1) GMs and IMs are studying and analyzing games with the help of engines, too. The style of engines influences the style of the student and may contribute to higher match-up rates (as believed by some GMs).
This just makes it more suspicious if you beat a current GMs matchup rate.
2) She is playing corresponding chess. As Gatecrasher says, some of the best CC players of the past had above average match-up rates.
No, reread it. He states that they had a lower matchup rate than current Super GMs.
3) We do not know the match-up rates of current GMs in CC. It can not be excluded, that the limits for match-up would have to be raised to higher numbers.
With the amount of preparation that a GM puts into his games I doubt that they would average much higher rates than they already do.
4) As you said, every statistic has its outlier, in particular when the starting conditions are violated they are more likely to happen. IF they happen, that does not mean, that they are not possible and have to be excluded because of low chances of appearing.
What conditions have been violated?
Originally posted by tomtom2321) This just makes it more suspicious if you beat a current GMs matchup rate.
[b]1) GMs and IMs are studying and analyzing games with the help of engines, too. The style of engines influences the style of the student and may contribute to higher match-up rates (as believed by some GMs).
This just makes it more suspicious if you beat a current GMs matchup rate.
2) She is playing corresponding chess. As Gatecrasher says, ...[text shortened]... e to be excluded because of low chances of appearing.
What conditions have been violated?[/b]
The limits for the matchup system are based on players from the pre-computer era. Because no computers existed to mess up the numbers. This principle flaw is acknowledged by people like Zygalski.
To speak as if a scientist would have to make the experiment: One compares a limit which has been derived from a different kind of experiment, because the limit from current games can not be dervied, because possible computer-users are messing up the experiment.
This doesnt make the experiment necessarily useless. It just has to be taken into account, that it COULD make it useless.
2) No, reread it. He states that they had a lower matchup rate than current Super GMs.
Gatecrasher said (I am quoting from Zygalskis post):
Even the top echelons of correspondence chess in the pre-computer era had similar match-up rates as today's regular GMs (and less than current super GMs).
and:
While it is true that super-GMs get higher match-ups than regular GMs, who in turn get higher match-ups than IMs, who in turn get higher match-ups than FM's, etc,
So, we all reread it now. This means, some of the best CC players have ABOVE AVERAGE matchup rates. Of course, if we assume, that regular GM's have above average matchup rates...
3) With the amount of preparation that a GM puts into his games I doubt that they would average much higher rates than they already do.
You can doubt as much as you like. You can also hope, wish and have oppinions. I am simply not sure, what kind of matchup rates current GMs like Carlson and Kasparov would have in CC. But I am very interested. Because it makes a HUGE difference playing OTB or CC - or are we now going to start discussing again, how equal both things are? They are two different worlds - so who can know matchup rates beforehand? It would be at least unscientific (again, you are invited to make an educated guess).
4) What conditions have been violated?
I am not sure, maybe none. But the possibility alone, that there could be a player who only ever studied from engines - he would most likely be an outlier, because he violated the basic assumptions of the matchup system - we all learned to play as the GMs who are responsible for the current limits...
Game 7709078
According to Rybka, Skeets is down 3.01 pawns and if Kaoslos finds 35 ...... Qxe2 the game is resignable as White either loses a piece or gets mated.
Game 7709081
Kaoslos has a rather clear advantage here, but blunders with 48 d6? allowing a simple series of moves which liquidate his strong passed pawn and give Skeeter an easy endgame win.
Those were the first two on SB's list and my analysis does show high matchup rates. But they don't look like engine games; perhaps there were a few exceptions but it never seemed like IM or Weyerstrass were anywhere near losing. I'll work on the rest though.
- Joined
- 08 Apr 09
- Moves
- 20184
Hey Zygalski,
I have got a few questions about the method:
- out of curiosity and because many posters seem afraid of drawing conclusions from statistical evidence: what is the chance of false positives, i.e. that the matchup rates indicate a cheater while the player is honest? It's probably very hard to put a number on that, but maybe it's possible to give some (calculated) estimate.
- i have little/no experience with using engines for analysis. So I'm wondering how engines matchup with each other. E.g. how would the results for a rybka vs. rybka game look like using houdine to calculate matchups? How "alike" are engines? --> This seems important to me, because you can't know which engine someone may be using. On the other hand, it would mainly produce false negatives, rather than false positives.
