Non-rated games already don't have any impact on players' ratings, which is as it should be. Likewise, they should have no impact on W-L-D records.
What if the current #2 player wanted to coach some players here? Perhaps he would want to play some practice games against them--maybe giving them odds, or setting up endgame scenarios, etc., games that he was "supposed" to lose to the trainee. He might be discouraged from doing that if he likes seeing that he has 0 wins.
Perhaps he goes ahead and loses 30 unrated games. He doesn't care about his record, just his rating. But others on the site observe his record and say, "He's got a great rating, but he's beatable, and he's a lot worse than some lower-rated players, who have a lot fewer losses."
Including non-rated games in W-L-D records distorts the various presentations of statistics, confuses people who seek information from the records, discourages teaching and practice, and punishes people who might want to play "friendly" games that might be used to attract people to the site or to chess in general.
The W-L-D statistics are on very little value.
You can easily get a 100%-0%-0% statistics just by playing far inferior players than yourself.
Or you can have a 0%-100%-0% statistics just by playing far superior players than yourself.
So the only thing that can discover a members true skill (as far as possible) is by looking at his rating, and look at his green graph. If his graph is stable, then his rating is reliable. The W-L-D statistics doesn't say much.
I agree with FabianFnas that the W-L-D record requires context to understand and cannot be a true comparison, and that the rating reigns supreme.
Nonetheless, taking the non-rated games out of the W-L-D record makes it more statistically useful. The rating would still be the best measure, but an observer could look at the W-L-D record in context and get even more useful information about the player's history, style, and ability--with no confusion from non-rated games.
What about the ability to just filter the rated/unrated games from your WLD records? I believe the more ways to view your own data, the more you can learn from it.
Originally posted by HolyT Non-rated games already don't have any impact on players' ratings, which is as it should be. Likewise, they should have no impact on W-L-D records.
What if the current #2 player wanted to coach some players here? Perhaps he would want to play some practice games against them--maybe giving them odds, or setting up endgame scenarios, etc., games that he was ...[text shortened]... ndly" games that might be used to attract people to the site or to chess in general.
I told the admins about this after playing my first unrated game at RHP... I ran a few 'tutorials' on mating with low material... setting myself up as the one in a losing position.
Once I realized they counted as lost games, I stopped doing it because the admins didn't do anything to fix the problem they admitted they had.
Unrated games should not count against the record.
Originally posted by tomasino What about the ability to just filter the rated/unrated games from your WLD records? I believe the more ways to view your own data, the more you can learn from it.
Thank you again for the recs. The simplest thing would be just not to count the games in the W-L-D record. Extra filters or options would be great, too--anything that makes available information clearer and less misleading.
I told the admins about this months ago and went back and forth on e-mail. Eventually, they said, "That's a good idea" but nothing came of it. I realize I have less clout as a non-sub (6 games is a lot to me!) and can't submit this for a vote. Where do we go from here? Thanks.
02 Jan '08 23:17
Back to Top