draw causes lose in rating

Sort:
FancyShoesVlogs
Why is it when I draw, I lose points? I thought drawing was a tie! This is some BS
Martin_Stahl
FancyShoesVlogs wrote:
Why is it when I draw, I lose points? I thought drawing was a tie! This is some BS

 

Rating Change depends on the results along with the rating and rating deviation (RD a measure of rating uncertainty).  The rating formula also takes into account expected outcomes. So, you can lose rating for draws 

Blinknone

If you are the same rating as your opponent, a draw will not affect either rating. 

But if there is a difference in rating, the higher-rated player will lose rating points if he draws against a lower-rated player.  And the lower-rated player will gain rating points. 

Randomdude640

Random question but hypothitcly what would happen if a grandmaster drawed aganist a 200 rated player

tlay80
Randomdude640 wrote:

Random question but hypothitcly what would happen if a grandmaster drawed aganist a 200 rated player

Assuing both players had previously played a lot of games (and without a long gap since playing them), the grandmaster would lose 8 points, and the 200 rated player would gain 8. If either of those players' ratings was based on only a few games (or if the games were a long time ago), that player's rating change would be magnified, perhaps considerably (maybe up to 50-100 points -- I forget). I believe the more established player's change wouldn't be affected (still 8), but it's possible the change would be slightly diminished if their opponent's rating were less reliable.

Someone do correct me if I have that wrong, but I think it's right. At any rate, it's pretty close.

VenemousViper
Randomdude640 wrote:

Random question but hypothitcly what would happen if a grandmaster drawed aganist a 200 rated player

There's an article about rating changes like that that'll give you a formula. But I expect it is not all proportional and it stops at some point, so I would say -20.

VenemousViper
tlay80 wrote:
Randomdude640 wrote:

Random question but hypothitcly what would happen if a grandmaster drawed aganist a 200 rated player

Assuing both players had previously played a lot of games (and without a long gap since playing them), the grandmaster would lose 8 points, and the 200 rated player would gain 8. If either of those players' ratings was based on only a few games (or if the games were a long time ago), that player's rating change would be magnified, perhaps considerably (maybe up to 50-100 points -- I forget). I believe the more established player's change wouldn't be affected (still 8), but it's possible the change would be slightly diminished if their opponent's rating were less reliable.

Someone do correct me if I have that wrong, but I think it's right. At any rate, it's pretty close.

No no no that is not right. You see, when one player wins, if that player is rated much lower than opponent, then chess.com thinks "Wow, that player is good, he managed to beat someone rated 300 points higher than him!" and gives him lots of points. What you described is true, but it would not be what would happen in this case.

tlay80
EnPassantAvalanche wrote:
tlay80 wrote:
Randomdude640 wrote:

Random question but hypothitcly what would happen if a grandmaster drawed aganist a 200 rated player

Assuing both players had previously played a lot of games (and without a long gap since playing them), the grandmaster would lose 8 points, and the 200 rated player would gain 8. If either of those players' ratings was based on only a few games (or if the games were a long time ago), that player's rating change would be magnified, perhaps considerably (maybe up to 50-100 points -- I forget). I believe the more established player's change wouldn't be affected (still 8), but it's possible the change would be slightly diminished if their opponent's rating were less reliable.

Someone do correct me if I have that wrong, but I think it's right. At any rate, it's pretty close.

No no no that is not right. You see, when one player wins, if that player is rated much lower than opponent, then chess.com thinks "Wow, that player is good, he managed to beat someone rated 300 points higher than him!" and gives him lots of points. What you described is true, but it would not be what would happen in this case.

Okay, why not? Where am I wrong? Happy to be corrected.