ELO rating inflation

[Sean Evans]

I wonder if this article would also apply to the SSDF and computer chess in general?

http://chessbase.com/newsdetail.asp?newsid=5608

Cordially,

Sean

[Laskos]

I wonder if this article would also apply to the SSDF and computer chess in general?

http://chessbase.com/newsdetail.asp?newsid=5608

Cordially,

Sean

Sonas is probably right. I also investigated the problem, and came to the conclusion that the players reaching the rating floor can deliver inflation or deflation in rating. If he has data that the players reaching the rating floor are overrated, then the trend is inflationary. If they are underrated then the trend is deflationary. I do not have this data, so I couldn't feed my computerized rating simulation with new entrants. The bigger the K-factor, the larger is the influence of the new entrants, and these new entrants are a key factor. But be careful, if all new entrants are like Magnus Carlsen, the the trend will be heavily deflationary .

This paragraph synthesize the Sonas' article

Now let's think about how it was back when the rating floor was 2200. Consider a hypothetical group of active players, all of whom have a performance rating of 2000 across all their games. Some of those players will certainly outperform their true 2000-strength for a short time, and others will underperform. Only those players from our group that outperform their true strength will make it onto the rating list, whereas the players who underperform will not be anywhere on the list. This means the players who show up on the rating list just above the rating floor, are (as a group) significantly overrated, just waiting to donate rating points to the rest of the pool. Even worse, while these overrated players keep temporary possession of their 2200+ ratings, other players may also receive inflated initial ratings as well, based partially on games against the overrated players. Over time, the overrated players will do worse than their ratings suggest, and their excess rating points will ultimately be distributed throughout the entire rating pool.

If that is correct, then the trend is inflationary.

Kai

ps I do not know about computer chess ratings, as many new engines are Magnus Carlsen - like, and many are not. Besides that, in the computer chess there is no K-factor, just Elostat and BayesElo.

[towforce]

I wonder if this article would also apply to the SSDF and computer chess in general? http://chessbase.com/newsdetail.asp?newsid=5608

I find myself to be in disagreement with this article by Jeff Sonas. Of course the availability of computer training and analysis is going to result in a surge of chess skill - and this is what has caused human ratings to rise so rapidly.

What he wrote about automated analysis by checking whether GMs choose the same moves as computers is nonsense: the way to automatically check is to look for tactical errors - the more of these there are (from which a recovery is not made - which could indicate a brilliancy rather than an error), the lower a player's rating will be.

[Sean Evans]

I did enquire a few years back with Hyatt why his Crafty program ELO had decreased over time rather than increased, which one would expect with better hardware and software updates.

Hyatt's response was that more computer opponents play on ICC, which have caused a decrease in ELO, i.e. higher ELO opponents lowers the your ELO score.

This would seem to dispute the argument that human ELO is increasing due to players have increased ELO strength.

Cordially,

Sean

[towforce]

I did enquire a few years back with Hyatt why his Crafty program ELO had decreased over time rather than increased, which one would expect with better hardware and software updates. Hyatt's response was that more computer opponents play on ICC, which have caused a decrease in ELO, i.e. higher ELO opponents lowers the your ELO score. This would seem to dispute the argument that human ELO is increasing due to players have increased ELO strength.

A stronger computer getting a lower rating would indicate rating deflation rather than rating inflation.

[Norm Pollock]

Elo systems have their weaknesses as seen with inflation/deflation.

Another system, and imho, a more mathematical system of player ratings is the "Chessmetric" system.

http://db.chessmetrics.com/CM2/Formulas.asp?Params=

[Laskos]

I did enquire a few years back with Hyatt why his Crafty program ELO had decreased over time rather than increased, which one would expect with better hardware and software updates. Hyatt's response was that more computer opponents play on ICC, which have caused a decrease in ELO, i.e. higher ELO opponents lowers the your ELO score. This would seem to dispute the argument that human ELO is increasing due to players have increased ELO strength.

A stronger computer getting a lower rating would indicate rating deflation rather than rating inflation.

I think that's a problem of chess servers allowing engines to participate, when the new computer entrant has an arbitrary initial Elo of 1600 or something, but the real rating is 3000+, with some K-factors. In this case, each new engine entrant contributes to rating deflation. I do not know if the new entrants affect SSDF, CCRL, CEGT, etc., as they use completely different rating schemes.

Kai

[Sean Evans]

I did enquire a few years back with Hyatt why his Crafty program ELO had decreased over time rather than increased, which one would expect with better hardware and software updates. Hyatt's response was that more computer opponents play on ICC, which have caused a decrease in ELO, i.e. higher ELO opponents lowers the your ELO score. This would seem to dispute the argument that human ELO is increasing due to players have increased ELO strength.

A stronger computer getting a lower rating would indicate rating deflation rather than rating inflation.

Incorrect, an even stronger computer getting a higher rating, would indicate inflation. Depends on your perspective -:)

[towforce]

I did enquire a few years back with Hyatt why his Crafty program ELO had decreased over time rather than increased, which one would expect with better hardware and software updates. Hyatt's response was that more computer opponents play on ICC, which have caused a decrease in ELO, i.e. higher ELO opponents lowers the your ELO score. This would seem to dispute the argument that human ELO is increasing due to players have increased ELO strength.

A stronger computer getting a lower rating would indicate rating deflation rather than rating inflation.

Incorrect, an even stronger computer getting a higher rating, would indicate inflation. Depends on your perspective -:)

Eh? An "even stronger computer" would be expected to get a higher rating, wouldn't it?

[Laskos]

A stronger computer getting a lower rating would indicate rating deflation rather than rating inflation.

Incorrect, an even stronger computer getting a higher rating, would indicate inflation. Depends on your perspective -:)

Eh? An "even stronger computer" would be expected to get a higher rating, wouldn't it?

And if it enters with an arbitrary Elo 1600 and eats 2500 players' points, being really 3000+, whose points does it eat, Sean? Is it an inflationary factor or deflationary? It just increases its points and deflates all other.

Kai