Yes, I remember seeing this data in another thread a few months back. While it's impressive, it only illustrates part of the story.
I was actually making a statement about 5000m running in two eras, the 70's compared to the 90's, and didn't differentiate between African and non African athletes. Africans were competing at the highest level in the 70's. Rekrunner's data fails to take into consideration that far fewer Europeans were going in to T&F by the 90's, for various and somewhat complicated reasons, so just focusing on that group isn't really useful for the point I was making. I don't think it can be refuted that T&F in Europe was far more popular in the 70's than it was in the 90's!
Moreover, a pool of just the top 5 is a very small sample, and doesn't take into consideration outlier talents.
I did a similar analysis for the top 10 performers (not performances, which would have made the point even more clearly) in the 1970s and the 1990's:
Av. 1970's (top 10, African and non-African) - 13:13.02;
Av 1990's (top 10, African and non-African) - 12:50.61.
That's 23 seconds, or about 180m in distance.
The argument you made (or supported) was that drug use was rampant (something about the 'Wild West') in the 70's and 80's. And if Rekrunner's data claims, everyone was on steroids and blood transfusions. Which is somewhat ironic, because apart from a quote about US athletes generally (nothing specifically about male middle and distance runners) and a Biathlete, he is doing exactly what he criticises others for doing; using non specific quotes and using them to make a leap of faith that 800m and 5000m runners must have all been doping in the 70's and 80's. Now I know he doesn't say that specifically, but the notes after each decade, e.g.' unlimited blood transfusions, steroids virtually undetectable', imply that quite strongly.
So, if the 5000m lists from the 70's are meant to reflect steroid and blood doping use, and the 90's lists reflect steroid and EPO use, then we have to ask why the average time improved by 23 seconds. Had we taken top 10 performances, then it would have been closer to 30 seconds. We should also remember that the rate of progression over time should slow down.
The improvement in tracks and spikes might count for 5 or 6 seconds?
So what about the remaining 17secs?
If steroid use is a constant (which some imply, but which I disagree with for middle and distance male runners), then we can only summise that the vast improvement shows EPO is far more effective than blood doping for improving 5000m ability!?
The other alternative is that the 1970 data was not affected as much by doping. The men running then were amateurs, there was little money in the sport, they wouldn't have got into the sport as kids in the 60's thinking about big money and careers that would set them up for life. By the 90's there was big money to be made, all the more so by those from poor countries.
If you look at the top 10 non-African performers in each decade, then the gap is much closer and more closely in line with the rate of historical progression one would expect to see during such a time scale.
Av. non-African top 10 over 5000m:
1970's - 13:14.16;
1990's - 13:02.11.
N.B. The fastest 5k non-African runner from the 90's was convicted of doping (Baumann).
That's an improvement of 12 secs. Subtract 5 or 6 secs for improved tracks and spikes, and the improvement looks more like 6 or 7 secs.
I've had a look at the top 10 non-African men over 800m since the 1960's, taking just the best time for each, and then taken the top 10 times/performances from non-African men over the same period.
The first thing I noticed, as I went through all the athletes' names until I reached the No. 10 non-African, was just how much longer it took to reach that point with each succeeding decade. While that wasn't much of a surprise to me or probably anyone here, the sheer number of Africans that were ahead of the 10th non-African increased enormously in the 90's and 00's.
I don't have the time to type up all the athletes (African as well as non-African) at present, I will give an overview of what I found.
1960's - only 1 African born athlete (Kiprugut) in the top 11. The top 10 guys were: - Snell, Ryun, Doubell, Kemper, Bell, Dyce, Plachy, Farrell, Sandison & Crothers. Interestingly only Snell had a 2nd performance (1:45.1) that made it into the top 10 times.
Top 10 men - average time = 1:45.02
Top 10 performances - average time = 1:44.97
1970's - Only 2 Africans (Boit & Kipkurgat) in the top 12. Of the top 10 non-Africans - Coe, Wohlhuter, Juantorena, Fiasconaro, Beyer, Van Damme, Marajo, Susanj, Ovett & Wottle - only El Caballo has more than 1 time that appears in the top 10 times, with 4 performances!
Top 10 men - average time = 1:43.69
Top 10 performances - average time = 1:43.53
1980's - There are 3 Africans that make it in to the top 13, with Koskei at No.3, Konchellah at No.6 and Ereng at No.7.
The top 10 non-Africans are: - Coe, Cruz, Gray, Cram, Barbosa, Mack, P. Elliott, Wuyke, Druppers and Jones.
Interestingly, the top 10 times are provided by just 4 athletes (Cruz, Gray, Coe & Cram), with half, 5 of them, coming from Cruz, and 3 from Gray. This does distort the stats somewhat and makes for a bigger disparity between top 10 men and top 10 times.
