I wanted to address repeated criticisms that, despite my own appeals to "statistical analysis of real performances", I have never shown any data based on real performances.
How often do we see these kind of EPO related claims:
- After the availability of synthetic EPO, distance running world records dropped dramatically.
- EPO can have a large effect for "high-responders" on individual performance, up to 3-6%, or more.
- EPO was widely used, up to 57% according to recent surveys.
- Notably, Morocco and Spain are among the best examples of EPO-abuse.
If these claims are accurate, we should be able to observe the large effect from EPO on the best performances, pre-1990, versus post-1990. It may not be the 3-6% individual improvement, but nevertheless, we should be able to see some sort of "EPO footprint", showing how the "game changed", resulting from a drug often described as a "game-changer".
With that in mind, I decided to make simple, easy to understand, comparisons of pre-EPO performances to post-EPO performances, to better understand how performances have changed, where performances have changed, and curiously, where performances have not changed as much.
I thought it would be interesting to break down the performances by ethnic heritage. Why?
- If EPO works in East Africans the same as it works "on everyone else", it is interesting to break out East Africa separately, and observe any "sameness".
- Morocco and Spain are considered among the best examples of abusers and beneficiaries of EPO, so it also makes sense to look at these populations separately.
We should also be able to see evidence of "works the same on everyone" improvements, in East Africans, in North Africans, featuring the EPO-doped Moroccans, and in the rest of the world, featuring the EPO-doped Spain.
All of the raw data comes from listings at "www.alltime-athletics.com" as of 15 Jan 2018, including performances that were annulled due to doping.
I counted "performers" and not "performances". If an athlete has multiple performances, I count only the best one.
I chose the cut-off as the average of the top-5 performers before 1 Jan 1990, as the massive drop in times started after 1990.
From this cut-off, I calculated two simple measures:
1) Quantity: A simple count of how many athletes performed better than the pre-1990 top-5 average.
2) Quality: Improvement, as a percentage of the pre-1990 cut-off, of the average of the post-1990 top-5 performers, if 5 existed, or the average of all of them if less than 5 existed.
For the men, I analyzed six events: 1500m, 3000m, 3000m steeple, 5000m, 10000m, and marathon.
In addition to global improvements, I repeated the same measures for specific ethnic sub-groups.
I collected separate numbers for 1) East Africans, 2) North Africans, and then 3) "everyone else", with special mention of Spanish athletes.
NOTE: Africans imported to other countries were counted as East Africans or North Africans, depending on their heritage, and not citizenship.
NOTE: South Africans did not factor much, but when they did, they are counted as "5 continents".
For the women, I analyzed five events: 1500m, 3000m, 5000m, 10000m, and marathon.
I collected separate numbers for 1) East Africans, 2) Chinese, and then 3) "everyone else". Moroccans and North Africans did not seem to be a factor for the women. Russia could also be of interest, but I did not break them out.