But you do raise an important question: "how many are elite BECAUSE they have doped?"
A follow-up question is "how can we answer that question with reasonable certainty, or without reasonable uncertainties?"
Many have asked the question, and many have assumed the answer, but has anyone provided an answer without reasonable uncertainties?
Can we answer it with short term studies on non-elites, producing non-elite times? Not without risks that come from assumptions and extrapolations and projections onto elite athletes.
Can we answer it with "anecdotes", like Boulami, Kiprop, Jeptoo, etc.? Not without addressing any likely significant confounders, not to mention details about which times are actually doped, and which are not, and the conditions and tactics specific to that time.
Can we answer it for men's distance events, with anecdotes of cycling, race-walking, cross-country skiing, or women taking steroids? We shouldn't have to.
Can we conclude it from historical evidence, like 3 out of the top 30 were "caught", or 2 out of the top 16, or 2 out of the top 100? Looks like there is a lot of room for interpretation of the remaining 27 out of 30, or 14 out of 16, or 98 out of 100.
We should note that Russia doped a lot, and only had limited narrow success with women and race-walking. If one were to attempt to draw correlations to doping and performance, Russia offers us good data.
When we do have rare glimpses of doping prevalence by country, e.g. the IAAF leaked database with high blood values, we see that East African countries like Kenya and Ethiopia did not particularly stand out, compared to countries like Russia and Morocco. When you rank countries by high blood values, and then rank them by performance, you obtain two very different rankings, suggesting correlation between blood doping and distance event performance is low. (NOTE I don't even speak of causation, but say that the correlation is low).
Regarding your WADA statistics, you should be more careful not to confuse different measures:
- The WADA statistic of 1-2% is of samples that test positive, and not a ratio of athletes testing positive. Consider the doped athlete with 10 tests, and finally 1 sample tests positive, and how that impacts the ratio of samples testing positive versus a true doping prevalence. If he is in a pool with 9 other clean athletes, each with 10 results, the doping prevalence is 10%, while samples that tested positive is 1%.
That claim simply begs the question as to how many are elite because they have doped. In an environment where doping is an identifiable issue it must follow that some of the elite will be doped. But we won't know who. As WADA (that now rather unreliable organisation) attests, only 1% of athletes test positive as against an estimated 15-40% who are actually doping. No country - and certainly not Kenya - can boast of its elites without inviting the question of how many are clean. Russia has afforded salutary caution against the illusory emptiness of such claims of excellence.