Top 10 men - average time = 1:42.97
Top 10 performances - average time = 1:42.44
1990's - This is where there is a big sea change in the 800m event. Before reaching the 10th fastest non-African man over 800m, I had to get to No.25. So, there were 15 Africans and 10 non-Africans in the top 25. The other thing that struck me was the sheer quantity of performances from the African athletes compared to the non-Africans, and that's not just from Kipketer. I counted 96 performances from African born athletes that were quicker than the 10th fastest European, Martin Steele at 1:43.84. That works out at c. 6 times/performances (on average) for each of the 15 African runners. In contrast, only Rodal (with 3) and Gray (with 2) amongst the non-Africans, set more than 1 time that made it into the top 10 non-African performances.
The top 10 non-Africans were: - Rodal, Gray, Tellez, Bucher, P. Elliott, Barbosa, Everett, Kenah, Douglas and Steele.
Top 10 men - average time = 1:43.13
Top 10 performances - average time = 1:43.00
2000's - We have to go as far as the No.19 fastest man to reach the 10th fastest non-African. In other words, there are 9 Africans and 10 non-Africans, so there is more of a balance than compared to the previous decade.
The top 10 non-Africans were: - Borzakovskiy, Bucher, Lopez, Baala, Czaplewski, Som, Milkevics? K. Robinson, Reed and Longo.
The top 10 times set from the above were provided by just 5 athletes, and 2 of those (Borza & Bucher) provided 7 of them!
Top 10 men - average time = 1:43.27
Top 10 performances - average time = 1:42.92
2010's - Like the 1990's, I needed to go as far as the 25th fastest man from this decade (well 8 years) before reaching the 10th fastest non-African. Also, like the 90's, there were some 97 performances from 15 African athletes that were better than the 10th fastest non-African. In contrast, there were 20 performances from the 10 non-African athletes that were faster than the 10th fastest non-African athlete, Lewandowski at 1:43.721
The top 10 fastest non-Africans were: - Tuka, Bosse, Soloman, Murphy, Symmonds, Kszczot, Berian, Brazier, Lopez and Lewandowski. Of those, Bosse, Soloman and Symmonds all produced 2 performances that made it into the top 10 fastest times by non-Africans.
Another standout is the fact that half of the non-Africans are from the US. That's a greater number than in any of the previous 5 decades!
Top 10 men average time = 1:43.14
Top 10 times average time = 1:43.01
In concluding, I think the stats show that there has really been very little change in the depth of standards over 800m since the end of the 70's. If we look at the average of the top 10 performers (top 10 men):
1960s: - 1:45.02
1970s: - 1:43.69
1980s: - 1:42.97
1990s: - 1:43.13
2000s: - 1:43.27
2010s: - 1:43.14
NB, some of the performances from the 60's would have been on cinders, so if that were taken into consideration, then possibly 0.5 secs could come off the average.
These stats show that over the last 4 decades, the average time for the best 10 non-African men over 800m, has fluctuated by just 0.3sec.
If we now look at the fastest 10 performances/times run by non-African men in each decade we see a similar pattern.
1960s: - 1:44.97
1970s: - 1:43.53
1980s: - 1:42.44
1990s: - 1:43.00
2000s: - 1:42.92
2010s: - 1:43.01
We see that for non-African athletes, the average of their best times set over the past 3 decades have differed by just 0.09secs.
The one slight anomaly from the stats above is the 1980's being on average about 0.5sec faster than the decades that followed. This can partly be explained by the fact that half of the ten fastest times were set by just one person, Cruz. But, when you then compare to the previous table (that of fastest 10 performers), one can see that there is actually little change in the 80's to the overall trend.
If a similar analysis of African performers and performances was carried out, then I'm sure there would be a greater drop in average times viz a viz non-Africans.
Of course, that is partly due to socio-economic issues in Africa since the 60's, just like there has definitely been a drop off in participation in T&F in non-African nations since the 1980's; although this is not so obvious from the data I've produced.
For me personally, the difference in the 800m lists from the 80's to 90's is the most marked. There was a definite great African emergence in the 90's, and the sheer number of performances in those lists would indicate that there was likely something happening apart from socio-economic influences. I find it difficult to look beyond EPO as the most likely cause of this.
I would suggest that the data above for the average fastest 10 performers from a non-African background over 2 laps; showing very similar results for each of the last 4 decades; indicates that the 800m is probably, either the least affected distance from PED use, or the event where drug use has the least impact on performance.
It could also be argued that the stats provide some evidence that drug testing outside of Africa has been successful over the past 40 years. I can't recall a single name of those athletes listed in the non-African ranks that have been banned for drugs. There were a few names I came across in the lists from African countries that have served a ban, and I think it is a universally acknowledged fact that whatever one thinks about the robustness of testing over the last 40 years, there has certainly been more testing in place for athletes outside of Africa than inside.