A modest proposal regarding time control terminology

Running Chess Tournaments
1

Increment, delay, bonus, Fischer, Bronstein, cumulative, non-cumulative – how are these terms used, nationally and internationally?

In the USA, “increment” generally means time added cumulatively, while “delay” is used for time added non-cumulatively.

Over at FIDE, “increment” refers to time added either cumulatively (bonus) or non-cumulatively (delay). FIDE uses other terms, such as Fischer, Bronstein, bonus, delay, etc. to distinguish the two.

These differences have created problems at both ends. I propose the following for both USA and FIDE:

  • USA should gradually phase out the word “increment” for time added cumulatively.
  • Both FIDE and USA should avoid the use of proper names, like “Fischer”, “Bronstein”, ”USA style”, ”Euro style”, etc. Keep everything in lower case!
  • For time added cumulatively, use “bonus” instead of “increment”. “Bonus” is already a well known term both internationally and in the USA, as a majority of clock manufacturers now use this term on their clock faces and in their manuals.
  • Use a plus + sign (and phase out the abbreviation “inc”) in tournament announcements. The plus + sign is already a standard abbreviation everywhere. Organizers who wish to make it clear that plus + means bonus rather than delay can simply add the word “(bonus)” in parentheses, e.g. “G/90 +30 (bonus)”.
  • Continue to use the word “delay” for time added non-cumulatively. This term is already used both in the USA and within FIDE. Also continue to use (for example) “d/5” or “d5” in tournament announcements.

Now, how should we distinguish the two common ways of displaying (non-cumulative) delay? Sometimes the name “Bronstein” has been used when the delay time is added to the main time in a single display, while words like “straight”, “normal”, “regular”, or “simple” are used when main time is displayed by itself and delay is displayed with a separate count-down digit or flashing colons or flashing letters.

I object to “Bronstein” because it is a proper name, and also because that name is sometimes used simply to distinguish delay (regardless of display method) from bonus.

Further, I object to “straight”, “normal”, “regular”, and “simple” because such words imply that the other method is somehow “crooked”, “abnormal”, “irregular”, or “complicated”. We need a non-pejorative lower case description.

I once proposed “discrete”, but as Tom Doan noted, that sounds like “discreet” which is almost the exact opposite.

So how about:

  • Refer to the two methods of displaying delay as “single display” (aka Bronstein) and “dual display” (the USA method).
  • This distinction need not appear in the tournament announcement. Just call it “delay”, abbreviate it as “d/5” or “d5”, and let it go at that. Let the players use whichever method their clocks provide.

Oh, and one more thing:

  • If there is a secondary control, use a semicolon to group and separate the time control elements. For example, “40/90 SD/30; d/5” if the delay is to apply to both controls. If the delay is to apply only to the second control (shudder), write it as “40/90 d/0; SD/30 d/5”.
  • A similar remark would also apply to bonus, of course.

What do you think of those ideas, Baba Looey?

Bill Smythe

2

I can live with ‘bonus’ even though it is less descriptive than ‘increment’. FIDE’s imprecision in its choice of terminology has polluted the water, but there are many chess terms that have multiple meanings. ‘Bonus’ has a fairly specific meaning in the US Chess ratings formula but not in FIDE’s ratings formula. ‘Bonus time’ might be an improvement, at least it makes it clear that you’re talking about time rather than ratings points.

With regards to the last suggestion, US Chess’s time control formatting does not currently allow time controls where the delay or bonus is different in one time control period than in another. Allowing that might impact whether an event is eligible for dual rating or not. (I think dual rating has outlived its purpose, but that’s a subject best handled separately.)

I think the parsing code already knows what to do with a semicolon, though. However, as US Chess gets closer to the point where it starts defining what the next generation of tournament reporting and rating software looks like, that’s the time to start looking at what time control structures to allow, so that we know how to parse them.

Interestingly enough, earlier today I was talking to someone and suggested the possibility of a time control such as 40/60,d30;SD/30,+30. The logic was that being able to build up time (which bonus does but delay does not) is more important in the later stages of a game than in the first X moves. Working backwards, if you want rounds that are 3 or 4 hours apart, what initial time control would you use, assuming d30 in the first time control and +30 in the second?

3

Neither term is inherently descriptive, i.e. there is nothing in either English word that inherently means cumulative rather than non-cumulative.

Having both bonus time and bonus rating points is hardly pollution. Those two uses are so far apart that they don’t really interfere with each other.

What the U.S.Chess time control formatting allows or does not allow may not be the same as what the rules (at least the FIDE rules) allow. Occasionally one sees international events where the bonus does not kick in until move 41 or move 61.

And even if U.S. Chess rules require (rather than just strongly encourage) any bonus to be in effect in all time controls, it still makes sense to encourage the use of a semicolon to eliminate any possibility of confusion. The unpunctuated “40/90 SD/30 +30” looks as though it could mean either “40/90 SD/30; +30” or “40/90 +0; SD/30 +30”.

The new short Chronos actually enables that possibility, and is most likely the only clock to do so. Bonus and delay are independent of each other and can change from one control to the next. All of the following are available options with the short Chronos:

  • G/90 +30 d/5 (single control with both bonus and delay)
  • 40/90 d/5; SD/30 +30 (two controls with delay in the 1st and bonus in the 2nd)
  • 40/90 +30 d/5; SD/30 + 10 d/3 (two controls, each with both bonus and delay, with 2 different bonuses and 2 different delays)

I think I’ll run a Short Chronos tournament. Everybody who wants to play must furnish a short Chronos so the game can be played with the specified time control.

