table below shows.  Then you can be more conclusive about calling out 'mutants' (but still not 100% sure, because we are limited by models.  The big picture would complete this, taking it from 99% confident to 100% sure)

Froome's performance on Ax-3-Domaines, whether it is 6.3 W/kg (Ferrari method), 6.4 W/kg (Vayer), or even 6.5 W/kg, lies below that line of "certainty", but above a line of "clear plausibility".  If there are zones - red for certain doping, green for very realistic undoped, and orange for "could go either way", we're in orange on Saturday.

The historical analysis of the performances since 2009 (using the pVAM method) suggest that Froome's performance, and the physiology driving it, lie closer to the limit than anything we've seen since the biological passport.  And the way Sky went 1-2 on the day says they have found a competitive advantage, certainly.  The source of the advantage is however not as clear cut, and so Vayer's interpretation of a value as plausible vs suspicious vs mutant/miraculous is too certain.

This in mind, the method of analyzing performance is not worthless, and I fully understand what Vayer is doing and support it.  In concept.  However, it does a disservice to the sensible interpretation of data when a number is categorically held up as proof of doping.  

If I did that, and wrote that "Froome is certainly doping based on my calculations that his power output was 6.5 W/kg", I would hope that many of you reading this would call it out and run me out of town on the basis of exceeding the strength of the analysis with my conclusion!

The value - the long term process, the hypothesis and the big picture

The value of this analysis is not the definitive answer is provides, but rather the process it asks of people.  All these "tools" - VAM, pVAM, CPL, estimated power output, physiological implications, are merely designed to help generate theories and hypotheses that, over time, might either be proven or disproven by results.  If that hypothesis is that the Tour has gotten slower as a result of being cleaner since the biological passport and sponsor/media driven pressure, then let the analysis show it, over time.  Because over time, all that error that seems to freak people out would become smaller and smaller, and a better understanding of the big picture would emerge.   Not in isolated efforts, which is the point of what I tried to express immediately after the stage on Saturday.  It's the process, folks, not the outcome, as is the case with sports science all the time.

On that note, the performances of Sky riders (with the exception of Froome) on Sunday gave some cause for a bit more optimism, because they revealed that perhaps Saturday was a supra-maximal effort, the costs of which were paid on Sunday.  The racing on Sunday was sustained through the whole stage, but the climbs were actually not as fast as they were Saturday - we were projecting between 5.0 and 5.2 W/kg on the final two climbs.  The fact that the front group was as large as 30 pretty much the whole way up both those Category-1 climbs confirms that the pace was not fast.  Yet Sky were absent. It is indeed ironic that we look for signs of weakness to dispel our cynicism.  Once again, that day in isolation means little, but it was, after Saturday, a reason to think twice about Saturday's performances.  Avoid performance pixelation, step back, see the whole screen.

More reading and listening

I'll call in a night there, and leave you, if you feel like more, with one more article to read and one podcast to listen to, on this subject.

This is a piece I wrote for a local paper - very short, very general, but summarizing what I've written about in the last few days.

And then this is a podcast of an interview I gave on SA radio today, discussing some of the questions I've attempted to address in the post above.

Tomorrow, I have a more opinionated piece on cycling and why it should continue to attract questions, particularly for Sky