And yet, M2 was sometimes lower, and sometimes higher than the "minimal estimate" M1. Strange...
In any case, you are still not recognizing the elephant in the room:
Well yes, fine, but I was not comparing the minimal estimate from the samples to the minimal estimate for the athletes, but to the real doping prevalence. As pointed out before, the real prevalence is a lot higher because all the dopers not pushing their Hb beyond the threshold remain unflagged here.
Referring to the example in the introduction to use same real numbers:
"If a blood sample is collected from 200 of these athletes, between 1 and 9 of them (4 on average) should present a value higher than 164 g/L. If 30 of these athletes presented a value higher than 164 g/L, then between 21 (11%) and 29 (15%) presented a value that was too high."
So anyone doping from 150 to 160 g/L (to use their units) would not count as a doper here. Anyone not tested while going from 150 to 170 g/L (e.g., because of poor OOC testing in Kenya at the time, or because of a quiet doorbell) would not count here either.
And exactly that is the main problem of the ABP (aside from the demonstrated corruption), as demonstrated in the Ashenden Eur J Appl Physiol 2011 paper, where the athletes doped - unflagged - from an average of 155 to 164 g/L.
Finally. We only know that the real numbers are significantly higher.
Oh, and running was also worse than cycling in 2001 and 2002 (about 18 to 12%), before the cat-and-mouse games began.
Btw, Kiprop and Jeptoo got away with their blood manipulations for quite a while before slipping up. So did Poistogova etc. That happened also to Ullrich and Landis, and Armstrong actually, so there is a comparable amount of sophistication from that point of view.