I didn't think there were professional tournaments using Fisher. So I'm not sure how it really works to do the comparison.

In addition, my sense from watching long Japanese matches is that the opening and midgame involve the professionals using more time than the endgame.

_________________
Occupy Babel!

Go watch the videos and see whether the pros are using their 30-second byo-yomi periods up to the end of the game and are pressed for time as they are trying to choose their next play. Short answer, they are. If they were using Fischer instead of byo-yomi, would they be accumulating time? Well we won't really know as long as they don't use that system. I will continue to think the answer is no.

The sense that pros use more time in the opening and middle game is true. But we are describing the problem not the answer! The pros have more time early in the game and they use it - too much of it. Otherwise we would not have the timing systems that we do. Think of all the commentaries that you have read that say something like "but (s)he was already in byo-yomi..." or "(s)he made a mistake in byo-yomi..."

Consider the possibility that the reason we amateurs use less time in the endgame is because we have neglected that boring stuff in our studies and are simply poor at it (this is a good description of my own case - again someone generalizing from personal experience ). Is the 'solution' to our poor endgame skills to systematically decrease the amount of time available at the end of the game?

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

Probably true, but it seems also possible that we use less time in the endgame because there are generally fewer points to be made. In the very late endgame, you're fighting over single-digit point moves, so if you make a mistake, it'll cost you less than missing a 20 point move in the middle game.

_________________
be immersed

In my experience, strong players often use up all their time for a byoyomi period even when their current move is entirely obvious (e.g. your opponent has just answered your ko threat and you are about to retake the ko). They use that time not to think about their current move, but to think about potential future lines of play. Effectively, they are redistributing their time themselves. I think that if they were using Fischer, they would just play and let it accumulate. They might then use it later, or they might not, depending on the flow of the endgame.

It seems that you have found a loop hole in the tournament classification. When A, B and C-classes were introduced, it was assumed that the overtime method would continue with the same speed until the end of the game. Now the Maidenhead tournament has 60 minutes main time and 15 minutes for the first 60 in the overtime, but later one does not get another another 15 minutes for another 60 moves. Instead one has to play 150 moves in the next 15 minutes.

I don't know if I would call it a loophole. Tournament class A also accepts 75 minutes sudden death, and this format provides more time than that.

It does not necessarily provide more time, but I can live with the fact that Maidenhead is in the A class.

Matti, were you involved in introducing that A/B/C classes and feel the accelerating overtime at Maidenhead is against the spirit, if not the letter, of the rules? I wasn't aware it was considered sneaky at all. As Herman says the comparison with the allowed 75-minute sudden death makes me feel it is legitimate, though you are correct that it doesn't necessarily allow more time, for example if you play only 9 moves in minutes 60-65 then you lose on time with the accelerating Canadian, but in sudden death you don't.

One problem with accelerating Canadian overtime is there is a perverse incentive to play more slowly (or fewer moves, e.g. avoiding kos) so as to postpone entering the faster overtime periods. With the 10 then 20 then 30 etc in 5 minutes I don't think this is such a problem, but with a British Open a few years ago the proposed overtime was 3 periods of 20 stones in 5 minutes (so A class) and then a final 2 minutes of sudden death (do you think this should be class A?). I pointed out that a problem with this system is a player who enters overtime early (say move 100, in a 340 move game) has 120 moves left to play (and that's assuming the opponent isn't mean and plays nonsense moves inside territory to prolong the game), but is only allowed to play 60 of them in the next 15 minutes and then has 60 more to play in the final 2 minutes. I suggested that rather than prescribing exactly 20 stones in 5 minutes (at which point the clock is reset and any excess time is lost) that it would be better to say "at least 20 stones in 5 minutes" as that way a player could play 20 stones in say 3 minutes, and then use the remaining time to play some more stones and thus not have to rush and play so many stones in the final 2 minute sudden death. Quite a few people liked this idea, and indeed I wasn't the first to suggest it: it's known as Milton Keynes overtime here in the UK and had apparently been used before. Others didn't though, one criticism being it is more complicated, another that punishing slow players is a good thing, and another that it's harder to referee/notice if your opponent has lost as the clock falling doesn't mean you lose if you are over the 20 stones (but that's a straw man as you can simply say that it does mean you lose: if you choose to play more than 20 it is your responsibility to ensure you stop and count out the next 20 before the 5 minutes is up). I don't actually remember what happened to the tournament, maybe the final sudden death was increased to 5 minutes.

I was involved. Ales Cieply provided data and I analysed it and we concluded that longer thinking time reduces variance in players' performance.Sudden death was not included in our proposal, but was proposed in the meeting and ad hoc decision was to accept 75 minutes sudden death in the A class.