Wednesday, July 18, 2012

Is Doping A Necessary Evil?

It’s TdF time again. Unfortunately, nowadays, that means everwhere you turn an ear in the circles of endurance sport you’ll hear all kinds of folks, some qualified, most not, postulating on who’s on what.

Most of these opinions will be based on little more than conjecture and the individual’s general world view, however, some have put a little more thought into the implications of today’s performance level on the question of the prevalence of doping. Ross & Jonathon at the Science of Sport website ( have put a lot of time into looking at what some of the recent TdF stages  suggest about today’s top cyclists’ physiology and more specifically, how they compare to yesteryear’s athletes who competed in an era in which doping control was less advanced.
A thumbnail summary shows that today’s top athletes (Wiggins et al.) are putting out somewhere around 6.4W/kg over short climbs (~20min) & <6.0W/kg over the longer climbs (40min+) whereas cyclists from Armstrong’s era & earlier were ~6.8W/kg for the short climbs and ~6.3W/kg for the long. These figures for the fastest tour climbs in history (from w/kg calculations based on the VAM metric of  Michele Ferrari, 2009) are shown below.

I’ve added my own calculations of suggested VO2 max of these athletes based on an economy of 78W/L & a fatigue curve of 5% - typical numbers for pro level cyclists that I’ve tested/have data on, though the best cyclists in the world may be even higher. Lab data from Miguel Indurain (Padilla et al., 2000) suggested efficiency of 26% or ~90W/L.
The estimated VO2max of ~90ml/kg/min is marginally lower than the numbers that Ross & Jonathon came up with, presumably because of the differences in the assumption of %VO2max being held for the respective durations. I have good data to suggest that elite level cyclists are holding >95% WVO2max for durations in and around the 20min mark & >90% WVO2max for efforts in and around 40min. Again, these are elite, not world class athletes.
So, the big question becomes, how feasible is it for an athlete to ‘train up to’ a VO2max in the vicinity of 90ml/kg/min? Certainly, lab values in excess of 90ml/kg/min are exceedingly rare though not unheard of. Cross Country Skier Bjorn Dahlie, cyclist Greg Lemond & ultra runner Matt Carpenter have all cited test values of >90ml/kg/min. Several of the cyclists in the above list have been tested in the mid-high 80’s. Were all of these numbers the result of doping? Given the relatively low economic incentive of sports like XC skiing or, especially, ultrarunning, coupled with the fact that a year of  EPO use costs in the vicinity of $30,000 (Riis, 2012) I would  question the economic reality of that assumption. But without doping, are these numbers even plausible?
This is a tough question to answer, but an important one. It’s dangerous for a physiologist to say that just because the VO2max numbers implied by these performances far exceed what they have measured from athletes in their own lab that they are physiologically ‘implausible’. History is full of athletes with ‘implausible physiology’ who were at their time head and shoulders above the rest of the field. This was true long before doping reared its ugly head. It will be a sad day in sport when any performance that exceeds the norm for the day by a large margin is looked on with suspicion. I hope we are not there yet. I hope that we still at least entertain the possibility that athletes will come along every so often who take the sport to the next level.
So back to the question, how feasible is it for an athlete to ‘train up to’ a VO2max of 90ml/kg/min? I spend a lot of time looking at the training response of various types of athletes. I rigorously monitor the relationship between training load (TSS) & fitness (VO2 score) for a wide array of athletes of all different levels and abilities.
What do these observations suggest for a (non-doped) athlete targeting a VO2max of 90+ml/kg/min?
10+ years of consistent, progressive overload.
While there are individual differences in training response, if we look at the average, a pattern of diminishing returns of a ‘half life’ of 15ml/kg/min per year would be a ‘normal’ training response, i.e. +15ml/kg in the first year of training, + another 15 over the next 2 years, + another 15 over the next 4 years etc. For an athlete starting with an untrained VO2max of 35ml/kg, 10 years of consistent, progressive aerobic training will theoretically yield a gain of ~55ml/kg, i.e. 90ml/kg/min. Of course, lots can happen over 10 years of training & a true consistent progressive build over that time is rare. Though for the truly driven athlete, the type that might go out for training rides in the middle of chemotherapy treatment, say, not impossible.
While an optimist & someone admittedly clinging to the purity of sport, I’m also not naïve to the possibility that the type of athlete who is that driven to be the best, might resort to short cuts and, to be sure, the physiologist in me knows the relative benefit of these short cuts. Even small increases in haematocrit of 10% will yield similar improvements in VO2max, i.e. for a domestique of say 75ml/kg/min, it would be very tempting to skip 3-4 years of development and jump ahead to a VO2max in the mid 80’s and a team leader position (with an extra few hundred grand a year). Before the HcT limit of  50%, the potential for gains was even greater which is  why you would see young riders come out of nowhere for a season and then literally drop dead!
However, for athletes who have ‘paid their dues’, it is my position that these levels of performance (consistent with a VO2max of ~90ml/kg/min) are indeed possible. Similar levels of performance have been seen both in sports without the financial draw of cycling & before doping was widespread. As the endurance sport with the highest financial draw, cycling will continue to see those looking for shortcuts however I truly hope that we don’t fall into the pattern of automatically discounting the ever-true performance benefit of devoting a decade or more to developing your craft.
Train Smart (& clean),


Nicolas said...

