Posted: 19 Apr 2006, 10:16
Interesting. So there is significant disagreement on Holmes, Marciano, Dempsey, Lewis, Holyfield, Jeffries, Langford, Tunney, Wills. 5 out of these 9 fighters are from before 1930, where video evidence is limited and when the color line was preventing significant matchups. 2 (3 if you count Homes) are recent fighters whose historical standing is difficult to assess at this stage.Decagon wrote:1. Ali 2-1-2-2 Average: 1.75 Standard deviation: 0.500
2. Louis 1-2-1-3 Average: 1.75Standard deviation: 0.957
3. Holmes 3-3-4-10 Average: 5.00 Standard deviation: 3.37
4. Foreman 4-7-7-4 Average: 5.50 Standard deviation: 1.73
5. Frazier 9-5-8-5 Average: 6.75 Standard deviation: 2.06
6. Johnson 8-4-9-7 Average: 7.00 Standard deviation: 2.16
7. Marciano 14-9-5-1 Average: 7.25 Standard deviation: 5.56
8. Liston 10-6-6-9 Average: 7.75Standard deviation: 2.06
9. Dempsey 13-11-3-6 Average: 8.25 Standard deviation: 4.75
10. Tyson 11-12-10-8 Average: 10.2 Standard deviation: 1.71
11. Lewis 5-10-15-16 Average: 11.5 Standard deviation: 5.07
12. Holyfield 16-8-14-18 Average: 18.0 Standard deviation: 4.32
13. Jeffries 20-15-11-11 Average: 14.2 Standard deviation: 4.27
14. Langford 6-25-12-14 Average: 14.2 Standard deviation: 7.93
15. Charles 19-16-13-12 Average: 15.0 Standard deviation: 3.16
16. Tunney 7-13-22-21 Average: 15.8 Standard deviation: 7.09
17. Walcott 18-17-16-13 Average: 16.5 Standard deviation: 2.16
18. Patterson 15-22-18-15 Average:16.5 Standard deviation: 3.32
19. Schmeling 17-19-19-17 Average: 18.0 Standard deviation: 1.15
20. Wills 12-29-17-22 Average: 19.5 Standard deviation: 7.25
The smaller the standard deviation, the more we agree on that fighter. The larger the standard deviation, the less we agree.