The "BAP" System and the GM Slugfest Tournament
One Clint Ballard has had enough of draws, and is putting his money where his mouth is. He begins an open letter to GMs and IMs like this:
After asking these rhetorical questions, Ballard offers a brief historical excursus recalling the days when draws were automatically replayed, notes that there are few draws in computer chess (indicating that the draw "problem" is not caused by GMs' nearly perfect understanding of the game), points out that even the NHL has recently eliminated ties to make their fans happy, and then goes on to propose the BAP system and plug his tournament. I'll say a little about his system and tournament, below, but first I'd like to address his questions.
In one of my first classes in graduate school, I was told to avoid rhetorical questions. (Who could have anything against rhetorical questions?) The reason is that someone might actually answer them, and not in the way I had intended! So it is here.
To the first question, how many of my 10 most brilliant games were draws, the answer is at least two. (Yes, I know I'm not a GM or an IM, but (a) the quick-draw problem doesn't just belong to those guys, and (b) I've played quick draws (and real draws, for that matter) with GMs and IMs.) One of those games is in my top two, the ridiculous game with Barcarola from the 1999 US Amateur Team tournament where I sacrificed two queens, and practically everything else besides. The second was a draw in the late 1980s with Larry Christiansen where we took turns sacrificing material for the attack, only to wind up in a drawn rook ending with a symmetrical pawn structure.
The second question is silly, of course, in the context of his overall proposal, because it doesn't show that anything is wrong with draws, but only - at worst - that something is wrong with (very) short draws. Thus the Sofia/MTel rules (no draw offers; draws only allowed when it's a forced repetition, material is insufficient or the arbiter permits it) are sufficient here. Further, there have been brilliant games meeting the question's condition: two that come to mind are Fischer-Tal from the 1960 Olympiad and Sax-Seirawan (a draw involving a queen sac that completely overturned the theory of that variation of the Pirc).
The third question fares no better, for while it's difficult to defeat someone using the White pieces as a fire extinguisher, it's hardly impossible. (Take a look at all the 0-1s in the history of the Exchange French, for starters.) Further, there's nothing preventing technique-and-safety lovers from continuing to play that way, waiting for the opponent to go astray. Players like Ulf Andersson, Tigran Petrosian and Anatoly Karpov became legends in that way, and drawing against them with Black wasn't very easy at all.
Likewise, his fourth question doesn't really address the draw problem, but one's tolerance for risk. Unless the penalty for a draw is Draconian, one isn't going to turn a Tigran Petrosian (the late ex-world champion, not the young GM) into a gambiteer.
So let's have a look at his proposed "BAP" system:
Black wins: 3 points
White wins: 2 points
Black draws: 1 point
White draws: 0 points
Any losses: 0 points
Two things are clear about this system. First, having the Black pieces is now a colossal advantage, far bigger than White's 55% expected edge in the usual case. And second, those with a more volatile style have an advantage against those with a more "correct" approach, even if by almost any current objective standard the former are clearly weaker.
So why do this? Ballard probably thinks the BAP system leaves chess as is, only serving to improve it for the fans, who crave decisive results. That's where the parallel to the NHL comes in, I suppose - the game will be played the same way, but the need for a decisive finish adds to the drama and makes the fans happy. But the sports parallel is inapt for an important reason. Overtime periods, extra innings, shootouts and so on do little to change how the game is played, except perhaps slightly at the very end of regulation/in the late innings. The BAP system, however, fundamentally changes the way the game is to be played, penalizing not just the wimpy and the lazy (or rather, the wimpy and the lazy when they have White), but anyone playing an excellent game of chess that isn't decisive.
Another outcome is that players with less imbalanced styles are punished; worse, they are almost prevented from displaying their art. There's nothing wrong with playing like a Tal, a Shirov, a Korchnoi or a Morozevich. I enjoy their chess at least as much as the next fan. But imagine what would happen if we changed the rules and created what we might call the Cunningly Reversed AP system:
White wins: three points
Black wins: three points
White draws: two points
Black draws: two points
Any losses: no points
In this, the Cunningly Reversed AP system, motivation to take risks for a win is seriously reduced, and it's the safety-first, technique-lovers who are disproportionately rewarded. Worse still, if the Tals of the game value results, they will have to change their styles and give up their art (or at least a significant and very special part of it) to succeed in this Not-So-Brave New World.
But the same goes for BAP: there's nothing wrong with playing like Karpov or Petrosian, if one is playing full games. Their games are masterpieces of the highest caliber, and it's terrible to force them to choose between relative competitive failure and playing the beautiful chess that's the finished, mature product of their hard work, mental gifts and aesthetic sensibility. If the choice is between losing fans who are not just averse to wimpy/lazy draws, but draws per se, and who wish to destroy a large part of the game as we know it, I say good riddance to the so-called fans. (To be honest, I think they are pretty rare; most players, even if they aren't moved by an Andersson, can be educated to appreciate his play even if they'd rather see a rip-roarin' sacrificial melee.)
