How to determine all time best?

[curious]

I was having a discussion with someone about how to determine the best alltime performer in a sport.

Should it be based on # of major titles/championships, best performance, consistency, or something else. I think track lends itself to this kind of discussion.
Cheers

[Jon]

Whatever criterium you use, the debate could go on for ever and ever and ever and ever and ever and ever, etc.....

[Guest]

Why would you use a bicycle race to determine the best track athlete?

[Hunter]

i've thought about this quite a bit. There are a lot of different variables, especially when you consider athletes now vs 80 years ago.

I think the best way is to look at consistency over time. Who was ranked #1 or in the top 3 in their event the most times?

I think we can find these answers somewhere on this site.
H

[schmke]

I compiled a spreadsheet with the top-10 800-10000 runners since 1968 and with a few equations arrived at a ranking of the best. The equation was based on the number of times ranked in the top-10 minus the average ranking. Thus the best by this definition would be someone who was ranked in the top-10 at different distances, ideally #1.

Using this definition the top 10 (well 11, 10th is a tie) are as follows:

Geb - 20.2
Aouita - 14.1
Coe - 14.0
Morceli - 13.2
Komen - 9.3
El G - 9.3
Ovett - 9.0
Tergat - 8.3
Cram - 7.7
Yifter - 7.0
Ondieki - 7.0

Geb's being #1 is due to being ranked in the top-10 at 3 different distances for a total of 22 top-10 rankings and an average ranking of 1.8. Aouita is close behind with an average of 1.9 but was only ranked in the top-10 16 times. What is impressive about Aouita is that he was ranked in the top-10 in a 5 distances.

Kevin

[Guest]

For my project I saw no reason to reinvent the wheel. I took T&FN's world ranking criteria and adapted it to an all-time setting.

http://personal.bgsu.edu/~jsquire/intro.pdf