CCRL live lists with 100 Elo reduction

[Adam Hair]

I don't understand those ELO reductions.

CEGT ratings were about -50 ELO comparing to CCRL. This difference was not too big.

Now CCRL ratings are 100 ELO lower and CEGT ratings 200 ELO lower, so the difference between CCRL and CEGT increased to 150 ELO.

IMO it would be better to select ONE very known engine with large amount of games, give it constant ELO and keep it the same on all ranking lists.

My suggestion would be for all the different ratings agencies to get together and agree on a constant ratings with one starting point, otherwise how do we know which agency rating system reflect the most accurate ratings

How do you judge which is more accurate?

[pichy]

I don't understand those ELO reductions.

CEGT ratings were about -50 ELO comparing to CCRL. This difference was not too big.

Now CCRL ratings are 100 ELO lower and CEGT ratings 200 ELO lower, so the difference between CCRL and CEGT increased to 150 ELO.

IMO it would be better to select ONE very known engine with large amount of games, give it constant ELO and keep it the same on all ranking lists.

My suggestion would be for all the different ratings agencies to get together and agree on a constant ratings with one starting point, otherwise how do we know which agency rating system reflect the most accurate ratings

How do you judge which is more accurate?

And how do you judge which is more accurate now

[Adam Hair]

I don't understand those ELO reductions.

CEGT ratings were about -50 ELO comparing to CCRL. This difference was not too big.

Now CCRL ratings are 100 ELO lower and CEGT ratings 200 ELO lower, so the difference between CCRL and CEGT increased to 150 ELO.

IMO it would be better to select ONE very known engine with large amount of games, give it constant ELO and keep it the same on all ranking lists.

My suggestion would be for all the different ratings agencies to get together and agree on a constant ratings with one starting point, otherwise how do we know which agency rating system reflect the most accurate ratings

How do you judge which is more accurate?

And how do you judge which is more accurate now

Let me ask this in a slightly different manner:

How would you, Jorge, judge which agency's ratings were more accurate, if they used the same starting point to compute the ratings? I am not questioning whether you could or not. Rather, I am asking about the method you would use.

[Piotr Cichy]

I don't understand those ELO reductions.

CEGT ratings were about -50 ELO comparing to CCRL. This difference was not too big.

Now CCRL ratings are 100 ELO lower and CEGT ratings 200 ELO lower, so the difference between CCRL and CEGT increased to 150 ELO.

IMO it would be better to select ONE very known engine with large amount of games, give it constant ELO and keep it the same on all ranking lists.

My suggestion would be for all the different ratings agencies to get together and agree on a constant ratings with one starting point, otherwise how do we know which agency rating system reflect the most accurate ratings

How do you judge which is more accurate?

And how do you judge which is more accurate now

Let me ask this in a slightly different manner:

How would you, Jorge, judge which agency's ratings were more accurate, if they used the same starting point to compute the ratings? I am not questioning whether you could or not. Rather, I am asking about the method you would use.

What impact has starting point for rating accuracy? I thought, we can add a constant to each player's rating and the accuracy will be still the same.

According to Graham, the reason for ELO reduction was not the accuracy, but too high ELO ratings comparing to other lists. So the goal is to have the most similiar ratings in different lists, isn't it?

[Adam Hair]

What impact has starting point for rating accuracy? I thought, we can add a constant to each player's rating and the accuracy will be still the same.

You and I both know that the starting point has no relevance on statistical accuracy. I am thinking Jorge is referring to comparing engines to humans, but I am not certain.

According to Graham, the reason for ELO reduction was not the accuracy, but too high ELO ratings comparing to other lists. So the goal is to have the most similiar ratings in different lists, isn't it?

In the CCRL, it was suggested that our ratings were too high, and a majority of the members agreed. The lowering was definitely not to improve statistical accuracy. As you know, the only way to do that is to reduce sources of error and increase the number of games.