I thought my choice of 11 different events, from the 1500m to the marathon, men and women, would be representative enough.
Nevertheless I gave the exact "method" I used, so everyone can feel free to confirm my numbers, or evaluate other events.
Generally, I was afraid that the picture with EPO in the 800m is not so clear because of the bigger role steroids might play.
At quick glance, the men's 800m would have about 5 athletes (all East African from 4 countries) meeting the cutoff (around 1:42.3), progressing maybe 0.5%, with only two athletes beating Seb Coe's 1981 world record.
In the women's 800m, 5 of the top 6 times are pre-EPO, in the 1980's. The only post-1990 result would be one performance from Pamela Jelimo.
Either way, EPO doesn't appear to have made much impact for the 800m since 1990.
For the women's 3000m steeple, I don't find any results before 1996, so can't generate a pre-1990 cutoff.
I think the 800m needs to be factored in. Like I mentioned, the 800 is an endurance event and VO2max is an important component of that distance. You're right that 5 out of the 6 fastest times are pre-1990 but 3 out of those 5 and 6 out of the top 10 all-time best are runners from Soviet block countries. In these cases you can't rule out blood doping as it was being used extensively with the Soviet block nations in the 70s & 80s. In addition, there have been several top female 800m runners who tested positive for EPO (e.g., Ceplak...who has the 18th fastest time). Maybe Ceplak and the others who tested hot for EPO were also using steroids - but we don't know conclusively since their positive was for EPO only. I would thing though the EPO/blood doping & androgens combo might be the perferred strategy for those 800m women who were doping.