A last critique, a sort of added "bonus" of the BAP system: the prospects for cheating are greatly increased. Suppose the World Open adopts this system, and going into the last round we have two players tied for first with, say, 18 points. These two have played each other, so they're paired down into the 17-point group, one with White, the other with Black. The 18-pointer with White cannot even assure himself of a tie for first, nor can anyone else but the Black 18-pointer. Guess which board is most likely to feature a bribe, and multiply it by all the score groups in which the difference between a White win and a Black win has a substantial difference in the prize fund.
All that said, Ballard is to be praised for supporting his ideas with action, personally funding a GM/IM event in Bellevue, WA from October 13-15 of this year. You can read all about the tournament, and his complete open letter with his entire rationale for the BAP system here.
How many of your 10 most brilliant games are draws?
How many of your 10 most brilliant games are draws by agreement less than 10 moves out of known theory?
How many times have you won against a strong player who was playing safe as white and not taking any chances?
How many times have you avoided playing the most exciting and daring lines because doing so did not make sense due to external factors, eg. prize money, rating points, tournament standings, etc.?
After asking these rhetorical questions, Ballard offers a brief historical excursus recalling the days when draws were automatically replayed, notes that there are few draws in computer chess (indicating that the draw "problem" is not caused by GMs' nearly perfect understanding of the game), points out that even the NHL has recently eliminated ties to make their fans happy, and then goes on to propose the BAP system and plug his tournament. I'll say a little about his system and tournament, below, but first I'd like to address his questions.
In one of my first classes in graduate school, I was told to avoid rhetorical questions. (Who could have anything against rhetorical questions?) The reason is that someone might actually answer them, and not in the way I had intended! So it is here.
To the first question, how many of my 10 most brilliant games were draws, the answer is at least two. (Yes, I know I'm not a GM or an IM, but (a) the quick-draw problem doesn't just belong to those guys, and (b) I've played quick draws (and real draws, for that matter) with GMs and IMs.) One of those games is in my top two, the ridiculous game with Barcarola from the 1999 US Amateur Team tournament where I sacrificed two queens, and practically everything else besides. The second was a draw in the late 1980s with Larry Christiansen where we took turns sacrificing material for the attack, only to wind up in a drawn rook ending with a symmetrical pawn structure.
The second question is silly, of course, in the context of his overall proposal, because it doesn't show that anything is wrong with draws, but only - at worst - that something is wrong with (very) short draws. Thus the Sofia/MTel rules (no draw offers; draws only allowed when it's a forced repetition, material is insufficient or the arbiter permits it) are sufficient here. Further, there have been brilliant games meeting the question's condition: two that come to mind are Fischer-Tal from the 1960 Olympiad and Sax-Seirawan (a draw involving a queen sac that completely overturned the theory of that variation of the Pirc).
The third question fares no better, for while it's difficult to defeat someone using the White pieces as a fire extinguisher, it's hardly impossible. (Take a look at all the 0-1s in the history of the Exchange French, for starters.) Further, there's nothing preventing technique-and-safety lovers from continuing to play that way, waiting for the opponent to go astray. Players like Ulf Andersson, Tigran Petrosian and Anatoly Karpov became legends in that way, and drawing against them with Black wasn't very easy at all.
Likewise, his fourth question doesn't really address the draw problem, but one's tolerance for risk. Unless the penalty for a draw is Draconian, one isn't going to turn a Tigran Petrosian (the late ex-world champion, not the young GM) into a gambiteer.
So let's have a look at his proposed "BAP" system:
Black wins: 3 points
White wins: 2 points
Black draws: 1 point
White draws: 0 points
Any losses: 0 points
Two things are clear about this system. First, having the Black pieces is now a colossal advantage, far bigger than White's 55% expected edge in the usual case. And second, those with a more volatile style have an advantage against those with a more "correct" approach, even if by almost any current objective standard the former are clearly weaker.
So why do this? Ballard probably thinks the BAP system leaves chess as is, only serving to improve it for the fans, who crave decisive results. That's where the parallel to the NHL comes in, I suppose - the game will be played the same way, but the need for a decisive finish adds to the drama and makes the fans happy. But the sports parallel is inapt for an important reason. Overtime periods, extra innings, shootouts and so on do little to change how the game is played, except perhaps slightly at the very end of regulation/in the late innings. The BAP system, however, fundamentally changes the way the game is to be played, penalizing not just the wimpy and the lazy (or rather, the wimpy and the lazy when they have White), but anyone playing an excellent game of chess that isn't decisive.