I guess I'm more of a cynic and skeptic, but unfortunately I'm much more pessimistic about doping than you are. I agree that it looks like the peleton has become cleaner over the past few years, judging by lower VAMs, lower average speeds (purely subjective impression) and higher number of crashes, which would indicate tired riders. However, there are a few points that point me to a more skeptical view, especially with regards to the past:

1) Doping being a relatively recent phenomenon: from your article I got the impression that you suggest that doping is relatively recent as a phenomenon. I would say quite the contrary: Tom Simpson dropped dead on the Ventoux in 1967 whilst on amphetamines. Eddy Merckx was apparently tested positive three times. Laurent Fignon, before passing away from cancer, blamed his health problems on his extensive doping. So I think doping has been around for a long time rather than being a more recent phenomenon.

2) Lack of incentive to dope in cross country skiing: unfortunately that sport also has a rich history of doping: Johan Mühlegg in Salt Lake city was a spectacular case; the whole Finnish team got caught a few years back as did the Austrian team (own-blood transfusions). One mustn't forget that in Scandinavia cross country champions are heroes and that olympic sports mean medals and thus another incentive to dope. Bjorn Daehlie, for example, has a successful high-end clothing company these days - so the sport is not completely without money. Even in ultra running Dean Karnazes seems to have had commercial success.

3) In cycling almost every tour winner of the last decade or two has been found doping at some point. If we accept that doping has a big effect on performance and someone beats those - in all likelihood doped - riders easily, then the performance becomes even less plausible.

4) There is so much circumstantial evidence surrounding so many riders that cannot be used legally but that gives paints a consistent pictures (Hamilton 04, Armstrong retests, etc), the chance of those unlikely performances being a freak of nature or very very good training just seems less and less likely.

I would like to believe that sports are clean, but my senses tell me that it's not the case. Sad, I know.

Alan Couzens said...

Hey Nicolas,

Fair points all.

The primary assertion that I wanted to rebut was that a certain level of performance is impossible without doping which seems to be where the science of sport guys are heading.

I, like you, recognize the draw to doping in a professional sport like cycling and realize that as long as there is a decent amount of money to be made (& a large gap in that money between team leader & domestique), there will be folks who compromise their ethics. I just hope that we continue to root these folks out in the lab rather than a) in the court of public opinion or even worse, b) via 'experts' deeming their performance physiologically impossible.

1. Regarding recency of doping: Very true, doping, as a whole, has been around for a long time & was no doubt more rampant before the mid-60's when it was outlawed. I guess I draw a bit of a distinction between stimulants/recovery aids & methods to increase O2 transport (i.e. EPO & Blood transfusion), which are more recent & presumably what the Science of Sport guys are talking about when they set non-doping limits on O2 uptake.

2. Doping in other sports: Sounds as though XC skiing may not be the best example :-) What about East African runners? Despite being head and shoulders above the field (& with performance levels indicative of similar O2 demand), Geb was not subject to the same suspicion as Lance. Again, not saying Lance didn't dope & not discounting that cycling has a long history of doping, just that high levels of performance alone don't 'prove' doping.

3. Is what I have a problem with. Without question, Lance trained differently, dare I say more intelligently, than the traditional means of preparation for Tour racing. He had been training at a high level for his entire life. Is it not at least possible that the x% difference between Lance & the rest of the field came from differences in preparation as from drugs? To be frank, based on what I have seen, the training programs of some cyclists even at the highest level are ad-hoc, poorly organized & leave a lot to be desired.

4) True but what of the circumstantial evidence that points to innocence? The years of negative tests. I find it hard to believe that he was doping to any level that had a significant impact on performance over all of those years and yet never had a true positive? Think about the number of tests both domestic & international. The best that they could come up with was 1 highly suspect 99 sample? He certainly wasn't at his best in 99.

Given the history of doping in cycling & the confirmed doping of his peers, it certainly wouldn't surprise me if some hard evidence of Lance doping did surface. However, I would suggest that we don't write the guy off purely on the basis of him being very good.

