Originally Posted by MattMattMatt
It would be more informative to see the results presented graphically. It's pretty easy to do, just select all and copy and paste into excel. A couple of mins cleaning up, then sort them in descending order and plot both fighters stats on the same chart. You can see at a glance if one fighter has a consistently better record, if they have only a couple of better wins, or if they have a dozens of meaningless record padding victories.
Obviously there are cases where the ratings don't reflect the strength of the opposition very well, since scores are reduced by lay-offs (which don't affect everyone equally), questionable decisions are still scored against the losing fighter (e.g. Pac), questionable victories are still credited to the winner (e.g. Bradley), DQ wins in fights that you were losing are still credited too highly (Bowe against Golota), beating a fighter that has significantly declined but has yet to have that 'exposed', or is fighting too far from their natural weight results in too much credit to the victor (which is then propagated throughout the whole system).
All of that evens out better than many biased human assessments. Boxrec is a reasonable starting point if you know and understand it's limitations.
A good example from that list is DaVarryl Williamson being rated higher than David Tua. There is no way that any sane boxing expert would ever say that Williamson is a better victory for Wlad than Tua is for Lennox. In fact, if Wlad had that win over Tua I think many people would rate that as one of, if not the, best wins for Wlad. The Rahman/McCall victories are a strange situation, I don't feel that it is appropriate to give Lennox more credit for beating someone that only has a high rating because he beat you by upset KO in the previous fight. In reality we have to take losses into account as well then to even those cases out.
Nice post. The part I highlighted about Rahman is what jumped out at me while viewing the list. It is definitely a strange situation. But the more I think about it, it actually seems OK.
Lewis still ended up being down 193 points, even after he KO'd him quicker in the very next fight.
In terms of comparing it to Wlads wins and giving Lewis credit. The fact is Lewis KO'd a guy in 4 rounds who was good enough to KO him in 5 rounds. Sounds really strange but it's true.
To try and change it would belittle Rahmans achievement in beating the top guy in Lennox Lewis. And/or belittle Lewis's achievement of destroying a guy who previously bested him.
On another matter Wlad has Byrd, Peter and Thompson all on the list twice which is also strange. I'm not sure what to think of that, except to say I'd prefer fighters not to be repeated. It instinctively seems better to have 15 different guys, but it may make no difference?
To clarify I believe Byrd, Peter and Thompson should be on the list twice, I just wish they wasn't.