[0 carbon]

Since the Mitchell and Howell movements preordain who plays who in each round, why does it have to be in lockstep? Why wait if your next opponents are available?

When both pairs next to play each other are both available, why not let them play? This could give players a little extra time to finish their next round, minimizing the need for adjustments, and speeding the entire tourney.

Just run out the clock on the slowest table(s) each round to keep to the max time for a tourney.

-- tom

[pescetom]

Not yet downed enough coffee to think clearly, but isn't this going to lead to monumental pauses when a frontrunner pair is due to play the slow clock driven pair (or one of their recent victims) towards the end of the tournament?

[mycroft]

Yes, you will eventually get what happens IRL when the (fast) players take over the running of the rounds - but instead of "Sorry I can't get you a board, they're playing it" (because online, there are infinite sets of boards), you get "Sorry, I can't move you to the next table for 10 minutes, they're the slow pair".

Now I don't mind (as long as I'm prepared for it) the Calgary Flight A tradition that "after the skip" (which, now that we web, means round 3) "if they're ready, let's start", as long as they're not obnoxious about getting a board "before the round is called". But they just need to know that eventually, they will pay for it with an "unscheduled break". Usually - not always, but usually - they're pretty chill with "yeah, you play fast. You've finally caught up to the people playing to time."

Some aren't - "well, why don't you pull a board?" "Because *they're* on time. *You're* 15 minutes ahead."

I do expect explaining that last online to be harder than IRL, especially when the rounds "just keep coming" at their pace to begin with. Plus, the online interaction feels like it takes more time/effort than live, because they can't "see" the cards in the opponents' hands.

However, from a director's standpoint, I don't see what this gets you. Your game is still tied to the slowest pair/the clock; if everyone plays 13 minutes a round rather than 14, either set the clock low or let the game move when everyone's ready (which it does currently). *Some pairs* will get a faster last round; *some pairs* will get 3 or 4 rounds in at "their time" and then get pulled back to normal. (This is a serious question, by the way. I could very easily be missing something).

[pescetom]

Having slipped to the other extreme of the caffeine spectrum, an alternative accelerator for barometer Howell/Mitchell did come to mind: how about automatically varying the round duration in real time on the basis of early finishes? Many Directors do this manually playing f2f or on RealBridge that other playing platform, bumping the clock down a minute or two when it's clear that almost everyone is well ahead after one or two boards. The software could do it automatically, perhaps even in both directions to allow for that really tricky round (just this week at the club, in a non-barometer Mitchell we hit a boardset with two consecutive tricky slams that effectively added 1-2 minutes to most rounds, I wasn't going to stress them for that). Say recalculate the nominal time for each board to the 80th percentile of ranked effective time (or whatever). Of course it would need a few rules, like not counting All Pass (unless every table does, fat chance) or pairs with one or more robots (if there are human-human pairs). But I think it should be feasible and save time without stressing any but the slowest pairs.

[mycroft]

I mean, it is possible to add time to a round manually, if it looks like too many people are going to miss.
It is possible to reduce time to the rounds, if 90% are going faster than "normal".
The system will move everyone to the next round if *everyone* is done. Arguably, this would be a good thing to change to "when 90% are done, move what you can, and keep the deadfall counter going on the 10% left" - but I'm again not sure how much that will get you (and yeah, the 10-10-10-10 hand that every table but mine (playing Precision) goes All Pass is a special case that would have to be handled well).
I would feel uncomfortable if we were unduly rushing "those pairs" over and above "you're allowed 7 minutes per board. We expect you to be done in time" known at start of game.

FtF, I don't mind "leaning on the slow pairs" (and I absolutely do, if it's one or two) as long as it's clear that "you do have the time, just don't waste it". Even "pull the board for the next table" is a subtle prompt, and if it's done before the round is called officially, it's a pretty good subtle prompt. But I don't have the "round's over, we're DDing the result" axe hanging over their heads when I move to "19 minutes/round instead of 20 because most people can handle it" (and insisting on having a few always in Damocles territory).

[pescetom]

mycroft, on 2024-April-19, 11:39, said:

I mean, it is possible to add time to a round manually, if it looks like too many people are going to miss.
It is possible to reduce time to the rounds, if 90% are going faster than "normal".
The system will move everyone to the next round if *everyone* is done. Arguably, this would be a good thing to change to "when 90% are done, move what you can, and keep the deadfall counter going on the 10% left" - but I'm again not sure how much that will get you (and yeah, the 10-10-10-10 hand that every table but mine (playing Precision) goes All Pass is a special case that would have to be handled well).
I would feel uncomfortable if we were unduly rushing "those pairs" over and above "you're allowed 7 minutes per board. We expect you to be done in time" known at start of game.

FtF, I don't mind "leaning on the slow pairs" (and I absolutely do, if it's one or two) as long as it's clear that "you do have the time, just don't waste it". Even "pull the board for the next table" is a subtle prompt, and if it's done before the round is called officially, it's a pretty good subtle prompt. But I don't have the "round's over, we're DDing the result" axe hanging over their heads when I move to "19 minutes/round instead of 20 because most people can handle it" (and insisting on having a few always in Damocles territory).

Sure, I'm with you in the the FtF timekeeping philosophy, which is also what I used to do online with that other site.
Online with BBO I felt no need to speed them up (they were already getting close to 5 minutes per board, which is hilarious considering how they screamed about 8 minutes FtF) and I confess I never dared try to reduce time, I had too many experiences of tournaments exploding for no valid reason to risk that.

