Casual obsever was right that I ignored the "eliteness" of the Kenyan subjects, falsely accusing them of being sub-elite on the basis of their 9:00 time trials. No longer ignoring it, thinking this through a little more, and trying to put it in perspective, the "eliteness" of the Kenyan subjects makes a complete mockery of this study and any conclusions that these improvements were significantly the result of EPO. I remind everyone that the conclusion from 2013, and again in 2018, is that the Scots and Kenyans showed similar improvements, ~5%, after the EPO interventions, and this is the basis for "calling out CANOVA over study confirming EPO works on Kenyan runner". Putting aside whether the difference between 5.7% and 4.6% is statistically insignificant, we have a comparison between the Scots who tried (RPE = 18-19) and the Kenyans took the time trials less seriously (RPE=14.6-16). If Aragon says 1 RPE is about 10 heartbeats, my personal experience is that is around 20 seconds per kilometer increase in pace. In any case the 1.5% RPE increase was about 26 seconds (roughly the same without considering heartbeats). This means, if the Kenyans tried harder, we are talking somewhere in the ballpark of the Kenyans running 1 minute faster than they did, if they had tried just as hard as the Scots. I've never been confident in RPE being objective, so how can we put these efforts in a different perspective? What if we use Daniels VDOT to determine equivalency? Recall we are talking about 1:03/2:12 runners performing 3000m time trials, on EPO, about as fast as Mary Cain's high school record of 8:58, as an 18 year old. Simple VDOT calculations means that 1:03/2:12 runners should be capable of running 3000m under 8:00. Further VDOT equivalencies: The baseline time trial was 9:22, or marathon pace, or 29 sec/km slower than 3K pace The first EPO time trial was just under 8:58, or half-marathon pace, or 21 sec/km slower than 3K pace The fastest Kenyan in all the time trials ran around 8:30, or 10K pace, or 13sec/km slower than 3K pace The 4.6% difference for the Kenyans was a 26 second improvement, or less than 9 sec/km. Recall, the fastest Kenyan, around 8:30 in the EPO time trial, should be able to run sub-8:00 CLEAN, or 6.25% faster than the fastest EPO time trial. I think it is fair to say that these borderline elite athletes could have easily improved their 3000m time trial performance, from marathon pace to half-marathon pace, not to mention 3K pace, without any EPO. The best of these borderline elite athletes should have been capable of another 6.25% improvement faster than the fastest EPO time trial, 8:30, COMPLETELY CLEAN. In other words, the study showed a 4.6% improvement for athletes running at 85-90% of their CLEAN potential, rather than 100-104.6% of their CLEAN potential. At the very least this underscores the need to conduct a better study, before we can say "Take that CANOVA! Scientists confirm EPO works on top Kenyans too!"
Aragon wrote:
Not to mentiom that your summary leaves the impression that elite runners were improving their elite times in an "all in" type effort, there is even a clear error, because the group wasn't on average 1.03/2.12 runners, but those whose personal record was full/half marathon had on average those times.
...
The key conclusion from my part is this. Any change in guys doing submaximal and relatively easy efforts can be as well motivation than something else because the key limitator isn't physiological.