Greg on Chess: Stop the Draw
By IM Greg Shahade   
August 16, 2011
Greg200.jpgGreg on Chess is a new series of CLO editorials by IM Greg Shahade, founder of the US Chess League and the US Chess School. His latest column was on "The Round Robin with a Swiss Twist." Greg's opinions do not reflect any official USCF views and we encourage discussion in the comment thread or on CLO's twitter, and facebook fan page,

In American sports, no one likes ties. The National Hockey League used to have frequent ties for many years, but about ten years ago they caved and instituted shootouts, so that every game would have a winner. In poker, deals at the final table to share the money used to be completely standard, but when poker hit big time and ESPN, such deals at TV final tables were eliminated or at least kept hush-hush.

Approximately 50% of professional chess games end in a draw. This extremely high draw rate does hurt our chances for commercial success. I suspect there have been quite a few media outlets that had some interest in chess in the past, and then lost a good chunk of it when they realized that the game ends in a tie so often.

One of the most disastrous things that happened in the past 10-15 years of chess, was when GM Garry Kasparov, the same man who has done so much to promote chess to the mainstream, took a draw in a man vs. machine match live on ESPN. The chances of chess making the big time plumment when the final game of a dramatic match with millions of potential chess fans watching ends in a draw in a complicated position. Anyone who thinks that draws don’t hurt us is living in a dreamland. If that type of match was played without the draw being a possible result, it would increase the chances of chess enjoying a popularity boom.

Recently, at, GM Rustam Kasimdzhanov put forth the controversial idea that draws must be eliminated (also see the reader response article). He suggested that any time a game ended in a draw, the colors should reverse, and you should play a new game with something like 20 minutes per side. If that game is also drawn, again play a faster game until someone wins.

I agree with Rustam’s sentiment, although I’m not a big fan of jumping into rapid play right away. Because draws are harmful to chess's popularity, I think serious steps should be taken to lower their frequency. I will propose a few solutions, one being extreme, and two being a bit more modest.
Solution 1: Start the game normally, whenever the game ends in a draw, you reverse colors, and you keep the same clock time from the previous game. This continues until there’s a winner.

As you can see, this is similar to Rustam's idea but I like it more because the games flow into each other, rather than being distinct entities. For instance, if you are playing a time control of 40/2, 20/1, SD/30, as soon as the first game ends, all of the left over time will carry over to the second game. So if you don’t finish the first time control, both sides will get to keep all of their remaining time and a bonus of 1.5 hours. This will often result in a second serious chess game at a relatively slow time control. At some point, if the draws keep coming and coming, the time control will get faster and faster, but it’s definitely not a guaranteed blitz game, simply because the first game ended in a draw. It also adds a whole new dimension to things, as people will have to play a little bit faster than usual in the first game (mainly if it gets to one of the later time controls) or else fear starting the second game with not much time. Also there should always be increment of at least five seconds in this type of game, as this is enough time so that even if you start the game with a minute, you still have the opportunity to play semi-reasonable moves.

This time control would be extremely interesting and fun. It would result in many great games, and would also result in a lot more tension, drama, and excitement. The players with white will be loathe to give a draw and have to play the next game with the black pieces, and so you will see people routinely doing everything they can to squeeze every last drop of play out of a position, especially with white. Because you get all of the time from the later two time controls when you start a second game, players won’t have to be too careful to move quickly until the very last time control, because as long as the game is drawn before move 60, they will always get at least 30 minutes for the next game.

 The downside to this idea is that players will have much longer games on average, and will be more “tired”. While I do agree that playing chess can be a tiring and stressful activity, I think that the complaining by players on how tiring it is to play a bunch of chess games is a bit ridiculous. If you are so tired from playing long games either relax a bit during the game, get in better physical condition or find a new occupation/hobby!
 I do suspect that this solution will be too radical for most people, but I’d sure like it, and hope that someone puts together a serious tournament with this type of system. In fact my sister and I have already put together an event with this structure, and you will be hearing more about it soon.

Solution 2: Start the game normally and whenever the game ends in a draw before move 40, you reverse colors and keep the same clock time from the previous game.
This system will simply be aimed at stopping short draws. Black may still want to draw relatively quickly, as then they would get a shot at victory with the white pieces but whoever has white will certainly not want a quick draw. I think this would be a good compromise solution, as it would stop all short and aimless draws, yet extremely hard fought , long and legitimate drawn games are more likely to simply be counted as draws, which is fine.
This system has a few benefits over the Sofia rules or the “no draws in under 30 moves” rule. The problem with these rules is that the final decision as to whether or not a draw should be allowed usually falls to the arbiters. However it is impossible to find arbiters who are able to judge repetition claims by elite players. For instance, if two elite players go into a repetition on move 30, how can an arbiter who is not a grandmaster fairly judge whether this is a forced repetition or an attempt to skirt the rules? A possible solution to this is to have a "Grandmaster" advisor or advisory board, but even this is flawed. The GM board could easily be either too harsh if he or she is using computers or too lenient, if they desire to avoid controversy. And because there are no codified rules on this, most organizers and arbiters seem to lean toward avoiding controversy, which gives players all the more incentive to draw if it helps them. Note that I don't really blame the players for this kind of thing, and rather think the rules should be designed in a way that doesn't allow for obvious loopholes.

Solution 3: If you offer a draw, your opponent has three choices. Accept the draw. Refuse the draw. And a new third choice: Switch sides and keep playing!

I think that this is an amazing idea. It does nothing to harm the quality of play, and also practically forces players to play on if they feel they have even the tiniest edge.   All of those quick draws that Grischuk kept offering with the white pieces in the World Championship Candidates cycle would now be fraught with danger. I’m quite sure that Kramnik would have been extremely happy to play those positions with the white pieces and a miniscule edge. I cannot imagine a solution which has as many positives and as few negatives as this one. The only downside to this rule is those situations where both players clearly benefit hugely by drawing, then they will usually find a way to split the point.

Someone told me this idea years ago, but I don't remember who. I think it’s brilliant, so if you know who came up with it, please leave a comment and the article will be updated to give credit where credit is due! 

I hope I convinced some of you of the value in coming up with creative and aggressive ways to prevent draws. I applaud the London Chess Classic for using the 3 points for a win/1 point for a draw scoring system, even if I don’t believe it’s the greatest solution. In attempting to curb draws, experiments are always welcome! Let me know what you think of some of the ideas above and look out for the TV pilot I created with Jennifer and Daniel Meirom to see one of the above ideas in action.

Also stay tuned for my next article on the Manhattan Open, August 17th-21st. which will be a recap of my comeback to competitive chess (you can read a little more about my approach on my personal blog).