Seems to distort the game less than the 3 points for a win, 1 point for a draw that has been used in various events. It awards a wee bit of extra compensation for drawing when the opponent gets the first move. And, it seems as though it would cut down the number of tied scores in an event.
This seems so simple, I wonder why it hasn’t been used. Or, has it?
That’s a little different from 3-1-0 scoring, because it gives an edge to Black.
It would make Black play more conservative and White play more risky. Especially risky for White because Black will try not to give him opportunities. It’s part-way to the Armageddon idea of draw odds, but without the offsetting time advantage.
I don’t think any of this is needed, but if you will pay the GMs to play, you can probably make any rules you want.
So then you would have Black trying even harder for a draw. Stop tinkering and let chess be chess.
My point was that this seemed a less drastic solution than the 3 points for a win, 1 point for a draw scoring system that’s been used in several recent events. You worry about “Black trying harder for a draw” but at least he’d be trying OTB, which seems to me more like “chess being chess” than just signing the score-sheet after a few moves. Is it any less like “chess being chess” than an Armageddon playoff with unequal time and draw odds?
Both you and artichoke seem to overlook that these tweaks to the scoring are intended primarily to inhibit short, prearranged draws, not the longer, more technical draws where one player tries to avoid giving the other opportunities. It seems to me that players of White would be less likely, in general, to agree to a short draw if it gave their opponent the edge, although I’m sure one could concoct scenarios where both players might consider it advantageous. As you both point out, it does have some unintended consequences in terms of Black’s strategy and motivation.
I’m thinking more of the likely effect of the rule, without caring much about the thought behind the idea, the intention of its creator.
The result of the rule would be to make Black prefer draws more, and White less, than what we have currently. It might well make White less willing to agree to a short draw. But even in a long and taxing game, White would still be punished the same for producing a draw.
Is it the players who hate draws? In their games or in someone else’s? If it’s in their games then they have all the power. They can propose no draws and never agree to them. If it’s in the games of others, then again they are spectators!
Why do we care so much what spectators think? I guess professionals should care if it affects their sponsorship and earnings, and indeed people have dreamed up a variety of rules and the GMs turn out to play under all of them, if the prize money is there. But let’s be quite honest: USCF tournament chess for almost everyone is a sport where they pay for the experience of participating and playing tournament chess. The experience should be designed to the preference of the players.
Some games cannot, with correct play, be won by either player. A half-point for each player is a correct result. If anyone asks, you can tell them I said so.
To further illustrate the point:
Let’s make Tic-Tac-Toe more exciting by awarding .4 to “X” for a draw, and .6 to “O”. Will this make it any more likely for X to play risky and to start with say a “side” move, rather than the center or a corner?
Once chess is solved, and humans have evolved to the point where we can remember the solution, your analogy will be spot on.
True. But since some of chess is solved…and since some of it can be remembered by humans…is it appropriate to have this kind of weighted scoring in all situations?
I don’t think it’s appropriate or needed in most situations, but it would lead to a few less place ties in events. I do think it’s a less drastic solution, does a little less violence to the spirit of the game, than the 3 points for a win, 1 point for a draw, 0 for a loss weighting that I’ve seen used a few times. To me, the Armageddon game to break ties, where one player gives time odds and the other gives draw odds, seems a greater distance from “real” chess.
Whether one has White of Black, “x” or “o,” depends on the pairing. For any given probabilistic event, the pairing is a stochastic factor which ought not to affect the result in a standardized tcontest of skill. Perhaps you’ll agree.
I take it you never heard of the Swiss Gambit? One would expect a few less of these if they only gleaned White .4
Similarly with last round draws to divvy up a prize fund and short draws in a round robin to preserve energy.
Are these strategies good or bad? I would say, in themselves, neither, but if one wants fighting chess, for its own sake or to attract sponsors, or to generate public interest, they ought to be discouraged.
You don’t need draws if you are intent on playing a swiss gambit. Just lose a game and you have dropped by two half-points, not just one. A round-robin is where .4/.6 starts to make sense IF you play an even number of rounds. I have never played in a tournament where the public interest might be excited by individual results. And at my age I don’t expect to reach that level of cpmpetition.
