Grand tours queen stages, MTF, TT's seem to be a battle of the two greatest forces a cyclist overcomes, gravity for climbing, and wind resistance for TT's. I've personally had enough of the diets, reinvented equipment, and transforming illness stories in cycling and don't see why you need to add peripheral factors into your models when all the athletes are riding almost identical equipment.
I basically agree on MTFs that if you account for wind and drafting (and wet heavy clothes) then the rest is mostly an argument of distraction. TT position is still too big of an unknown.
Do any of your w/kg prediction models take rolling resistance into account? Teams that pay higher attention to such things will be seeing faster climbing speeds for the same power but as your models look at speed would this will be interpreted as a higher w/kg prediction rather than getting more "bang for buck" from the same power?
the coefficient for rolling resistance is standardized. fortunately since the power needed to overcome rolling resistance is small and speed dependent it would take a huge difference in tires to create significant error in the power estimates.
I remember at this year’s Tour, David Lopez turned up at the Tour. At the table, he was eating Nutella. Because he was relatively new to the team, he had no idea of how that would offend his teammates. But it did offend them. They just didn’t like it, as in, “That stuff is no good for you. We’re here to eat the right food, and you bringing that to the table is lowering the dietary and nutritional standard.” I just thought that kind of attitude was impressive. That’s what Sky have created, and I found it impressive. Maybe the other teams are doing it better, but from the bits and pieces I heard, it doesn’t seem that all of them are.
“Berglund B, Hemmingsson P, Birgegard G. Detection of autologous blood transfusions in cross-country skiers. Int J Sports Med1987;8:66–70.”—does that mean it was considered enough of a problem back in 1987 that researchers were already looking for ways to detect it ?
“In addition, a variety of rumours and innuendo suggested that at least some endurance athletes were using this technique in an effort to gain a competitive advantage in international competition.13–15 Thus, the term “blood doping” was coined. Although it is clear that blood doping improves performance, it is unclear how widespread it was in the 1970s and 1980s as detection was difficult because athletes received a reinfusion of their own red blood cells.”—BJSM randomness saved on the Macintoshphone
Rephrase: Sorry. Given the weirdly random factors of real world bike-ride conditions, how can race data (as used to estimate GT podium VO2max) indicate anything in comparison to lab data? Also, haven't aren't you worried about poisoning your conclusions with bias and bad methods (ala Ioannidis JPA (2005) Why Most Published Research Findings Are False. PLoS Med 2(8): e124.)?
one way but not the other yeah.
so right the idea that i’m coming to is that
for cycling a functional VO2 max should probably be defined
as what can be hit at a relevant time in a race
that lab conditions
shouldn’t be set up to max out VO2 kinetics
to reproduce what is likely seen in the real world
so yes forcing the fit of race data into a lab derived model
may always be square peg wrong whole
but the opposite
doesn’t necessarily have to be true
which as far as i know
my attempt is one of the few to even try to work out
what’s going on metabolically in a large field derived data set
any one of the data points might be complete junk
but on the whole
maybe this approach is inching a bit closer to a less wrong
Nice math; thanks. Problem (maybe it's the beer/wine): U say VO2's maybe a lab condition. Wait: how U measure any aspect of race performance in-lab? Dolomites, good luck putting those in lab, with weathers 'n all. That hill Sagan paperboyed in Tirreno? Or a guy bein' 42, got no respec' yet (I know)? Ex-teammates? And people who ride know for no reason they go fast one day by surprise and slow one day when all they done was train. Data from races? Also: haven't you read Ioannidis?
i am so trying to understand what the hell you are saying but i’m not
W/kg =44.24 * (1/vClimb) + 1.90 - 0.00035*vClimb + 3.95 * air density
so why this equations
so this equation for now looks like a good candidate
to cover the time period 10-50 ish minutes
where grand tours are won or lost
because it has a structure that at least plausibly aproximates
the metabolic components
that make up the total power
at any given time
as a caveat though
its worth noting that the OLS used to derive the coefficients
find 1/vClimb to be statistically significant
but i kept it in anyways
i “know” its there
because there has got to be some anaerobic contribution
to the effort
even if it is tiny
which it appears to be
given the noisiness of the data set
i think its just getting lost statistically due to the noise
that being said
i do find it interesting just how little the anaerobic component
(which i can show is tiny by manually turning it up and breaking the curve as in the latest video)
though it does make sense considering
1. the climbs even short ones are at the end of stages so that even at the start of the climb many matches have already been burned.
