http://www.chess.com/news/karjakin-is-o ... match-4133

Chess has caught up with the format I have dubbed Win & Continue, which goes back in go to the early 20th century in Japan, though I suspect go got it from martial arts.

And by all accounts it has been a terrible failure as a chess format. The nature of the game, with the preponderance of draws and the advantage of the white pieces, does not make this format suitable or exciting in the same way as when it is applied to go.

It's hard to see how it can work, for fans or players. Since there are so many blitz tie breakers, you're essentially watching two formats at the same time. Serious games and mickey mouse games. That seems unsatisfying to me, but perhaps I shouldn't knock unless I've actually tried observing it. Are there any of these tournaments happening now or soon on twitch or youtube?

I'm not sure this has to be the case. The format is for a team event but there is also scope for rewarding individual effort. In go this is usually, on top of the team prize, a big prize for someone who wins five in a row.

If we modify this slightly to take account of chess's draws, we could envisage a format in which a player gets a substantial extra prize for every game he wins. In the case of a draw, both players drop out (i.e. Win and Continue literally). That would incentivise both players to avoid drawish lines so that they can stay in the match and earn more prize money. As I understand it, White's first-move advantage in chess shows up mainly in his extra ability to direct the opening into drawish lines. Once one or, better as here, both players commit to risky lines, first-move advantage becomes hard to demonstrate.

The advantage of this idea from a sponsor's point of view is that blitz replays are not necessary, except in the case of the very last game, but then it becomes an exciting and highly televisual climax.

Chess players being gamesters, I can foresee the possibility that some will try to disrupt the purity of the match by, say, accepting draws under team orders to force a player on the opposing side to drop out, but that can be dealt with through the magic of the sponsor's invitation cards for next time, or by upping the individual prizes.

I don't think there is any good way to get around the problem that in chess a draw is a likely outcome. Dropping both players in the case of a draw would just create different and probably worse problems. For example, in a clearly drawn position, both players would have an incentive to toss a coin to decide who resigns.

That doesn't seem to make sense. If one resigns that means one wins, and he so continues. In my solution both drop out (the draw stands) and both teams move on to their next man. As I said before, the only time that becomes a problem is in the very last game, but then the problem becomes an asset because that final game can be replayed as a televised blitz.

Is anyone aware of any research on the effect of rule changes on the draw rate in chess? For example, if they changed the rules to make stalemate a loss instead of draw, what happens to the draw rate? This would seem quite possible to do with the strength of current chess programs.

_________________
Dave Sigaty
"Short-lived are both the praiser and the praised, and rememberer and the remembered..."
- Marcus Aurelius; Meditations, VIII 21

Suppose I am in a drawn position. Now there are two scenarios: I drop out and take my opponent with me, or one of us drops out and the other continues. In the second scenario, if the winner is decided by a random coin toss, then statistically this helps my team as often as it hurts my team, so even if I am motivated purely by team altruism I need not feel guilty. And if there is some significant individual reward at stake, then it is clearly better for me to risk the coin toss.

Getting back to the actual event where this format was applied, the reason for the failure seems to be quite different than the generic issue of draws or the advantage of playing white.

Classical time control chess is an entirely different beast than rapid or blitz. For each classical game the players have large teams investing dozens of man-hours to prepare for the game, and generally they know beforehand what types of positions they will end up playing each day. When the classical game ends in a draw, you suddenly switch colors and go into a rapid format, where you are "on your own" so to speak, with no specific preparation.

The Chinese players, due to the structure of the Chinese training system, excel in classical controls utilizing team preparation. Without this specific preparation, they are much less experienced than the elite Russian players. Classical and rapid really are different games in chess, much more so than in go.

I see where you are coming from now, but we are still on different tracks because you are assuming the players get to decide the rules. I would always start with the premise that the organisers/sponsors would try in advance to rule out coin tosses or any other unwanted manipulations.

But this aspect of ending up at a different destination depending on where you start takes on bigger significance in in cdybeijing's report. If I understand him correctly, he is reporting that it is the Chinese who are regarding this particular Sino-Russian event as a "failure", whereas others, such as the Dutch reporter in the original link, seem to think it is a success. I would infer from that that the Chinese only regard an event as successful if they win it. Cynical but plausible, and it makes me wonder yet again how much effort they are putting into tilting go events in their favour. We already know they support front-line players in international events by providing coaches to do research on foreign opponents, one reason I suspect the Japanese are left floundering. Are locations and tournament formats given a similar amount of forethought?