Bill Smythe

4

I virtually GUARANTEE you that some players will (intentionally or otherwise) think ‘bonus’ has to do with ratings points if we start using that term to refer to time controls, but as I noted earlier we’ve got more than one chess term that is, to use a programmer’s term, overloaded. It’s more of a ‘deal with it’ problem than an insurmountable one.

I haven’t looked at the time control parsing code in a while, I don’t know how much work it would take to have it understand having different delay/bonus in each time control. The current assumption is that whatever the delay/bonus is set to is constant from move 1 on. That makes it easier to compute total time for a 60 move game, which is the basis for deciding whether it is dual-rated or regular-only (MM+SS > 65).

If we assume that any time control period has to be at least 30 minutes for regular/dual rated sections (which I don’t think is an actual requirement in US Chess rules or procedures, just something that I thought was once in the rule book, though I can’t find many recent events that weren’t blitz or quick-only that might not have at least 30 minutes in a time control), then nearly any realistic two-stage time control that has an increment or bonus of at least 1 second in at least one stage would have MM+SS>65, so maybe that’s not as big an issue as I thought it was upthread.

5

If the U.S. Chess rating software assumes (for rating purposes) that any listed bonus or delay applies to the entire game, then that works for me. There should be no need to modify the rating software.

In my post I suggested that organizers might want to add the word “(bonus)” in parentheses to any time control that uses the plus + sign as an abbreviation for bonus – at least temporarily until players get to know that the plus + sign means cumulative time added. I guess I agree with your suggestion that “(bonus time)” might be even better, at least for a little while.

It was once a rule, before the mm+ss concept was invented. Then 30 became the lower limit for mm+ss. This prompted a handful of organizers to run events at G/5 d/25 and still have them rated as regular/dual. IMHO this is an abomination. There should be a minimum of 25 for mm alone, as well as the minimum of 30 for mm+ss.

Bill Smythe

6

In the TD Tip after rule 5b1c that I helped write and was recently added to the rulebook, I mention that bonus is another commonly used term for increment.

TD Tip after rule 5b1c:

TD TIP: There are two forms of delay, “simple” (or “US”) delay and “Bronstein” delay. Both are mathematically equivalent, equally acceptable for use under the rules, and just differ in how the delay is displayed on the clock screen. Simple delay separates the delay time from the main time, often by showing the delay countdown in digits or having something like the colons or the word “delay” flash each second during the delay countdown. Bronstein delay, by contrast, combines the delay time and the main time into a single display. If a player uses the entire delay time or more for a move, the clock will add the delay time to the main time. If a player uses less than the delay time for a move, the clock will go back to the time it had at the beginning of the move. The fact that the main time will never go above what it was at the beginning of a move with delay is what makes it different from increment. With increment, (also commonly known as “Fischer” and “bonus”), the increment time is added to the main time each turn, regardless of the amount of time it took the player to move.

7

The only thing the ratings system itself currently needs to know is which of the 6 ratings systems to use (OTB: regular, quick, blitz; ONL: regular, quick, blitz) and whether dual rating also applies for an OTB regular event. The time control information (and the online/otb flag) unambiguously determines that.

When displaying events on MSA, we want to show a more complete description of the time control, including whether it varied from one round to the next, so if we’re going to start seeing multi-stage time control events with different increment/bonus in stages, the parsing software should be modified to deal with it. I am now pretty much convinced that won’t affect how we decide if an OTB event is dual-rated or regular-only, so the parsing is largely just to standardize the notation shown. Bill’s notation with the use of a semicolon is essentially the same one I proposed in 2012, I think, but at the time the idea of increment/bonus varying by stage was not something we saw in the US. (I’m not sure clocks even supported it back then.)

Something I’m hoping the next reporting and event history database structures can support is tracking which schedule a player was in. There’s currently no way to report it, and I’m not sure WinTD even tracks that after the schedules are merged, I think Thad Suits told me that SwisSys does track it.

Bill, want to work on standardizing how events where the time control varies from one round to the next and multiple schedules are described? In the former, everybody plays at the same time control in any specific round, in the latter they do not.

8

I’m reasonably sure the DGT 3000 (and similar) clocks can do this. I’ve never actually tried it, but for each time segment you have to specifically select “Fischer” to get the increment, so it makes sense that you can choose “US Delay” instead for a segment.

I know in the rating report you can report the multiple schedules. You have to do it manually, of course. I don’t know if this information is actually saved anywhere. I’ve always thought it would be nice if the system would use that info to select the appropriate rating system for each schedule. That way the faster games which can be dual rated (or quick only) would actually be dual rated or quick rated instead of just regular only.

9

Whatever you see when editing the event is what gets placed on MSA. For an example of a section with different schedules see uschess.org/msa/XtblMain.php?202207317812.2

What it does not currently support is having a different increment/bonus setting for a secondary time control. I think we can safely assume that such an event is regular-only rated, which might simplify the coding somewhat.

10

Someone has to say it.

No.

11

This topic was automatically closed 730 days after the last reply. New replies are no longer allowed.