2021 World Championship Match Time Control Question

Running Chess Tournaments
1

In the rules for the 2021 World Championship Match, it states: “The time control for each game is 120 minutes for the first 40 moves, followed by 60 minutes for the next 20 moves and then 15 minutes for the rest of the game with an increment of 30 seconds per move starting from move 61.”

On the DGT 3000 and VTEK 300, if you set the clocks for 40/120,20/60,SD/15-with inc30 only on the SD/15, the clocks starts giving the increment time once White has completed move 60.

On the Leap KK9908, Chess Evolution Classic, longer Chronos, and Wholesale Chess Advanced/Leap PQ9912 clocks, if you set the clocks for 40/120,20/60,SD/15-with inc30 only on the SD/15, the clocks starts giving the increment time once White has completed move 61. (On the Leap KK9908, and Wholesale Chess Advanced/Leap PQ9912m you have to choose preset 6 for this to happen. If you try setting the clock manually for 40/120,20/50,SD/15-with inc30 only on the SD/15, it will start giving the increment time only when the main time in the first two periods have run out, regardless of how many moves have been made).

Which do you think is correct and why? I think the way the DGT 3000 and VTEK 300 do it is correct. If the time control had the increment starting from move 1, the players would get the base time in the first time control and the increment time at the start of move one (for example, if the time control was 40/100,20/50,SD/15 with inc30 from move 1, the players would start move one with 1:40:30 (one hour forty minutes and thirty seconds, which is the base time of 100 minutes in the first time control and the increment time of 30 seconds). Thus, with the increment coming into play starting on move 61, the players should get the increment (and the base time in the third time control) after they have completed move 60/at the start on move 61.

2

I agree. As soon as white presses his clock after move 60, he gets 30 seconds added for move 61. This is analogous to giving 30 seconds “after move 0” (i.e. when the clock is set initially or when it is first started) for move 1 – which, as I hope everybody knows by now, is the correct way to do it.

This fails to give increment for move 61, and is consistent with the way these other clocks handle increment generally, i.e. they fail to give increment for move 1. And the solution is the same. Instead of “40/120 inc/0, 20/60 inc/0, SD/15 inc/30”, set them for “40/120 inc/0, 20/60 inc/0, SD/15:30 inc/30”.

Incidentally, just as white should get 30 seconds added when he presses his clock after white’s move 60, black should get 30 seconds added when he presses his clock after black’s move 60. I assume the clocks handle this part correctly.

Bill Smythe

3

What is the point of giving incremented time? It is to ensure that a player need not ever have a flag fall so long as they continue to move timely. So the real question is why is increment not applied to the non-sudden-death periods as well? What’s sacred about the 40/120 and 20/60 periods that they should not likewise be subject to not having the flag fall so long as the player continues to move timely. Actually the real question might be why have the move limits at all and simply make it G/195 inc/15?

But as to which I think is correct, it’s really neither in a USCF-rules tournament and I think for the first sentence above. 7th Edition: “5E1. Increment or delay with mixed time controls. If a mixed time control includes increment or delay, the increment or delay should apply starting with the first move of the game, and the increment or delay time should be the same for all time control segments in the game” (p. 13)

I can’t quote the FIDE rules, I’m sure somebody might if there’s an applicable section. But to answer your question as you asked it, one shouldn’t start receiving the increment until the first move of the next time control. (Move 1 in SD does not occur until the sixtieth move is completed, so therefore it should be move 61, and giving the time at complete of Move 60 is tantamount to having given the increment before the first move of a SD is made then giving another at completion of move 1.) I may be wrong.

4

The Chess Evolution Classic automatically gives the increment for move one. The Leap KK9908 and Wholesale Chess Advanced/Leap PQ9912 automatically give the increment for move one under most of their settings.

By the way, the reason I got thinking about this subject in the first place was because I was playing around with the Xflyee clock I just got (which overall is a great clock). Preset 21 on that clock is 40/120,20/60,SD/15 with inc/30 only on the SD/15. The clock starts giving the increment to the Black player after he completes move 60 but doesn’t start giving the increment to the White player until after he completes move 61.

5

Ooh. This definitely sounds wrong. It cheats white out of one increment time compared to black.

A decent alternative might be to give both players an increment when black presses his clock after move 60, and then white gets another after move 61 (i.e. before move 62), black gets another after his own move 61 (i.e. before 62), etc.