And if you want last-round games to be played out to the last Pawn and the playing venue needing be shut down, then .4/.6 is a great way to do it.
It has been done and tested, by Clint Ballard, using his BAP scoring system, in his “Slugfest” tournament, circa 2006, here in Seattle.
BAP (Ballard Anti-draw Point) was more complex than your .6 vs .4, but the principle was the same.
Clint spent his own money to have strong players participate in his serious BAP tournament; include Alexander Shabalov and Gregory Serper.
I spectated some of the Slugfest tournament.
[AA] The players said they played, and considered draw offers, no differently than they always do, even tho the BAP scoring was explained to them.
[BB] I think the winner, Victor Mikhalevski, would have won also under normal scoring.
[CC] The USCF did rate these games normally, despite knowing about the major rule change.
[DD] The draw rate was the same as under normal scoring tournaments.
[EE] No rule was used that set a minimum number of move-pairs to complete before draw offers could be made or accepted.
[FF] The physical venue was Elliott Neff’s Chess4Life center (http://www.chess4life.com/, where the Seattle Sluggers also gather at.
[GG] I remember Gregory Serper being concerned about the pairings for a particular round. During a long discussion, Clint explained how BAP alters the usual pairing algorithm (I half understood at the time, but have since forgotten).
For more info about BAP…
Joel Benjamin, about BAP, 2006/Nov/16, on UsChess.org
Oddly, your initial post does not explicity state whatever specific problem that you are hoping to solve. Exactly what problem are you targeting?
Do you hope to reduce the rate of (aa) tied tournaments, or (bb) drawn games, or (cc) other?
Percentage-wise, chess has more traditionalists then even American Major League Baseball has. Chess traditionalists deny there is any “draw flaw” in modern elite chess.
When I push them to defend their stance, traditionalists says it would be fine if 100% of elite chess games ended as draws; as long as each is hard-fought and tense such that the W/L/D outcome is in doubt until the last few moves.
Of course, a few moments thought shows that the Even-100%-tense-draws-are-OK stance is self-contradictory: when 100% of games are drawn, the outcome can be in doubt only in an irrelevant theoretical sense.
The Anand-Gelfand 2012 match had 83% draw rate; and the Candidates’ 2011 tournament had a 90% draw rate. Those percentages are close enuf to 100% that the effect partly sets in.
Anti-draw Solution: Discard the “Random” from Fischer Random Chess.
These pieces on this board need not be so enormously draw-prone. Instead, blame the current rules of chess, and the people who control the rules.
I played in what I believe was Ballard’s first slugfest tournament, held on Bainbridge Island in December of 2005. It was the first rated tournament I’d played since 1975. I thought at the time that the three point for a win, 1 for a draw penalized the draw more severely than was right. I’ve read of some more recent events using this scoring method, although they don’t seem to have credited Clint for the idea.
I’d consider this change to have minimal impact, actually. The point of my idea (and I’m sure I’m not the first to suggest it or something similar) is to reduce the number of very short, often prearranged, draws. It also would function as a built in tie-breaker for many events. Once a game is in play, the opponent with the disadvantage would fight just as hard to save .4 as to save .5. And the player out to win would usually fight just as hard to gain that extra .4.
A logical way to “split the point” would be to base it upon the rating differences. If the higher player should win 55% of the time, he/she should receive .45 points for a draw. This leads to a draw with a person 400 points lower resulting in a score of 0, but that might be what they deserve. There some players, I am one of them, who have a better career score with Black than White. Should they be “rewarded” for this strange occurance?
Why is that logical? Would you change the scoring of wins also - if your going to do it with draws, why wouldn’t it be logical to do it with wins? So that If GM beats a 500 the GM gets .0000000001 and the 500 gets .9999999999.
I think that would make pairings a nightmare.
There would be as many score groups as there are players, with one player in each group.
Everybody would be paired against somebody in an adjacent group. It would be like 1-vs-2 pairings, throughout the tournament.
Bill Smythe
See Bill, you’ve just hit upon a great way to run a 1v2 ladder tournament!