GP,
It doesn't prove much trying to figure out whether a move is inspired by either tactical ideas or just computer evaluation. An initial mistake (which is typical human) may as well provoke a large sequence of logic moves and then turn into an advantage only after 10 moves. That's called luck. Nothing can be proved based on 1 move or even 1 game. I believe in numbers: the larger the sample, the better. But feel free to give your analysis on the moves you asked for. I hope we can learn from it, and then I mean about chess.
Originally posted by no1marauderExcuse me for interfering in my: "the kids are asleep Saturday red wine drunken state".
Game 7709078
According to Rybka, Skeets is down 3.01 pawns and if Kaoslos finds 35 ...... Qxe2 the game is resignable as White either loses a piece or gets mated.
Game 7709081
Kaoslos has a rather clear advantage here, but blunders with 48 d6? allowing a simple series of moves which liquidate his strong passed p ...[text shortened]... never seemed like IM or Weyerstrass were anywhere near losing. I'll work on the rest though.
Yes my boss has costs me many lost points when he enters and I just move! - instead of doing the wise and just close the screen...
The "mach-up" will never be able to catch cheaters!
The first point is that you will have to manually go through all games in your analyses to catch a cheat, which will take way to much manpower.
Why?
1) anytime multiple moves leads to a win you have to disregard the move played if it’s one of those (a prior post state about the same).
2) If an engine user plays a non engine user - the engine user will expand his/hers advantage the more complicated and/or open the position becomes - hence the engine user will (if smart) choose complicated or open" over "0.10".
A (bad player and) engine user will often enter a clearly draw endgame one or two pawns up - where a decent human player would keep the game alive - settling for a smaller advantage.
point 2: When I prepare for OTB games - of-cause I switch on the engine when entering the books suggested variation to my database - but I never chose an engine move over a book move without very-very deep preparation - the engine simply fails to "see" deeply enough in the position!
Point 3: The official base may stop and you have no more played or analyzed moves in a particular line, so you start the analyze here... but who says the players haven’t played the line before somewhere and/or analyzed it with computer - hence finding the "best" moves?. A good player who sticks to his repertoire these days has many very deeply analyzed lines in his base...
Finally and most important: As Kramnik said after "toilet-gate", you can't determine if it’s a computers moves, because the "best" moves changes the longer you let the computer analyze...
Originally posted by KaoslosYou've been playing chess since you are seventeen, with a wife and 3 kids?
Excuse me for interfering in my: "the kids are asleep Saturday red wine drunken state".
Yes my boss has costs me many lost points when he enters and I just move! - instead of doing the wise and just close the screen...
The "mach-up" will never be able to catch cheaters!
The first point is that you will have to manually go through all games in your ves, because the "best" moves changes the longer you let the computer analyze...
Cheat!
-m.
Edit: I would say a lot more, but you just aren't woth the time nor the effort.
Enjoy your career as a cheat!
Originally posted by mikelomno -I just been playing with other chessplayers ;-) - yes my profile is true! If you put enough at stake I'll show you...
You've been playing chess since you are seventeen, with a wife and 3 kids?
Cheat!
-m.
Edit: I would say a lot more, but you just aren't woth the time nor the effort.
Enjoy your career as a cheat!
Edit: and I'm probaly one of the highest if not the highest rated otb players on this side - I'll show you that too if you put enough at stake. So the question is - will you put your money where your accusations are? :-)
Excuse me for interfering in my: "the kids are asleep Saturday red wine drunken state".
Getting your kids off to sleep by giving them red wine is not responsible parenting no matter what day it is.
Yes my boss has costs me many lost points when he enters and I just move! - instead of doing the wise and just close the screen...
Good god man. How many points would you have if you were unemployed ?
The "mach-up" will never be able to catch cheaters!
It's just part of the evidence.
Finally and most important: As Kramnik said after "toilet-gate", you can't determine if it’s a computers moves, because the "best" moves changes the longer you let the computer analyze...
I spend a lot of time on the toilet but it hasn't improved my rating.
Originally posted by KaoslosWell with your rating and climb to fame are you asking me to be your sponsor now?
no -I just been playing with other chessplayers ;-) - yes my profile is true! If you put enougt at stake I'll show you...
You already said you have a boss/manager.... wot's up - does he cream your income from OTB??....
Can you give us an example of one of your OTB games, venue/opposition etc plus date, and some kind of annotation of your game?
Even a game with your anno would suffice.... prepared to do that to qualify your position here?
-m. 😏