2. the shortest climbs are likely being stretched by the fighting for position and coming into the climb already near full boil before the timing even begins
more surprising though is that if you set the 50 ish minute effort at 85% of VO2 max
(which doesn’t seem to be much challenged)
the equation suggests that riders may not holding VO2 max on even the shortest 6-7 min climbs included despite lab studies showing that VO2 max can be held for between 3 - 12 minutes
instead if you anchor 50 ish minute effort at 85% then the best that the riders are sustaining is about 95 % VO2 max
which may simply be that the short climbs are effectively stretched due to tactics as mentioned above
B (the much more interesting possibility)
the VO2 max is an artificial lab condition specific phenomenon
you can only really hit VO2 max with a protocol designed to put you into the state which is dependent on being very fresh (ie having the reserve to go very deep) and not having some other central governor in play
is supported by the reverse ramp protocol study which showed that the reverse ramp protocol generated higher VO2 max numbers
may help explain why VO2 max (besides just the issue of differing efficiency) is a not so stellar predictor of racing success
instead of worrying about producing a “max” value
what we should be looking for in performance prediction
so that instead of looking at a max value
what VO2 is sustainable for various durations
if this is true of course
why bother with VO2 at all
since VO2 * efficiency just = power anyways
especially the functional VO2 profile isn’t tightly tethered enough
make VO2 max meaningful?
although on the otherhand
identifying why the untethering occurs
may shed some light on a new target for training techniques
if you are already hitting high fractions in your profile
target something else in your training
if your profile is low compared to VO2 max figuring out how to bring it to the VO2 max suggested potential will likely be one of the best ways to unlock watts
on the doping front what might be interesting is how O2 vector doping affects the VO2 profile as compared to training adaptations
my speculation would be
doping moves VO2 profile AND VO2 max up proportionally
while training moves up the VO2 profile more than VO2 max
What's your view on moving evidence for doping using the ABP and its successors from statistic to forensic? Instead of using p>0.01 as the limit the likelihood of doping. It gives room for doping within limits. An adverse analytical finding could (also) be formulated to what is known in professional experience. Wording could be something in the line of "without a doubt in professional experience this is an adverse analytical finding...
whether you are running the statistics on a computer or in your head its still an attempt at probability. there is no such thing as “without a doubt” the advantage of the professional opinion is that humans tend to be good at pattern recognition and can incorporate a huge number of parameters. the down side is that experience tends to overweight the extreme outcomes. for example you may go through an intersection a thousand times but only remember the time you got in an accident. humans also have a tendency to project meaning into chaos such as watching the clouds and finding ones that look like certain things.
either way whether its a computer or a human interpreter you still have to establish test the sensitivity specificity and prevalence so that you know what your positive predictive value is.
first one who can get me the relavant citiation for the normal reticulocyte distribution for an individual wins a case of email can tecate, the question to answer is normal or multimodal distribution ?
quick notes to myself on the direction of the metabolic component model
to use p = reserve/time + CP to best define the true reserve then it would make sense to use time points where VO2 max can be sustained for the entire duration, such as 1 min and 5 ish min. since the the times are closely spaced probably a couple of trials would be necessary to achieve a consistent reserve calculation (speculating)
then once reserve is established hold it constant and use it to define the dynamic CP by recalculating at multiple time points 5 min, 10 min, 30 min etc
then this should show you just how the fraction of VO2max falls off at various durations.
describe that shape with an equation, substitute it for CP and cap the whole equation at Pmax and you should have a reliable model for a much broader range of times
Dr. Veloclinic, Many of your readers don't really understand that the bio-passport could work okay if the sports federations were willing to open cases. We know for sure that the sports federations simply do not open cases even when there is the positive sitting there in the APMU. Sadly, the lack of dopers being caught is being explained as some fault of WADA, not the sports federations. :( This sugests the new leader of the IOC will enable more doping...
maybe but I’m also starting to think that some of it is that any biopassport case brought against a rider could be ripped apart
i could do it myself
the mill stone around the bioppassports neck is the 2 year (now 4 year ban) which in either case is effectively a sporting death sentence requires a proportional degree of certainty
the level of certainty the 99.9% that is claimed is not there in a blood parameter only model of indirect dope detection.
this is not to diminish the work put into the biopassport or the value but i honestly believe that that level of certainty is not there based on the available published research studies
Biopassport version 2 is desperately needed for new cases to be brought
Project idea: how much will effective testing cost, then can pro cycling afford it? Set an optimal schedule of testing then price it, including passport analysis by humans, and multiply by the number of world tour riders, then a reduced schedule for conti teams and see what we get.
the WADA study already did this and came up with 30,000 and I think that is easily on the low end as far as sensitivity you get with current approaches.
There will always be doping in sports. And tests will always be a couple steps behind the newest wonder drug or procedure. This "new clean era" concept is a complete BS. We have seen it before. There will never be a "new clean era". Human nature does not change. So what is the point on spending so much effort trying to elucidate all the nuances of doping? Let them have it. Just move on and do something more meaningful. Go for a ride, for example.
