Understanding and Pushing the Limits of the Elo Rating Algorithm

Discussion of chess software programming and technical issues.

Moderators: bob, hgm, Harvey Williamson

Forum rules
This textbox is used to restore diagrams posted with the [d] tag before the upgrade.
Post Reply

Michel
Posts: 2050
Joined: Sun Sep 28, 2008 11:50 pm

Re: Understanding and Pushing the Limits of the Elo Rating Algorithm

Post by Michel » Wed Oct 16, 2019 7:01 am

Daniel Shawul wrote:
Tue Oct 15, 2019 8:14 pm
Let me try again: Paired comparison with ties
You should probably contact them so that they can at least refer to your work.

But I think there is essentially nothing in the paper. As you say k-elo seems to Davidson.

The only thing that could be considered slightly original is that they identify the draw model corresponding to the pseudo likelihood approach to the elo model with draws counting as the average of a win and a loss (i.e. the contribution of a draw to the log (pseudo) likelihood is counted as (1/2)log(w(1-w)) where w is a function of the elo difference). But this is a triviality.
Ideas=science. Simplification=engineering.
Without ideas there is nothing to simplify.

User avatar
towforce
Posts: 9298
Joined: Wed Mar 08, 2006 11:57 pm
Location: Birmingham UK

Re: Understanding and Pushing the Limits of the Elo Rating Algorithm

Post by towforce » Fri Oct 18, 2019 9:29 pm

Just my opinion, but it seems to me that:

1. optimal chess is a drawn game

2. therefore, at a certain level of skill, all games will be drawn

3. therefore, any rating system that doesn't allow for this is fundamentally flawed
Love of truth is the best defence against ideological possession.

Fulvio
Posts: 155
Joined: Fri Aug 12, 2016 6:43 pm

Re: Understanding and Pushing the Limits of the Elo Rating Algorithm

Post by Fulvio » Sat Oct 19, 2019 12:48 pm

Michel wrote:
Tue Oct 15, 2019 8:30 am
https://arxiv.org/abs/1910.06081

Nothing that wasn't already known unfortunately.
tl;dr

Based on the ELO difference, the current algorithm gives the following probability of winning:
+0 --> 50%
+100 --> 64%
+200 --> 76%
+300 --> 85%
+400 --> 91%

What are values of the improved algorithm?

Dann Corbit
Posts: 10124
Joined: Wed Mar 08, 2006 7:57 pm
Location: Redmond, WA USA
Contact:

Re: Understanding and Pushing the Limits of the Elo Rating Algorithm

Post by Dann Corbit » Mon Oct 21, 2019 6:41 pm

A while ago, Kaggle had a contest for improved game outcome prediction algorithms.
Uri did pretty well in the contest, if I remember correctly.
Taking ideas is not a vice, it is a virtue. We have another word for this. It is called learning.
But sharing ideas is an even greater virtue. We have another word for this. It is called teaching.

Post Reply