Another outcome is that players with less imbalanced styles are punished; worse, they are almost prevented from displaying their art. There's nothing wrong with playing like a Tal, a Shirov, a Korchnoi or a Morozevich. I enjoy their chess at least as much as the next fan. But imagine what would happen if we changed the rules and created what we might call the Cunningly Reversed AP system:
White wins: three points
Black wins: three points
White draws: two points
Black draws: two points
Any losses: no points
In this, the Cunningly Reversed AP system, motivation to take risks for a win is seriously reduced, and it's the safety-first, technique-lovers who are disproportionately rewarded. Worse still, if the Tals of the game value results, they will have to change their styles and give up their art (or at least a significant and very special part of it) to succeed in this Not-So-Brave New World.
But the same goes for BAP: there's nothing wrong with playing like Karpov or Petrosian, if one is playing full games. Their games are masterpieces of the highest caliber, and it's terrible to force them to choose between relative competitive failure and playing the beautiful chess that's the finished, mature product of their hard work, mental gifts and aesthetic sensibility. If the choice is between losing fans who are not just averse to wimpy/lazy draws, but draws per se, and who wish to destroy a large part of the game as we know it, I say good riddance to the so-called fans. (To be honest, I think they are pretty rare; most players, even if they aren't moved by an Andersson, can be educated to appreciate his play even if they'd rather see a rip-roarin' sacrificial melee.)
A last critique, a sort of added "bonus" of the BAP system: the prospects for cheating are greatly increased. Suppose the World Open adopts this system, and going into the last round we have two players tied for first with, say, 18 points. These two have played each other, so they're paired down into the 17-point group, one with White, the other with Black. The 18-pointer with White cannot even assure himself of a tie for first, nor can anyone else but the Black 18-pointer. Guess which board is most likely to feature a bribe, and multiply it by all the score groups in which the difference between a White win and a Black win has a substantial difference in the prize fund.
All that said, Ballard is to be praised for supporting his ideas with action, personally funding a GM/IM event in Bellevue, WA from October 13-15 of this year. You can read all about the tournament, and his complete open letter with his entire rationale for the BAP system here.
Posted by Dennis Monokroussos on
Tuesday August 22, 2006 at 1:45am
I feel like this system is going to make people overpress, and while there will be more decisive games, they will be far uglier.
It's hard to vigorously criticize someone who's put his own money and time into creating events that bring to reality his sincere vision of how to make chess more popular.
But I think tournaments played under BAP are at best a diversion from real chess. Therefore they should not be supported; and furthermore, since BAP is materially different from real chess (as you demonstrated in your post, although you don't say so explicitly), SUCH TOURNAMENTS TECHNICALLY ARE NOT ELIGIBLE TO BE USCF-RATED, following the letter of the law.
I have contacted the USCF Ratings Committee asking them to reconsider the ratability of BAP tournaments.
My biggest problem with Clint Ballard and his system is that he is, shall we say, "disingenuous" about describing the distinction between his system and chess as we know it. He regularly states that BAP is not a "rule change," but merely a different "pairing method."
This is clearly not true. As I wrote when debating Clint on another blog, if changing the OBJECTIVE of a game doesn't count as a "rule change", then what does? No matter how you slice it, counting a draw as a loss for White is a material difference in objective, between BAP and real chess. Yet Clint systematically denies this very basic and self-evident fact.
Now Clint is a gentleman and I wouldn't want to accuse him of dishonesty. But I'm sure he knows that the USCF allows TD's all kinds of leeway in pairing methods for rated events; but very little leeway about rules of play itself.
Time controls, penalties, rules of personal conduct during the game, scoresheets, and other ancillary matters can vary. But if an event is to be rated, the rules of how the pieces move are sacrosanct -- otherwise, USCF would be rating bughouse, odds-games, theme tournaments (all games played from a pre-set opening position), and all manner of other chess variants.
In fact I believe BAP, since some of its most important rules (defining the various outcomes of a game) obviously and clearly deviate from those of chess, is actually a chess variant, and therefore should not be allowed to be USCF (or FIDE) rated, any more than bughouse should be rated.
As far as the underlying problem that BAP is meant to solve, I agree with Dennis that proponents of BAP and most other anti-draw systems are in effect using a nuclear bomb to swat a fly. Banning draws by agreement, either before a certain number of moves or at any time in a game (as done at Sofia M-Tel), solves the problem without transforming chess into a different game. I also share Dennis's view that draws aren't even a problem to begin with. Like Dennis, I count a few draws among my own most brilliant and entertaining games (two of which are slated for publication in a future Chess Life issue).
