|
|
|
Author
| Message |
|---|
|
|
| appalachia wrote: i never said the "rating system here is terrible"... |
I know. That's why I was asking GileCAR, by name, and why he was responding. If you'd read the earlier posts in the thread and read my post more carefully, this would have been obvious.
|
|
|
Well, I think now that 'k' rating should be even bigger :)
About ratings don't telling me anything: let's say 1700 player can put up a real fight, while 1800 can blunder a piece in first 10 moves. There is no "rule". When playing on FICS, I know by the opponents rating, how much trouble is he going to make, and ratings are good indicator of the game outcome. Here, I just don't see that.
Reason for this is people playing many games at the same time ... or not paying enough time on their games ... or what ever ... |
That doesn't actually make any difference. Ratings measure performance, rather than strength. |
I think that makes all the difference. I cannot prove it, or argue that it does, but that is only explanation I can find for "weird rating behavior". I think it's like having same rating system for blitz and standard games (some players play blitz better than others and standard time games worse than others , and vice versa, and in my case there is over 500 points difference between the two), and this might be creating rating inflatioin (I'm just guessing, not saying it really does). Because ratings measure performance and not strenght, you cannot know anything about opponents strenght by his rating (he might got his rating by play only 10days/all games)...
Just to clear things up... I'm not saying this site is terrible (I think it's pretty good, otherwise I would be playing here:) ). I don't think that there should be radical changes to the rating system. I don't think it's easy to tune the rating system to be more telling ... All I'm saying is that rating points here don't tell me almost anything about the opponent, and I don't understand how other people think 200 rating points difference here tells something.
|
|
|
If I don't know about a player's rating, I look at the games he has played on this site. That tells me a great deal about how strong his game is.
|
|
|
I accept what you're saying, GileCAR but I think the explanation is this. Some players are consistent and some players are not. Some players with, say, an 1800 rating play consistent chess: OK but nothing spectacular. Some players with an 1800 rating play pretty strong chess in general but quite often blunder a piece through carelessness (caused by playing too many games at once and not spending enough time on each). So, You might have a 50-50 chance of beating the consistent player based on the chess he plays. On the other hand, the inconsistent player has, say, a 20% chance of blundering a piece (in which case, you're `guaranteed' to win) but, in the 80% of games where he doesn't do that, he has a 63% chance of winning. Overall, you still have a fifty-fifty chance of winning but you have to work harder for the win if he doesn't blunder in your game.
The rating difference is still a reasonable measure of your chance of winning the game. It's just not a good measure of how difficult the game will be (i.e., how strong your opponent is).
|
|
|
The rating difference is still a reasonable measure of your chance of winning the game. It's just not a good measure of how difficult the game will be (i.e., how strong your opponent is). |
Well... you might be getting somewhere with this... There is one thing I didn't mention in my posts, and that is the "feel of the game" ... how much difficult game will be... how much trouble I'll have in the opening and so on... On FICS by opponents rating I can know that... here that's not the case ... and probably that is the main reason why I don't have impression the ratings are meaningful here. Thanks for the explanation :)
|
| Previous 1 2 3 4 |
|
|
|