I’ve been riding a lot lately actually in the about 14 ish hours per week, knocked out 2 MTB 50 milers already, won 1 and got thrashed in the other, all in prep for 24 solo at old pueblo in Feb. I say this only because your attempt at making a point is stupid, trite, and hackneyed. I can ride my bike all day long, but like doping, the act only has meaning in what we project onto it.
You can give Belkin a lot of crap for that (and they deserve it), but what about Ibarguren and Lefevre at Quickstep? What about Leinders at sky? Movistar paying Valverde while he was suspended? Mantova for lampre? Ferrari and Liquigas, where Kreuziger and Nibali worked with. Point is: Rabo had a past with dope and they're only the latest to be found out.
“Rolling out over the rolling hills, rolling out from Hamilton. A sunnier day than a fall day should be in the fall in Hamilton. But clouds were never a day away never more than a day away never out here in Hamilton. The country roads, except for townie cops and town line limit changes, were worth it for the drive alone regardless of the distance. We decided to stay off the god awful road, the main road up through Utica. Utica, wasn’t it, that the the town, the fuck at this point if I could even remember.”—
“Lance deceived everybody on the planet, us included, so obviously we wanted to believe it also - that he was winning the Tours clean,” O’Grady said. ”We’re all athletes out there suffering through the mountains and you’d like to think he was just training harder and working harder than we were.”—it’s like dope era cyclists can’t move their lips without saying some stupid
“I tried this technique on myself in the past few weeks: I noticed that lowering the head the full diaphragmatic expiration is easier, and if you ride at high RPM (like Froome… ), each exhalation is accompanied by a 15-20% increase in instantaneous watts (better venous return? Less myofascial tension?).”—Ferrari on Froome
continuing the what does zero vuelta positives mean discussion
The above figure visually illustrates how the volume expansion that naturally occurs during grand tours masks the effect of blood transfusion.
The baseline (blue) and transfusion (red) values were taken directly from Damsgaard (2006) where researchers transfused subjects with 800 ml of packed red blood cells.
Then assuming that a transfusion was timed to coincide with the volume expansion in the first week of a grand tour, I adjusted the values down by 10% to show what the transfusion might look like during a grand tour (orange).
What does this mean for the Biopassport which requires outlier values?
The volume expansion effectively blinds the biopassport software so that it can’t see the peak in the Hgb concentration or the peak in any of the combined parameters that include Hgb such as the OFF score.
This masking effect on Hgb concentration then leaves just reticulocyte count to possibly be flagged by the software.
Again using the numbers from Damsgaard (2006) the pattern that may be visually apparent is unlikely to flag the passport software as it would likely not cross any thresholds.
Now to be absolutely fair to the biopassport, the study data is averaged data from several subjects. Averaging the data is likely to cause some smoothing that may take out some of the more extreme values for any one individual.
On the other hand, the blood volume used is large and given in one transfusion. The study also doesn’t consider the effect of masking doses of EPO that could be used to keep the reticulocyte count from dropping as much. One of the issues not often discussed is that since the natural physiological level of EPO is only about half of the traditional Ferrrari “micro” dose the masking doses of EPO may still be well below the levels detectable by the current EPO tests.
Taken together, it would be hard to argue that the 9% sensitivity that Morkeberg (2011) showed for transfusions of 1 fresh blood bag using the athlete biological passport (or up to 32% for 3 blood bags) is truly applicable to catching this lower level of doping in a grand tour. In the grand tour context, the sensitivity for the transfusion of 1-2 blood bags may be close to zero.
Damsgaard R, Munch T, Mørkeberg J, Mortensen SP, González-Alonso J. Effects of
blood withdrawal and reinfusion on biomarkers of erythropoiesis in humans:
Implications for anti-doping strategies. Haematologica. 2006 Jul;91(7):1006-8.
Mørkeberg J, Sharpe K, Belhage B, Damsgaard R, Schmidt W, Prommer N, Gore CJ,
Ashenden MJ. Detecting autologous blood transfusions: a comparison of three
passport approaches and four blood markers. Scand J Med Sci Sports. 2011
the following study transfused either 1 or 3 bags of fresh or frozen blood
and used several methods
including the Athlete Passport
Mørkeberg J, Sharpe K, Belhage B, Damsgaard R, Schmidt W, Prommer N, Gore CJ, Ashenden MJ. Detecting autologous blood transfusions: a comparison of three passport approaches and four blood markers. Scand J Med Sci Sports. 2011 Apr;21(2):235-43.
to test how well the transfusions could be detected
the percent of transfusions detected
Hgb and OFF sore
3 bags frozen:
8% and 13%
3 bags fresh:
17% and 32%
1 bag frozen:
13% and 17%
1 bag fresh:
0 and 9%
not entirely terrible right?
until you start to throw in the 10% volume expansion of a Grand tour
consider how the volume expansion
what would be seen following the transfusion
to get actual numbers for the Hgb and Retic
we need to go back to an older paper from Damsgaard