Clint has told me he's certain that eliminating draws is the only way chess can have any chance of succeeding as a televised sport. Even if he is right, I would rather see real chess survive and prosper as an un-televised art if that's the way it must be, than see a bastardized impostor variant of the succeed and perhaps further confuse and dilute the public's appreciation for our wonderful, thousand-year-old game.
People who want to destroy chess in order to save it from draws remind me of people who said they had to destroy the village in order to save it from Communism. Even more apt is an 18th-century Edmund Burke quote about misguided children of their country who "hack that aged parent in pieces," throw it into a cauldron and pronounce "wild incantations," all in hopes of "magically restoring it to health."
And, what's this American obsession with sports having a decisive result? Football (soccer) matches may be drawn. And I don't like the 3:1 rule, not even in football.
I prefer league tournaments (rather than k.o.), like Linares, Wijk aan Zee, Dormund or the World Championship of San Luis, where the player who really wants to take risks in order win can make a difference -remember Tal, Kasparov or Topalov. At the same time, as Dennis said, it would be unfair to punish the more "academic" yet excellent players with the BAP system or the 3:1 rule. There's no need for a change, in my opinion.
Btw I don't know about Andersson, but both Petrosjan and Karpov were very capable of beating strong opponents with Black in attractive style:
Tal-Petrosjan, Moscow 1973
Larsen-Petrosjan, Milano 1975
Sax-Petrosjan, Rio de Janeiro 1979
Timman-Karpov, London 1984
Kasparov-Karpov, Sevilla 1987 (2)
No doubt there must be more.
Imo worth studying for the attackers amongst us.
Finally I think chess becomes less interesting, if everybody plays like Tal or Sjirov.
Of course, my point wasn't that these greats could never win with the Black pieces, but that their styles tended towards a relatively high percentage of draws compared to someone like Morozevich or, in an earlier generation, Larsen. And this penalizes them under the BAP system despite their superior playing strength.
Although Tal is probably my all-time favorite player, I would agree that chess loses something if everyone plays like him. Fortunately, not everyone does - and for that matter, very few really can!
I was not contradiction you. I just wanted to point out, that the safety-first reputation of these three is not entirely justified.
Since this thread has material inaccuracies, thinly veiled personal attacks and even an example by the moderator with the intentionally derogative acronym of "CRAP", it is clear that there is a lot of animosity and downright hatred toward my BAP system. I doubt that a rational discussion is possible and I am used to that, but I will present FACTS for the unbiased reader to consider and hope it won't get censored. Of course, all the critics will accuse me of being irrational because I won't simply agree with 100% of what they say, even if they are accusing me of being disingenuous, that I am conducting tournaments that should be banned, etc.
BAP doesn't solve all problems, so the difference in value of a win against a strong opponent vs. weak one, is something BAP doesn't fix. Then again it is the same with existing tournaments, so not sure how this is a negative reflection on BAP. In a GM tournament, this is less of a factor anyway, but of course all GM's are not the same. The traditional strategy of drawing against the strong opponents and beating up on the weakies won't be optimal with BAP. I claim that is a good thing. BAP doesn't fix the cheating that already exists, either. I am sure that BAP doesn't fix a lot of other things too, but the resistance it is getting by addressing just two issues indicates that having it fix more things will increase (if that is possible) the resistance to BAP.
The doomsayers claim that BAP will totally change chess as we know it based on zero experience with it. However, that is not what the games that have been played have shown. A player's best chance to win is to play in a way that is consistent with their style! If anything, it magnifies the differences in style, as a strong positional player will probably play even more positionally while a strong tactical player will play more tactically. What is it called when definite statements are made about a topic without any data or actual experience with something, all the while ignoring the actual games that have been played with BAP?
The GM Slugfest (gmslugfest.com) is an experimental test that will put to rest once and for all if GM BAP games are chess or not-chess. The games will speak for themselves and I am confident there will be a lot very interesting games, positional, tactical and hybrid. I predict around 25% draws between the titled players. While there will be blunders, I doubt there will be any more than normal. I would hope that there will be objective analysis of the games from the GM Slugfest. It is this hope of mine that has compelled me to personally fund the quite expensive event and brave the ridicule for actually DOING something about the draw problem. The success BAP has had at the lower levels was dismissed as just games played by weak masters, so it didn't count. I am sure that critics have already concluded that the GM Slugfest games are not chess and I guess they can already write their summary of the quality of games, etc. as the facts don't matter when you have your mind made up. For those who are open minded and willing to wait for the results before declaring them invalid, why not make specific predictions that can be verified. Something along the lines of the scientific method. Based on what I know about chess, incentive systems, math, human nature, past BAP tournaments, etc. I am predicting around a 25% draw rate and the vast majority of draws will be of the interesting type that everybody seems to be rabidly trying to use as a criticism of BAP. If these exciting draws are what is good about chess and BAP increases them, it seems backwards that they are used to criticize BAP.
