See my above response to Norm. Chessmetrics does not do my calculation sequence - it doesn't try to compare everyone at once. It makes the assumption that the #3-#20 ranked players in the world have a constant Elo. (Unless I am totally misunderstanding that sentence, but I don't think I am..)
As for your point about rating staying constant over time, restricting the set of games for each player to maybe a 15 year window in the peak of their career would probably fix this, yet still allow the calculation to be done, although it probably can be wider than that for most players that stayed active for a while, like Lasker.
Finally, about Chessbase being able to do it, I didn't know that. It's a little too expensive for my grad student self right now, but maybe someday... BayesElo wasn't quite working for me either with such a big database, but I'll give it another go.
Best,
Pete
I think that the assumption that there is a 15 year window that player strength is constant is a wrong assumption.
The best way to use results to compare between players of different times is to allow some non deterministic computer chess program to play in tournaments
(when the program is weak enough not to beat everybody(when it may be designed to get rating of 2300 or 2500).
The program should be designed to have rating that is not dependent on the computer and it may always search 100,000 nodes per second(regardless of hardware and if the hardware is faster it is going to do nothing more time).
You are making an assumption that any database out there is filtered and without errors, and is a fair representation of each player's career.
Errors (wrong names, wrong moves) and near-duplicate games are everywhere. Plus games that were friendly exhibitions or blitz or rapid or playoff or handicap are often not indicated as such.
You are making an assumption that any database out there is filtered and without errors, and is a fair representation of each player's career.
Errors (wrong names, wrong moves) and near-duplicate games are everywhere. Plus games that were friendly exhibitions or blitz or rapid or playoff or handicap are often not indicated as such.
-Norm
Norm,
I totally agree. For arguments sake, say there was an easy way to clean and filter the DB...
Uri Blass wrote:
I think that the assumption that there is a 15 year window that player strength is constant is a wrong assumption.
The best way to use results to compare between players of different times is to allow some non deterministic computer chess program to play in tournaments
(when the program is weak enough not to beat everybody(when it may be designed to get rating of 2300 or 2500).
The program should be designed to have rating that is not dependent on the computer and it may always search 100,000 nodes per second(regardless of hardware and if the hardware is faster it is going to do nothing more time).
Uri
Uri,
For more recent players, you could even use shorter windows (like 5 years). For players from the 19th century, there might be some small distortion, but I doubt it would be a big deal. I don't see why Lasker's ability would have changed much between ages 25 and 40.. Indeed, he was WC for much longer than 15 years. But even for him, 5 years might be enough.
As for the computer in tourney idea, unfortunately they weren't running Fritz .00001 on Babbage's Analytical Engine back in the 1880's
