Increment of 5

Running Chess Tournaments
1

Chess friends I respect have recently pulled me into a debate on time controls, and particularly on the use of increment 5 (we are talking regular chess here, not blitz). One side argues that increment is meant to provide additional thinking time and that an increment shorter than 20 seconds should never be used; if you want to use a 5 second increment, you should use a 5 second delay instead. The other side argues that if you are going to use a 5 second delay, then using a 5 second increment is at least as good and probably better.

I can see both sides.

Thoughts?

2

The difference between 5 second increment and 5 second delay is minor. Indeed, if a player spends at least 5 seconds on every move, then there is no difference at all.

Typically, the difference appears only in time pressure. Even then, it is rare for someone to gain much time. Unless there is an obvious move (e.g. recapture), then chances are it takes a few seconds to find a reasonable response. And there goes most or all of the added time.

Michael Aigner

3

My basic feeling is that, except for blitz, delay should be for reaction time and increment for thinking time. Thus, I am opposed to delay greater than five seconds. I also don’t see a point to increment less than about 30 seconds. For blitz I do like to use a two second increment, but blitz is very different.

Alex Relyea

4

David Bronstein’s original idea on the use of delay was twisted for the convenience of a few organizers of tournaments. Bronstein said that players should play 20 minute games with a non-accumulating 15 second delay. If 20 minutes was not enough time to play, then he was not averse to a 2 hour game with 15 second delay. I notice that experiments are being done in some tournaments to allow a 10 second delay instead of 5 second delay, coming closer to Bronstein’s idea. He was opposed to the increment idea of accumulating time. He said that you should not gain time for nothing. A few repetitions, a couple of quick obvious moves, can add time under increment. This can be abused by crafty players.

One interesting facet of Bronstein’s delay method was to play, within the 4 or 5 hour period normally allotted for “serious” games, a 4 game mini-match with the opponent. Game 20 with 15 second delay would lead to 40 move games lasting about an hour in length. Instead of agonizing over a poor position dragging on interminably, the player gets a chance to put it aside and start a new game. If you think a match is a better test of skills, then Bronstein’s methods agree with you. Bronstein was interested in popularizing chess, in making it more enjoyable and understandable by the general public. Quicker play, the type normally seen in club games, was more fun.

5

One good reason to prefer increment over delay, at least in the USA, is the way delay is implemented on the most popular clocks.

The Chronos does everything well, but I’ve been hearing (on other threads) that that clock may be near extinction, with prices up to $300, etc.

Its likely replacement will be the DGT North American, which in my opinion is the best affordable clock currently available. But it has a crummy way of displaying delay. The word “delay” (in tiny print) is displayed for 5 seconds, before the main time starts counting down. You have no way, to know, at a glance, how far into your 5-second delay you are.

The Saitek is perfectly horrible in this regard, as well. It displays the delay seconds instead of the main time, in the very same area of the display normally used for the main time. That can be extremely confusing in time pressure.

The Excalibur Game Time does delay quite well, but I keep hearing rumors of that clock’s demise, as well.

On top of all this, there is ongoing disagreement as to the best method of displaying the delay. Should it be done as “straight” delay, or as non-cumulative addback (Bronstein style)? They are mathematically equivalent, but players have strong preferences one way or the other. Some like “straight” delay so that they can use 4.9 seconds on each move, to optimize the efficiency of their time usage. Others prefer Bronstein because a single display always tells you exactly how much time you have remaining. Ironically, proponents of “straight” delay are thwarted by clocks like the DGT North American, which never tell you how many delay seconds you have remaining.

Because of all this, and for other reasons, I tend to prefer increment over delay. For Blitz this preference is strong, with G/3 inc/2 being my favorite. For quick- or regular-rated events, if the delay or increment is 5 seconds, it doesn’t make a whole lot of difference. With 30 seconds, you’re into more leisurely time controls, for which increment seems better suited than delay.

Bill Smythe

6

If you are looking at your clock nervously, then you just lost a second of your 5 second delay. You’re only distracting yourself! Instead, you must get used to how long those 5 seconds really take. I recommend that time pressure prone students practice by playing 5 minute blitz with a 5 second delay.

This is just a very minor reason to dislike the DGT North American. And this opinion comes from an owner of two functional Chronos clocks.

Michael Aigner

7

Chronos clocks have been available at wholesalechess.com for $115 on and off for several months now. They will get 10-60 and they sell out in a couple days, then they get some more, and they sell out, and then nothing for a few weeks, and then it all repeats. They got a big shipment a bit over a week ago, and had both the button style and touch style, in black, white, and blue all in stock for over a week. They are now down to just white and blue button style clocks, and blue touch style.

A couple other sellers have received Chronos shipments over the last few months, but no one has gotten them as often as wholesalechess.com, and wholesalechess.com seems to have always got them when the others did, and has the best price, so anyone looking for a Chronos should probably just watch wholesalechess until they get more in (assuming the ones they have right now aren’t suitable).

For DGT clocks, it looks like the new DGT3000 is good from a delay point of view. It has both US delay mode and Bronstein delay mode. In US mode, it is like the Excalibur. It shows main thinking time and delay separately and simultaneously. It’s more than twice as much as the DGT North American ($99), but in addition to the better delay handling, it also has more digits so can show hours, minutes, and seconds. It’s the replacement for the DGT XL, and so has the electronic board support and the very flexible custom timing modes.

8

Is that true even if the delay is 2 digits, i.e. 10 seconds or more?

Bill Smythe

9

The delay display has 3 digits, formatted as M.SS, so it should handle any reasonable delay, and quite a few completely ridiculous delays.