Bill Smythe

6

Yes – at the start of the first move of the next control. The player is given a 30-second increment when he presses his clock after move 60, so that he will have the increment for move 61. Then, when he completes move 61, he is given another increment so that he will have it for move 62.

– which is correct, so that the player will have that increment for move 61 –

– which is also correct, because then it’s giving another increment for move 2 (i.e. move 62).

Here’s a simpler example. 40/60 SD/30; inc/30. Just two controls, and, unlike the above example, the increment applies to both controls.

The clock should add the increment for move 1, i.e. it should show 60:30, not just 60:00, when white’s clock is started at the beginning of the game. That way, he gets 40 increments added by the time he presses his clock after move 40. If the clock did not add the increment for move 1, and the player exceeds the time limit at move 40, he has received only 39 increments, thus has been cheated out of one increment and has time-forfeited unfairly.

Analogously, if the increment begins at move 61, then this increment should appear at the start (not the end) of move 61, so that it will available to the player for move 61.

I agree, however, with your main point:

Both U.S. Chess and FIDE now prefer that the increment start at move 1, i.e. that it be in effect during non-sudden-death as well. But this organizer has apparently gone outside the box. Having the increment only for the final control creates a huge difference in the “look and feel” of the game after move 60.

I would hate this. Whenever the total control is more than 2 hours, the first control should be 2 hours. I hate it when a dead-lost opponent sits there and thinks forever. Forcing him to at least make move 40 helps out here.

Bill Smythe

7

Doesn’t that depend upon what “starting from move 61” means? One could reasonably interpret that either as add-before 61 and add-after 61.

8

Doesn’t that, in turn, depend on what “add-before” and “add-after” really mean? I have seen these terms used in two different ways.

In one pair of definitions, “add-before” means that white’s clock increments when black presses, and vice versa, while “add-after” means that white’s clock increments when white presses, and similarly for black.

In the other pair of definitions, “add-before” has meant simply that the increment also applies to move 1 because the clock increments at the start of the game as well as between moves, while “add-after” means that the clock increments only between moves.

Under the first pair of definitions, almost all clocks add after. Apparently there are only one or two that add before, and one or two others that can be set either way.

Under the second pair of definitions, the available clock brands seem to be split about 50-50. Those that add after should, strictly speaking, be set with one increment added to the main time (e.g. G/60:30 inc/30 instead of G/60 inc/30) to compensate.

It is virtually certain that an “add-before” clock in the first sense also adds before in the second sense, but the converse is not always true, because the “add-after” clock needs an extra step in order to apply the increment to move 1.

Bill Smythe

9

True, although FIDE uses terms like “starting from move 1” for increment as well and you get the increment applied at the start of move one so I would interpret “starting from move 61” to mean you get the increment applied starting at the beginning of move 61.

10

Yet that doesn’t happen in sudden-death only tournaments. You make your move one, press the clock, and get the time at the press - Black does not get their increment time until they’ve pressed at the end of move 1 either so they would be penalized unless you add fifteen seconds to both clocks before move 1 (i.e. G/30 inc/5 clocks should be initially set to 0:30:05 then). I know absolutely I’m being pedantic at best and rules-lawyering at worst here. (EDIT: I see you addressed this below, and add to this after the quote-cut…)

I see what you mean here, though… My question is which clocks actually do that - assume the first press starts the time of the other player and adds an increment to them automatically at the start of the move? I know there have been threads here in the past and Micah is a clocks maestro and might be able to say. I’ve never played with a clock that didn’t increment the pressing player as the key is pressed. OTOH I’ve got my new-ish Wholesale Chess clock (3 years old but I have only played a handful of delay games on it d/t nursing school) and I think it might do that… I’ll test if Micah doesn’t chime in… I know my old Chronos didn’t.

Thanks for the info - was looking for this.

Actually I agree with you mostly. IIRC there always was kind-of and now explicitly is an out in USCF for a player to ask for a delay-of-game penalization and in an egregious case (mate in 1 no escape, for example) the director could use discretion to wrap it up IMVHO. Dangerous, but in such a case would be warranted. But I see what you mean as far as dead lost, no immediate mate, and let the clock play out… The thing I would do is again approach the director, get out my leisure reading and/or MP3 player and just sit there doing other things with my life while the other player tries their little mind game - which is denied and therefore ineffective. For the World Championship, though, I hope the players would be of a little higher social caliber and not do such things, and I question the effects of move controls at that level of play. How many excellent GM games have we seen go blunderful disasters because of move limit time pressure?