For those who claim there is no draw problem and that chess should not be changed in anyway and go so far as to try to ban BAP rated tournaments, well, first let me assure everyone that the USCF has said that using BAP for pairings and prize money is allowed. I brought this up at the US Open member meeting to make sure. For those who feel compelled to try to ban BAP rated tournaments, my question is why? With 99.99999% of chess games not using BAP, why the intolerance to BAP? Is a G/31 game the perfect pristine chess that BAP supposedly sullies? A G/31 game counts for rating points as much as 40/2 + G/60. A kind reader of Chess Today found an 1867 German reference to a tournament that is the earliest known (to me) use of the half point for a draw. Almost 140 years. So long that it seems like it is part of the rules of chess, but it isn't.
Also, if there is no problem with chess, why do we lose 99% of elementary tournament players by the time they are 25? Would the chess world be better off if we only lost 98% of these players? If BAP could have a beneficial effect to player retention, then why attack it? Nobody has made a credible claim that BAP would decrease the defection rate and if BAP was used in 0.01% of the chess games played how could it be a negative?
For those who admit that there is a problem and like the alternative solutions better, why not give BAP a chance? Clearly, a tournament with Sophia rules is more fun to watch than one without. What if BAP is even more fun to watch than a Sophia rules tournament?
As far as the ability of white to draw if white wanted to, if Kasparov (at his peak or even now) was playing to not lose, who other than Hydra would have a chance to beat him? In fact, isn't that what he ended up doing against Karpov with a zillion draws in a row? The draw margin in chess is big enough that Kasparov was able to draw with either color, at will. THAT is the problem with the 140 year old point system! Is that really what we want? Clearly, it seems to be what everybody is working so hard to defend. When 2 GM's agree to a draw because the position is drawn, but mere mortals scratch our heads wondering why they agreed to it, would it not be helpful if they actually played out the instructive endgame? If it is so easily drawn, they could just blitz it out and it won't take much effort on their part. The chess world will be able to see how to properly play these positions. Is this a bad thing?
BAP decreases the percentage of draws, but of the games that are drawn, it INCREASES the interesting draws. It eliminates the dull draws, at least among non-titled masters.
I will respond to some of the specific things in this thread that are inaccurate, misleading or downright wrong.
"indicating that the draw "problem" is not caused by GMs' nearly perfect understanding of the game"
This statement makes the implication that GM's have solved chess! Gee, I must have missed that announcement. Was it my imagination that Hydra DEMOLISHED Adams? One of the top 10 players in the world at the time, not prone to losing, got killed. The only draw was by agreement of the operators of Hydra, even though Hydra itself thought it was winning. Since Hydra is better than the GM's, shouldn't we have seen if it could find the win? So, with an actual result of 5.5/6 vs. Adams and maybe it should have been 6 out of 6, it boggles the mind that claims are being made seriously that human GM's have a nearly perfect understanding of the game.
"Two things are clear about this system. First, having the Black pieces is now a colossal advantage, far bigger than White's 55% expected edge in the usual case"
This so called 55% expected edge for white is not indicative of white's true advantage and is only due to the broken point system that such claims are being made. At the high levels of chess (let's not get into debating what is and isn't high level as I am just illustrating a point), the approx percentages are 30% white win, 15% black win, 55% draw. Based on the 1867 point system, this gives white 57.5% of the points and black 42.5%. Close enough to the cited 55%. This gives a bias of about 25% to white, so the current point system is biased, nobody is arguing that it isn't. The fact that a 2:1 advantage is changed to a 25% advantage is an indication that something is wrong, but yet, that is how the math works out. With BAP, the 15% black win, 30% white win, 55% draw will give black 100 BAP in 100 games and white would get 60 BAP. Aha! I am sure the critics will try to quote this out of context to "prove" that black gets an unfair advantage with BAP. This totally ignores the effect of the scoring system on the distribution of draws. IF, BAP does reduce draws to my expected level of 25% and white maintains the 2:1 advantage in winning, the distribution becomes 50% white wins, 25% black wins, 25% draws. Now, we get the result that there is no bias either way. Coincidence? Actually, no, I designed BAP to do this. Now, if I am wrong about the 25% draws and the 2:1 white ratio, then BAP will have a bias. However, the fair test of BAP is the magnitude of the bias. It would seem reasonable that if BAP biasing is less than or equal to the 1867 point system's biasing that it is no worse and any reduction in biasing would be an improvement in this area. Therefore, claims that BAP is unfairly biased toward black are totally unfounded. I admit that BAP might be biased toward black to a certain degree, but I would be surprised if it was more than the 1867 system. If the GM Slugfest proves me wrong, I will admit it and be open to tweaks to BAP. Until something has actual evidence that shows that it is broken, it doesn't make sense to fix it. There is no evidence that BAP is broken.
