Many people have been asking me tricky questions about the ratings system. It is not just a random number which I pull from thin air that determines your strength, but rather a complex statistical formula. And for those of you who can’t even pronounce what I just wrote, then you need not read on.

First of all some history. About 100 years ago a Professor named Arpad Elo decided that it would be nice if everyone could compare their strength without using only tournament results. A ranking system seemed logical and he accordingly developed the ELO rating system. To this day the system is being refined again and again, but the main aim still remains the same.

i.e. provide an accurate comparison of each players’ playing strength in the same ‘pool’.

A ‘pool’ of players is a group that competes against each other.

The higher your rating the better player you are, based on past results. Even the best chess player must sometimes make a mistake and lose to a weaker player, and for that reason nobody is ever expected to win 100% of the time. As an indication of what the higher rated player would be expected to score here are some figures.

Rating Difference Expected Score
0-3 50%
47-53 57%
99-106 64%
146-153 70%
198-206 76%
291-302 85%
392-411 92%
over 735 100%

Alternatively if you don’t have a rating, but want to know what it would be, here is how to work it out. If you can beat a player 50% of the time then you would get the same rating as your opponent.

Score Rating
99% +677
75% +193
60% +72
50% same
40% -72
25% -193
1% -677

After a tournament your rating changes based on the score that you were expected to achieve compared to the score you actually achieved. So your rating can go either up or down
at the end of a tournament or remain the same.