HRE,
Let me clarify one thing here. I'm not doubting (here and now) that EPO can be effective for slower runners, especially if they are poorly trained, but really just looking at athletes who have the potential to move their own national records.
I'm genuinely interested in why people say what they say, and why they believe what they believe.
Wipe your mind clean of all pre-judgement, and ask yourself, why do I believe that EPO works for elite runners?
I'm less interested in the ultimate truth of the belief, as to the arguments and examples as a result of this thought process.
Don't dwell so much on what can or cannot be conclusively proved as caused by EPO (except where something measurable, like race times, can put an upper bound on the realized effectiveness).
I wasn't trying to make a point so much as wanting to see someone positively defend the idea that EPO can work, or has worked, "at the top", without assuming the conclusion.
Renato says it doesn't work for top Kenyan athletes. Whatever you think about that, my race times observation is that it generally hasn't worked for top non-East Africans either, because, based on national and area record times, the number of potential candidates, over several decades, that could have cheated to surpass pre-EPO times, can be counted on two hands, and maybe two feet.
Whatever effect EPO has in studies, on amateur athletes, or BBC reporters, as a general rule, the best times of non-African nations have remained relatively static.
This places some kind of upper limit that says EPO has not worked at a time when there was no risk of getting caught taking it. I find that kind of odd, when we compare it to cycling, we don't find the same parallel.
Let me make an analogy (you might have already seen), and consider a fat-burning weight loss pill.
If I give it to an out of shape 250 pound couch potato, he might lose 75 pounds.
If I give it to a "well-trained" runner of 160 pounds, he might lose 15 pounds.
Now someone like Renato says, I believe this kind of weight loss pill will not work on top-Kenyan athletes who weigh 110 pounds.
Then letslose.com says "what?" Kenyans are humans too -- of course this drug will react on all humans -- Kenyans are no different. Some Kenyans even got caught taking these pills. Who knows what they would have weighed without it -- they've been taking it all along!
This is how I'm thinking of EPO -- it doesn't work at the top well-trained runners because there is very little margin left to improve.
Here's another exercise:
In various articles discussing the problem of who to give Lance's yellow jerseys to, there were all kinds of analysis that most of the cyclists themselves were doped. They were hard pressed to find anyone in the top 10 finishers who weren't already implicated in some kind of scandal.
Now look at any running event 1500m and above, men or women. Select the top-100 performances of all time. Tell me, as a percentage, how many of these performances have been linked to EPO, or any drug, for that matter, and compare that to cycling during the "EPO-era".