"There is a draw disease that strikes from time to time, but the outbreaks tend to arise in positions that aren't particularly interesting and are, I suspect, often foreordained"
I claim that these positions that aren't particularly interesting are by design. To get such a position, at least one, if not both of the players didn't want to get into anything too interesting. From personal experience that you are free to dismiss, since I am only an A-player over the board, almost all my games are interesting. I had a 2500+ performance rating during the first 6 rounds at the recent US Open. 4 wins against masters, all of them with material sacrifices that were sound enough to beat the masters. Considering that I was over 300 pts underdog I made many inaccuracies, yet I had as many wins as any GM at that time and I had the toughest pairings as of the first 6 rounds. How often do you outperform your rating by 600 points over 6 rounds? Could it be that I might actually know a little something about this game called chess? Oh, I forgot, rhetorical questions are frowned upon, sorry. For anybody curious about how an A player played like a GM for 3 days and 2 hours, you can read about it on slugfest.org.
"how many of my 10 most brilliant games were draws, the answer is at least two."
2 is 20% of 10, not 50%+. Also, this question was addressed to GM's, as the draw percentage drops significantly below 2400. My point with this is that the majority of almost all player's most brilliant/memorable/favorite/best games are probably decisive games. Since the majority of GM vs. GM games are drawn, it is backwards and doesn't match the distribution of the best games. Statistical indication that the ancient 1867 point system is not maximizing the number of brilliant and hence, interesting games.
"last round we have two players tied for first with, say, 18 points"
The assumption that both players are tied with 18 points, but have different colors is incorrect. BAP tournaments are color balanced, so the one due black in the last round got the 18 points out of one less possible points than the player due white in the last round. On a percentage of points possible basis, these two players are most definitely not in a tie for first place. It is perfectly normal and expected that a player that is ahead of another player going into the last round has an advantage over the other player. BAP has this characteristic and the example used to present BAP as flawed is itself flawed. Also, saying that a white draw is the same as a white loss is also incorrect, though understandable error. BAP is not a zero-sum point system. The effect of white drawing has a 2 BAP change to the overall point pool relative to a white loss. White winning has a 3 BAP difference relative to a white draw. Black winning has a 2 BAP difference relative to Black drawing. Black drawing has a 3 BAP difference relative to black losing. If you are going to comment on BAP math, please use the correct numbers.
"It's called active or rapid chess: 30 min pgpp"
As less and less time is available for a game, the error rate goes up. I want to minimize the effect of blunders on the result. Ideally, blunderfree chess is what I want and using rapids or speed chess to decide serious games/matches/world championship?? is better than just calling it a tie, but in my opinion only marginally so.
"Clint has told me he's certain that eliminating draws is the only way chess can have any chance of succeeding as a televised sport"
I do not want to eliminate draws. I admit that originally I was thinking about that, but experience with BAP has showed me that many draws are in fact very exciting and with BAP draws are decisive anyway. No need to eliminate them. What I said was that every game has to have a winner and a loser for chess to be a sport that has a chance of success as a TV sport. BAP makes draws decisive as there is a player that got 0 points. No need to eliminate draws, in fact there is significant evidence that shows that a properly played game of chess should end in a draw, so it would be silly to eliminate them. I am many things, but I try not to be silly.
"Another outcome is that players with less imbalanced styles are punished; worse, they are almost prevented from displaying their art"
This is purely conjecture of the worst type. This contradicts all the actual BAP games so far and it is simply stated as an "outcome" as if it were a proven fact. With all respect to the author of this statement, I would hope that rationality and logic will overcome the fear of the unknown. The effect of BAP on GM's will be known soon enough, no need to declare the incumbent 1867 system the winner beforehand. I believe the winning percentage of Karpov was one of the highest during his peak years. Karpov is known as Karpov because he won the games that he did, not because of the games he didn't win. How this is twisted into BAP preventing future Karpov's from displaying their positional masterpiece boggles the mind. Now, if the so called positional chess is inferior to the so called "imbalanced" chess and cannot win as much, well, then maybe positional chess should be discouraged. However, it is my opinion that positional chess and dynamic chess are both equally valid approaches to winning and thus both will be equally rewarded under BAP.