In the same way that the 'scientists' who deemed the 4min mile a physiological impossibility wound up looking a little silly, the SOS guys have a dose of humility heading their way when some guy comes out of nowhere & rides up Alpe D'Huez at 7W/kg.

I look on the notion of 'physiological impossibilities' more sceptically than just about anything else :-)



Nicolas said...

Hey Alan,

I agree that talking of impossibility is not the right way to approach it. However, it seems very very unlikely at some point that a performance is not doping-enhanced anymore. Judging by VAM data such as the Alpe d'Huez climb times that SOS quote the difference between the "high doping" (late 90s, early 2000s) and the "more likely to be clean times (more recently) are very significant (by the looks of things about 2min) and Armstrong went another minute faster. As I mentioned I find the size of the difference staggering, making me less convinced of it being possible.

Regarding your point that he trained smarter: what evidence do we have for this? Yes, there is a lot of folklore around how his cancer transformed him from a heavy powerful rider into a much leaner, more efficient rider, but this was debunked quite convincingly by a sports scientist whose name I forget. He showed that the claim of progression in power and power to weight from before to after the cancer was based on very poor measurement.

The fact that he hasn't tested positive is not surprising in my view: Armstrong was extremely professional in his preparation and, if this included doping, I would have expected that this meant taking necessary precautions. There were, however, suspicious incidents such as when he denied access to doping testers after a ride in his comeback preparations.

In the end we all get swayed in our argumentation by what we believe. I don't particularly like or hate Armstrong (probably a bit more of a dislike because of some bullying he has shown), but it seems to me that the majority of people put on blinders and argue the direction "oh I would really like to believe that he was clean so I will accept all circumstantial evidence in his favor and deny all circumstantial evidence against him, however overwhelming one might seem."

I am looking at the situation (vast superiority over people who already had a strong performance boost from PEDs; huge leap in performance after a debilitating disease that in all likelihood should have huge negative performance consequences; highly consistent testimony from multiple parties, including very loyal friends (Hamilton, Hincapie); tests identified retroactively as positive but unusable for legal reasons, close association with "doping pope" etc.) and find very little on the plus side (didn't fail tests officially - but then Ullrich didn't either wrt to Puerto).

I equally believe Ullrich was doped, Pantani, etc, etc. C'est la vie...

Alan Couzens said...

Hey Nicolas,
I get your perspective & recognize the circumstantial 'evidence' that might point towards a rider of that level doping.

I'm not sure what Tyler & Floyd et al. have to gain from naming Lance. Their accusations are certainly worthy of consideration/investigation but banning/premature conviction…?

With an average test rate in the latter stages of his career (the highest performance period) of ~2 unannounced tests per month, how does one pull off a successful doping operation? Considering RBC’s have a half life of ~55 days, the math doesn’t add up for me.

The only physical evidence that I know of is the B sample from 99 and, while I’m not generally supportive of folks arguing against evidence through mishandling, it’s hard to ignore in this case – 6 years post &, as Vrijman concluded the samples were simply not anonymous, they were numbered in accordance with finish place!

While I hope for the sake of the sport that Lance is innocent, that’s not my dog in this fight. My biggest problem is with the assumption that x level of performance = doping. Doping is a short cut and, to be certain, allows some folks to jump ahead to a fitness level that they don’t deserve. However, I don’t see it as setting a firm performance ceiling. In the same way that Phelps was beating folks with superior suits and the same way that we are starting to see swimming WR’s from the suit era attacked, it is not impossible for a (very special) non-assisted athlete with superior preparation to beat the assisted.

Was this Lance? The circumstantial evidence for this is also compelling. At 16, he was going head to head with the guy who would become the greatest triathlete in history – Mark Allen. So his genetics/training response is likely pretty unique. USPS were one of the first cycling teams to focus the season on a single goal – complete with things that were, up to that point, unheard of such as course reconnaissance! While other teams were busy filling their calendars with as much racing as possible to ‘toughen up the riders’ while simultaneously keeping the sponsors happy, Lance was methodically training for one event. What is this singular focus worth? What does an athlete lose with a (traditional) ad-hoc program of ‘racing themselves fit’ with event after event including multiple stage races? Heck, what is a power meter worth? How many athlete were training with power in the late 90’s/early 00’s?

Do these items account for the relative benefit of a 50 vs 45 difference in Haematocrit? Maybe yes, maybe no but at the very least we would have to concede that, given Lance did things a little differently, it is a possibility. For this reason, despite the negative side of the circumstantial evidence, I’ll continue to remain an optimist, continue to believe in the pure value of hard, intelligent work in sport & I’ll wait for some hard evidence to surface before convicting in my own mind.