Greengard's ChessNinja.com

No Sleep Till Brooklyn

| Permalink | 4 comments

Just a quick note to say I'm back home and that "daily" will be back in the Daily Dirt again as soon as I pay off my massive sleep deficit. Now I need to go over all the Kasimdzhanov-Adams games too. Even the cool as the other side of the pillow Mickey Adams has shown plenty of nervous play in the final. Kasimzhanov has played the strongest possible field in Tripoli, one of Linares caliber, and is more than holding his own. And yet I haven't seen a one win from him that I would call an excellent game, although I haven't had a good look at the final games yet. It's not Kasimdzhanov's fault; I blame the silly semi-rapid time control.


Not to open a can of worms, but I am tired about the "semi-rapid" comment about the FIDE time control. I agree, it is not perfect, but it is about, what? 110 minutes for 40 moves? Instead of 120 minutes? Yeah, too fast. And all these blunders in the "classical" zeitnot, when a player had to make 10 or more moves in one minute were "exciting", but now the blunders are the proof that the new time control does not work.

I agree, in the endgame the time is still too little, even after the 15 extra minutes. Maybe it should be 30 minutes, instead. However, that is a moot point; sometimes the inertia is too difficult to overcome. (What? 40 moves in 120 minutes instead of 150? No adjournments? Your are killing the chess - Someone in 1990)

Bad math. It's only 110 minutes for 40 moves if you make 40 moves! Have you ever played at this time control or watched an event that uses it live? Half the time the players are down to increment by move 25 or 30. You often see players down to five or ten minutes on move 30 or 35 in classical chess, but then they get that big, juicy hour added and they can relax and think. It's a huge difference, especially when you are used additional time controls.

It's not so much the total minutes, it's the nerves created by playing in a perpetual state of time trouble that create so many blunders. The endgame is obviously destroyed, but as we've seen in KO after KO (and Olympiad), the middlegame suffers tremendously as well. The number of serious blunders in these KOs is as bad as a rapid event.

This is all an interesting debate, but more convincing is simply asking what the positive results are of the time control change. Can't think of any myself. A few years ago I researched these KO events. There are't fewer draws, the chess is worse, sponsors aren't interested (I doubt Ghaddafi cared about the time control!) and compacting two rounds into one day has been resisted forcefully. (That would be a terrible but practical benefit of shorter games for organizers.)

I'm all for experimentation. The trend of shorter time controls is a permanent feature of the chess world. But this one, eliminating extra controls with increment, has been proven a disaster. I also think a line must be drawn somewhere to preserve the quality of the game. It's depressing to think that the best chess humans will ever play is already behind us. Computers forced the elimination of adjournments, so there's no way back there. But since most chess games are viewed long after they are played it would be nice if they were good ones!

Yes, I have been directing a few tournaments with this time control. And also have been on tournaments with the classic control, suffering Walter Browne time scrambles.

Let's face it: Many players would use all the time allotted, causing zeitnots. And the 30-second increment is better than the classical zeitnot, and infinitely better than the sudden death time limit (which is forced in the classical control, given the need to finish a game in less than 7 hours without adjournments). Besides, sudden death is the cause for most headaches for us arbiters (see article 10.2!)

OK, lets start from scratch, and design the "ideal" time control. Which would be the requirements?
1.- One session for all the game.
2.- Maximum time, no much more than 6 hours.
3.- Some kind of increment or delay, at least in the last control of the game, to prevent ugly zeitnots and games lost on time. (Even Nigel Short recently acknowledged this)
4.- Some players prefer having some time added after move 40 or so, to take a long breath before entering the endgame.

From there you can design a control that leave everyone happy. The three variables are:
1.- Initial time allotted.
2.- Additional time added at second (or even third) control, if any.
3.- Increment or delay on each move (from start or after the second/third control).

Obviously, the most defining variable would be the increment, as the other two would be limited by the fixed session time. So, which is a convenient increment?

The USCF has its own view: 5-second delay. This is fine, as it allows winning trivial endgames without risk of losing on time. However, it also means that writing down the moves is prohibitive. So, claims based on score sheet (50 moves rule, 3 repetitions) are almost impossible.

FIDE prefers 30 seconds, which seems the minimum still allowing writing down the moves. So, no need for special rules. Seems logical for me.

From which point should be time added? Starting from the 1st move is the most logical choice, as it also prevents first control zeitnots.

Once you get here, there is not a way to depart too much from FIDE control.

So, how would be distribute the initial and second control times. Let's say the longest game would last 100 moves, and let's give 6 hours (360 minutes) for it. It means you have 260 minutes left, or 130 minutes for player. You can divide it in many ways: All at start, or save some time for a second or even a third control.

For me, 100 minutes at start plus 30 minutes at move 40 (without 3rd control) seems logical. It means 120 minutes at move 40 (the same as classical), plus some time to take a breath before entering to the endgame. For me, it looks like a good compromise.

FIDE control is a bit faster, at 90 minutes initial time, plus 15 minutes at move 40. Previous to this championship, it was only 90 minutes initial time. It is encouraging that FIDE added that little second control, but it still is somewhat fast for me.

All these considerations were taking into account the rules and the convenience from an organizational point of view. The players most possibly have another view. One problem is the inertia. For a player, changing a habit may be difficult. Even if you give MORE time than classical (say, 110 minutes initial plus 40 added at move 40, w/30 sec. increment), he may feel hurried by a "semi-rapid control". Besides, FIDE has made so many blunders when trying to forcefully implement the increments, that most players would hardly agree to what is a good idea in principle.

About speeding up the games, this proposed time control does little reduction, as the game would last about 6 hours anyway. It only eliminates zeitnots, sudden deaths and most games lost on time.

OK, enough rant. Gotta work!

It can be made much simpler:

Ihr 20m each on the clock initially and an increment of 1 minute per move right from the start.

The theoretical 100 move game would then take 6 hours if each player used their time to the full.

Tom Rose

Twitter Updates

    Follow me on Twitter



    About this Entry

    This page contains a single entry by Mig published on July 12, 2004 11:20 PM.

    └ la Recherche du ╔checs Perdu was the previous entry in this blog.

    Gimme a K! is the next entry in this blog.

    Find recent content on the main index or look in the archives to find all content.