I hope 1% of the readers of this will be open minded enough to wait to see the games from the GM Slugfest before they declare to everyone that they know that BAP doesn't work, or that it ruins chess, etc. I hope that a rational debate will happen, but I predict that the usual BAP-bashers will use their non-experience with BAP tournaments to make unproven assertions that "prove" how bad BAP is. If they realize that doing so might make them look to be unreasonably biased, they could always call me names, call me a liar, etc. It is a free country, with free speech, so go for it. I have been equated to Hitler because of this, it probably can't get much worse than that. Spending close to $100,000 of my money to try to help fix the draw problem in chess that some say doesn't even exist, is what I am doing for chess. I have no illusions about ever recouping the money spent, nor even a fraction of it, but I hope that I have at least earned enough respect to have my ideas truly given a fair trial.
Clint Ballard
manager of uschessleague.com Seattle Sluggers
In general, if you can improve your position significantly without risking anything, it makes sense to try it. So, if BAP makes chess more exciting and we could end up with chess on TV to any extent, what is the harm in trying out a new system?
I guess for the chess purists who cannot stand any games being played anywhere that doesn't reward white draws equally to black draws, then the risk is intolerable.
Not trying new things is called stagnation, isn't it? When the environment changes but no changes are made, usually you end up in a suboptimal place. Since 1867 a lot has changed in the chess world. I think the most significant is that the ability to defend has gotten a lot better, not to mention tremendous increase in opening and endgame knowledge. Armed with this knowledge, a top GM ** could ** play to not lose and be quite effective at not losing.
Let's do a thought experiment. What if we had a GM tournament where a draw was worth the same as a win for white, but a draw is the same as a loss for black. Clearly unfair, but the key question is how many games will black be able to win assuming all players are 2700+
With draw percentages in the 55% to 70% range as it is, it would seem likely that we would see a draw percentage of around 80% or even 90% under those rules. I would also imagine most of the wins will be white wins as black will have to press for a win whilte white can just play solid and wait for an unsound move.
Now, let's flip the colors around using the same point system. Black drawing is the same as black winning, white must win as draws or losses are the same. I would guess that the draw percentage could actually come down a bit, but since black has strong incentive to draw, there are counterbalancing forces. Some crazy person would have to fund a GM tournament to find out the answer to this one. It could be anywhere from 40% to 80% draws, but the distribution of win, loss, draw will be significantly different than in the first case.
Now for the zinger. Let's think through what would happen if we scored the draw the same as a win for both sides, but we have the Sophia rules so no draw offers allowed. Would you be surprised if the draw rate under such conditions would be close to 100%? Everybody can get a full point every round and everyone can split the prize money equally and everybody would be happy.
I hope everybody agrees with my assessment of the 100% draw scenario even using Sophia rules. This is the fundamental flaw with the Sophia rules. A draw can be implicitly agreed to without a single word being spoken by either player. Maybe there will be some crazy players who play for a win anyway because they got bored out of their mind trading off pieces as quickly as possible and getting to a totally drawn endgame.
Let's reduce the value of a draw a little bit, say down to 9/10'th of a point. We probably now see some decisive games, but not too many as winning is a lot more than 11% harder than drawing. What about at 3/4 of a point? I would predict still over 90% draw rate as it is more than 33% harder to win than to draw.
See, that's the key, the difficulty of winning versus drawing is the breakeven value of a draw. If it is twice has hard to win as to draw, then the draw being worth half as much is the breakeven point where it doesn't matter whether you take the draw or go for a win.
Now, let's see how the math works using the 30% white win, 15% black win, 55% draw percentages. With these percentages, a white draw is a little bit less that twice as easy as a white win. White would be better off going for the win with the draw being worth half a point, but it is a close call. For black, it is entirely different. It is almost 4 times easier to draw as black than to win. A draw gets you statistically almost twice as many points for black as winning would. Hence the well known strategy of draw as black, win as white, unless you are playing against a stronger player, then might as well draw and go for wins against the bottom of the ladder.
For those who are still with me on the math, it should be clear why we have so many draws at the GM level. Black gets almost double the expected value from a draw, so it clearly makes sense to be satisfied with a draw. White is ambivalent toward draws. While a draw is less than the expected value, it is not by much, so other factors could tip the balance. For instance, if white is playing against a stronger opponent. Then, the expected value would actually be lower, as higher rated opponents win more often than the lower rated ones. At this critical threshold value, it becomes logical for white to prefer a draw as it gives white more expected value than trying to win. So, let's say there is a 100 rating point difference, with black being higher rated. Black is supposed to win significantly more than white, but even with a 2:1 advantage, that still makes a draw very close to the expected value!
It is this exact math of the 140 year old half point for a draw system that makes it logical for GM's to have 55%+ draw rates. Since GM's tend to be very logical, there is no surprise that we have an inflated draw rate relative to computers who have not been programmed to take this sort of stuff into account. Computers just play the best move that they can find. Changing the point system, changes the breakeven thresholds for when to accept a draw, when to avoid a draw, etc.
