Originally posted by greenpawn34Hmm, OK, but then you're just going to create massive rating inflation. 1200s will be gaining +32 points all over the place just because they were lucky enough to get matched with a cheater in a tournament, etc.
Not if you used his grade at the time the game was played.
The program will not alter the cheats grade.
OK this will benefit some players who played the cheat after he
cheated his way to over 2000 but the main thing is everyone
will get a rating increase. Even those he beat legit.
In fact, it becomes like a lottery, where about the best thing you can do in terms of your rating is deliberately try to get matched up with a cheater.
I can see the motivation behind your idea, but in practice it's a little more complex than you're acknowledging.
I think you are missing the point of the ratings. They are not like points you collect. the are a measure of who your peers in chess are. Ratings are easily manipulated to increase of decrease that "score", but the fact remains that you play at a certain level regardless of the number. Look at all the talk of sand baggers. My point is quit worrying about a few points here and there. That said I would like ratings to be calculated from the beginning of the game, not the end.
quote:
".....because they were lucky enough to get matched with a
cheater in a tournament,"
What is lucky about playing a game of chess in which you have
no chance of winning. Some have paid money to get a clean game
of chess. They have been cheated....CHEATED.
They must get the result reversed and the rating points.
Originally posted by cheshirecatstevensIf you count a 1200 as beating a 2200 just because the 2200 was cheating, that totally undermines the function of the rating as "a measure of who your peers are".
I think you are missing the point of the ratings. They are not like points you collect. the are a measure of who your peers in chess are. Ratings are easily manipulated to increase of decrease that "score", but the fact remains that you play at a certain level regardless of the number. Look at all the talk of sand baggers. My point is quit worryin ...[text shortened]... . That said I would like ratings to be calculated from the beginning of the game, not the end.
The fact is, the 2200 cheater is still playing at a 2200 level, even if they're not doing it legitimately. If you count everything against them as wins, all you do is massively distort the purpose of the rating.
Like I say, I can see the spirit behind the suggestion; I just don't think there's any way to implement it that makes sense on a practical level.
But it doesn't matter much, because I seriously doubt that Russ would ever implement a complex and unnecessary change like this. He seems to have enough on his plate with regard to changes that are clearly beneficial (conditional moves, etc.)
Originally posted by Tyrannosauruschexsay you got +16 points winning against a cheat. in the recalculation, you'd end up with 2,1 or even 0 points for that particular game. that would leave you with 14-16 lower rating after all is recalculated. double that for tournament and clan games.
You can only ever lose points for a win if you have a provisional rating.
for the top players, who have played against cheats the most, it would bring quite a drop for every recalculation crunch. surely that's a far worse scenario than not adjusting the points after a ban.
Originally posted by cheshirecatstevensI should be working now - you lot are going get me the sack.
I think you are missing the point of the ratings. They are not like points you collect. the are a measure of who your peers in chess are. Ratings are easily manipulated to increase of decrease that "score", but the fact remains that you play at a certain level regardless of the number. Look at all the talk of sand baggers. My point is quit worryin ...[text shortened]... . That said I would like ratings to be calculated from the beginning of the game, not the end.
Of course I know what ratings are and I also know the grades
on this site mean absolutely nothing. Grades should only apply
to OTB games - this is where chess is played. eyeball to eyeball.
That is true measure of your playing strength.
Not sitting in your bedroom reading opening books and moving
the pieces about.
But rating points are used on this site. Some people take
them seriously. So be it.
The victim's grade has been lowered by a cheat.
Why cannot it be reversed? (or at the very least void the game
so the victim does not lose anything.)
Why are we arguing about this?
Originally posted by greenpawn34introduces ratings inflation, by adding free points to the system. which would affect everyone, as the additional points would be redistributed to your future opponents. the amount of free points could potentially be huge, as cheaters tend to ramp up huge game loads and play them at fast pace without losing much.
say you got +16 points winning against a cheat....
[b] WINS AGINST A CHEAT DO NOT RE-CALCULATED
WINS AGINST A CHEAT DO NOT RE-CALCULATED
WINS AGINST A CHEAT DO NOT RE-CALCULATED
WINS AGINST A CHEAT DO NOT RE-CALCULATED[/b]
Originally posted by wormwoodActually the presence of cheats causes ratings to be artificially lower. Probably the optimal solution would be to simply erase ALL changes to anybody's rating when they played a cheat. The whole thing sounds like an unnecessary programming nightmare considering that most sane persons don't worry too much what their RHP rating is.
introduces ratings inflation, by adding free points to the system. which would affect everyone, as the additional points would be redistributed to your future opponents. the amount of free points could potentially be huge, as cheaters tend to ramp up huge game loads and play them at fast pace without losing much.
Originally posted by greenpawn34now that's just silly, why would otb ratings mean more than any other kind of ratings? either all ratings mean as much, or none of them matter.
Grades should only apply
to OTB games - this is where chess is played. eyeball to eyeball.
That is true measure of your playing strength.
I have absolutely no interest in otb. I've played maybe 10 games, always drunk, and without the slightest effort of playing 'real chess'. while it was okay & fun, I've never played less seriously than those games. I take even online blitz far more seriously, and that's not a lot. here is the most serious chess I play, often putting in hours of work for every move.
Originally posted by wormwoodI assume you are joking; you didn't learn how to play chess on the internet did you?
now that's just silly, why would otb ratings mean more than any other kind of ratings? either all ratings mean as much, or none of them matter.
I have absolutely no interest in otb. I've played maybe 10 games, always drunk, and without the slightest effort of playing 'real chess'. while it was okay & fun, I've never played less seriously than those games ...[text shortened]... lot. here is the most serious chess I play, often putting in hours of work for every move.
Originally posted by no1marauderthat sounds like a straigthforward solution, and would probably work just as well or badly as any elaborate recalculation scheme. and there's a very simplistic sense of fairness to it, unlike the complex recalculation which seems to always be a bit unfair to some players.
Actually the presence of cheats causes ratings to be artificially lower. Probably the optimal solution would be to simply erase ALL changes to anybody's rating when they played a cheat. The whole thing sounds like an unnecessary programming nightmare considering that most sane persons don't worry too much what their RHP rating is.
Originally posted by no1marauderactually I did. saw the kasparov vs deep blue documentary three years ago, started wondering how the pieces actually move, googled up the rules, and got hooked. my first games here are the very first ones I played against a human, ever.
I assume you are joking; you didn't learn how to play chess on the internet did you?
I've got everything documented in my blog:
http://burncastleburn.blogspot.com/2006/02/current-status.html
that's the first blog post 2½ years ago. I had everything since the first few weeks to 6 months in my training diary, but lost it all in a computer crash. then moved the diary online to prevent that from happening again.