[0 carbon]

mycroft, on 2024-April-19, 09:48, said:

Yes, you will eventually get what happens IRL when the (fast) players take over the running of the rounds - but instead of "Sorry I can't get you a board, they're playing it" (because online, there are infinite sets of boards), you get "Sorry, I can't move you to the next table for 10 minutes, they're the slow pair".
...
However, from a director's standpoint, I don't see what this gets you. Your game is still tied to the slowest pair/the clock; if everyone plays 13 minutes a round rather than 14, either set the clock low or let the game move when everyone's ready (which it does currently). *Some pairs* will get a faster last round; *some pairs* will get 3 or 4 rounds in at "their time" and then get pulled back to normal. (This is a serious question, by the way. I could very easily be missing something).

That is what happens in Unclocked tourneys. I often adjust when just 1 table is left in a preceding round and 2-4 other pairs are waiting. You can see this in Tournament Status.

However, I meant in Clocked tourneys, Howell or Mitchell, if players are fast, why not let them start playing the next round if their predestined opps are available? The clock should still reflect the time left to round end

This would give them a little more time in the next round, and everyone still gets moved along when the Clock reaches zero. Then if they have a difficult board, they have extra earned time to deal with it.

This reduces delays and still keeps the tournament duration to the calculated time. They are going to play those players sometime. Why not sooner?

Unlike unclocked tourneys, slow players don't end up playing other slow players. They still move at clock=0 - or sooner! No negative clocks!

[mycroft]

That would be fine.

But I'm E-W 1. My partner and I are such that we will finish our 3-board round, against average pairs, in 17 minutes. The "slow" pairs are ahead of me, E-W 3 and E-W 4. I move early to table 2. I might move a couple of minutes early to table 3. But table 4 has just spent the first two rounds right to time, and if they're not the fastest, this round will also go close to time. Sure, I have 26 minutes on my "20 minute" clock, but I still finish in 17. So my clock is on 9, and I won't be moving until it gets to 1!

But I still have to sit at the computer (rather than go refresh my water, or do cat boxes, or...) just in case the round starts at 5, and now *I'm* the slow one (even though the clock will still be starting at 25 minutes)!

Sure, if it works out, especially at the end, we might breeze through the last few rounds and finish as much as 10 minutes early! Sure, I agree, if we - and our new opponents - have gained some clock, we've "earned" it for this round. But when it's only us - and it eventually will, because the pairings are fixed - we lose that "earned" clock equity, only this time 9 and 10 minutes at a time (but only twice in a 21-board session), rather than 3 minutes every round.

[0 carbon]

mycroft, on 2024-April-21, 12:52, said:

But I'm E-W 1. My partner and I are such that we will finish our 3-board round, against average pairs, in 17 minutes. The "slow" pairs are ahead of me, E-W 3 and E-W 4. I move early to table 2. I might move a couple of minutes early to table 3. But table 4 has just spent the first two rounds right to time, and if they're not the fastest, this round will also go close to time. Sure, I have 26 minutes on my "20 minute" clock, but I still finish in 17. So my clock is on 9, and I won't be moving until it gets to 1!

But I still have to sit at the computer (rather than go refresh my water, or do cat boxes, or...) just in case the round starts at 5, and now *I'm* the slow one (even though the clock will still be starting at 25 minutes)! ...

Since everything is in lockstep, you will seldom accumulate time. Your extra 3 minutes is unlikely to stretch to 6 or 9 or 12 since 1) your new opps may be slow, or their opps. 2) you're not playing other fastest players like in Unclocked 3) the round clock is inexorable.. But those 3 minutes may allow you more thought for a hard hand, or a quick biobreak. Just tell your P, "BRB" and let her say to new opps "Sherlock will be back in a minute or two" and you will still have the allotted time right there, instead of exercising your Kegel muscles.

[mycroft]

Okay, that I can see. And that's a good idea. But it won't "speeding the entire tourney", which was my responses so far. It'll mean that the fast players will have a few rounds with a couple of extra minutes, and likely have their waits go from "2-3 minutes, 7 of 9 rounds" to "6 or 7 minutes, once or twice; maybe a minute or so the rest of the rounds".

But their "earned time" is still dependent on their next round opponents; so it isn't really "earned".

The other issue, of course - which we shouldn't be considering, but people will - is "I get advanced warning of the hands" when all players are playing the same boards at the same time, but some start them early. Not sure it's a big deal over "by the third board of the round, we're starting early"; in fact, I can't really see an extra issue. But somebody will raise the C- flag, I'm sure.

[0 carbon]

mycroft, on 2024-April-19, 09:48, said:

...
However, from a director's standpoint, I don't see what this gets you. Your game is still tied to the slowest pair/the clock; if everyone plays 13 minutes a round rather than 14, either set the clock low or let the game move when everyone's ready (which it does currently). *Some pairs* will get a faster last round; *some pairs* will get 3 or 4 rounds in at "their time" and then get pulled back to normal. (This is a serious question, by the way. I could very easily be missing something).

Two benefits:

• The tourney can move a little faster - potential slow boards have more time to marinate
• Players may earn a little more time to use in the next round for a harder board.

[0 carbon]

mycroft, on 2024-April-19, 09:48, said:

Yes, you will eventually get what happens IRL when the (fast) players take over the running of the rounds - but instead of "Sorry I can't get you a board, they're playing it" (because online, there are infinite sets of boards), you get "Sorry, I can't move you to the next table for 10 minutes, they're the slow pair".

One way to use those waits is bio breaks - stuff coming in or out... But an estimate of the time available would be useful. So....


When a table finishes the boards in a round, give that estimate like "New boards in less than X minutes" where X is the time to the round end.


This could even be refined by using the average time that the coming opponents took in previous rounds to calculate "New boards in about X minutes"