Generally speaking, we get what we pay for. Now, if all the other chess organizers out there want to structure the reward system to artificially boost the draw percentage, then that is certainly their choice and 140 years of tradition is hard to go against. However, if I am organizing an event and I don't want to pay for lifeless draws, I shouldn't be forced to use the obsolete point system. An unintended side effect of BAP is that the draws that do result are usually interesting, if not downright exciting, and at the very least instructive to see how a master tries to convert a better position into a win. Even as an A-player, if I set my mind to play to not lose, I can reduce my loss percentage dramatically, get lots of draws and end up winning games against the weaker players. That's how I became a teenage expert way back when. My draw percentage now is quite a bit lower and it is simply due to a change in my value system. I used to play for rating points as my metric, which of course encouraged me to draw unless I could be sure I would win. If I could do this, I know GM's can do it a zillion times better and they do. I used to have a higher rating, but really boring games. I now have a lower rating, but much more exciting games. There is no comparison at all. I never used to have a chance against masters, now I do. Of course I can lose to a B-player right after winning against a master, but the key is that every game of chess I play now is fun. Chess is a game. Games are supposed to be fun. Grinding out a draw was not fun for me and it was really a lot like work and that is actually the reason that I stopped playing for 20 years. Then I discovered gambits :)
As to the point on whether all this will keep 1% more of the elementary players playing chess, it certainly won't directly do that. However, if chess was on TV, or even had a lot more sponsors, there would be more money flowing into chess. Instead of $500 prizes at the weekend swiss, if it was a $5000 prize, then maybe 1% fewer 6th graders would completely stop playing. A small 1% reduction in attrition would DOUBLE the player base. Now, if we carried the ten times bigger prize funds into something like the World Open, I wonder how many players would return to the game?
$500,000 first prize in Open section
$200,000 class prizes!
All of a sudden, the equation of time allocation among former elementary chess players who are now adults will change, at least a bit.
Clint
I think my biggest problem with some of your comments was this view that technical chess is boring so you play one position basically eveary game as white after 1. d4 d5 2. Nc3 Nf6 3. Qd3 which I would consider to be quite boring to get one position with the idea of doing one thing (attacking) every game. The fact is there is beauty in all of chess and I don't think you will necessarily see more crazy attacks than you did before... Certainly I don't expect to see people deviating from the already inferior Nc3 with the move 3. Qd3, but I'm more than happy for you to encourage people to attack unsoundly. However I'm a little concerned that you're equating no draw with attacking chess worries me as the other parts of chess are at least as interesting as attacking.
In fact on another blog Clint suggested the reason I wasn't moved by his arguments is that I'm not a strong enough player to understand the nuances. He indicated that his system is designed primarily with masters in mind.
But Clint's opening choice, his own Class A rating, and the likelihood that (as both Josh and Dennis indicate) he clearly believes that wild attacks are the only type of chess that is interesting or beautiful -- all that suggests that Clint may have had the shoe on the wrong foot when he blamed lack of sympathy for his innovation on my insufficient understanding of chess.
Instead, it seems more likely that Clint's own limited understanding of the deeper aspects of chess (his great U.S. Open run notwithstanding), might be shaping his views about the way the game "should" be played.
My use of Qd3 is simply to get out of mainstream book positions, since I have never had the time, nor inclination to memorize the reams of opening "theory". I do half-seriously call it the Ballardsov and nobody complains about that since everyone thinks it is so bad.
Make fun of my play and openings and playing strength all you want, but when was the last time a past his prime 40-something outperformed his lifetime rating by 500+ points, if only for just 6 rounds and "only" against ~2200 players? I played with the accuracy of an A-player, but my performance was much better. This could indicate that I played a vastly superior opening, which you say it certainly isn't, so how can an aging A-player that plays an inferior opening, with little to no positional understanding, sloppy tactics and laughable endgame technique beat 4 ~2200 players in a row?
The answer that you are free to dismiss, but that I truly believe in, is that I played slugfest chess to learn chess so I can become better. Before you can be a strong positional player, you need to be a strong tactical player. I am certainly not a strong tactical player as I frequently drop pawns, entire pieces, even fall into mate in 1's.
I am now in about my fourth year of active tournament play. 2 years when I was a teenager and the last 2 years. There was no coaching or much of anything 25 years ago, so I am still very early in my chess evolution, unlike the clearly more superior and mature players that are already masters.
As far as Jon's comment about my limited understanding of chess, I am the first to admit I don't have a 32 piece tablebase. Then again, nobody has. Until we do, its anybody's guess as to what the correct way to play is.
What I know is that GM's not fighting to win everygame is no way to get closer to this correct way to play.
Clint