11

Apparently we now agree, or almost do, on most points.

Micah knows the specifics better than I do, but for the most part, the DGT-type clocks usually do this, but Chronos and Excalibur do not.

There are two ways a clock can do this. One is to simply show the increment time added to the main time when the clock is set. For example, with G/60 inc/30, as soon as the clock is set, it shows 1:00:30 (instead of 1:00:00) on both sides, before either button has been pressed to start the game.

The other way is to add the increment time to both sides when the clock is first started. In other words, when it is sitting there waiting for the game to start, it will show 1:00:00, but when white’s clock is first started, both displays will switch to 1:00:30. After that, 30 seconds will be added to each player when that player presses.

I suppose there could be a third way, but I don’t know if any clock does this. Start out at 1:00:00, but add 30 seconds to white’s side when white’s clock is first started, then add 30 seconds to black’s side (as well as to white’s, for move 2) when white presses after move 1, then revert to “normal” thereafter.

On at least one clock (one of the gray ones with a rocker arm), one of the presets (if you look on the bottom of the clock) is shown as “1:30:30” for G/90 inc/30. That’s fine, but unfortunately, that clock doesn’t have a preset for G/60 inc/30, so I presume you’d have to set that one manually by entering “1:00:30” rather than “1:00:00”.

There is a P.S. for the DGT North American. That clock shows only hours and minutes (h:mm) until the time drops under 20 minutes, so if you use that clock for G/60 inc/30, you won’t be able to tell whether it “shows” 1:00:30 or 1:00:00 because it is actually showing only 1:00. You’ll just to take its word for it. You’ll know it almost immediately, though, after the game starts, because it will continue showing 1:00 for the first 30 seconds before it changes to 0:59.

Bill Smythe

12

This is how the DGT, Chess Evolution, Omcor GT960, VTEK 300 (if it’s in the FIDE mode, otherwise the clock doesn’t automatically give the increment for move one), Leap KK9908 (if you are using a setting that automatically gives the increment for move one), Leap PQ9912/Wholesale Chess Advanced (if you are using a setting that automatically gives the increment for move one), and Pursun-1688/Xflyee (if you are using a setting that automatically gives the increment for move one) do it.

This is how the Tap N Set, ZMF-II, and ZMart Pro do it if you have the updated firmware versions which allows increment to be displayed in the preferred “add-after” method (what these clocks incorrectly call “Bronstein”).

13

Speaking of Bronstein, this is what the Chronos does in its Bronstein modes (e.g. AD-1). These are non-cumulative addback, i.e. delay, rather than cumulative addback, i.e. increment. So, if you set it to set it to G/60 d/5, while it is waiting for the game to start, it will show 1:00:00 initially, but as soon as you start white’s clock at the beginning of the game, it will switch to 1:00:05 on both sides, and thereafter add 5 seconds (or time actually used, if under 5 seconds) after each move.

I’ve always found it strange that the Chronos does Bronstein correctly, but increment incorrectly.

Bill Smythe

14

Why? It’s very clear what the point of delay is—you get ss seconds to make the next move without losing any of your main time. It’s not at all clear why increment is “you get ss extra seconds to make the next move” vs “you get ss extra seconds for each move that you make”. I suspect that there are other games where the latter is the interpretation. I wouldn’t be surprised if 20 years ago (when the Chronos was programmed??), the latter was the more common interpretation for chess players (since after all that’s how secondary time controls in chess work—you get extra time for making a certain number of moves). Given that the only difference is whether an increment is given prior to move 1 or not (i.e. an “angels on the head of a pin” level argument), the only compelling reason for the FIDE interpretation is that including an increment before move one simplifies the calculation of the precise amount of time available to make 40 or 60 moves and there are regulations based upon those.

15

All that matters is the current FIDE and US Chess rule for increment is that the increment is applied before move 1. Therefore, the Chronos should adopt this.

16

I seem to recall that Chronos was out of business at one time. Are they selling and or manufacturing them again?

17

There are (tens of) thousands of Chronos clocks out there that aren’t going to change. If someone, really, really, really wants to make a big deal of of this, one can just adjust the initial time. I suspect that very few people care about an extra 30 seconds at the start of two hours.