[pescetom]

Given that things are rather quiet, how about this one.
Both sides playing 2/1 with strong NT.

MP

---

After West makes the contract, South calls you to complain about the very slow double (acknowledged by EW).
How do you evaluate the situation and how do you proceed?
Would you look at the full hand diagram with double-dummy analysis if you had it available, and would it effect your evaluation if South had done so before calling you?

[barmar]

You find peers of West to poll. First ask them what they would consider and do after the double without telling them that it was slow; this provides you with the logical alternatives. Then tell them that the double was slow, and ask which of these actions they think are more likely to be successful given what it suggests about doubler's hand; these actions violate Law 16.

You only look at the hand diagram to decide if NS were damaged. If they can't make 2♠ they were getting -200 or worse, so -140 is a gain and result stands. But if they could have made the contract, you need to determine if pulling the double was suggested by the break in tempo.

[smerriman]

barmar, on 2025-January-08, 17:12, said:

If they can't make 2♠ they were getting -200 or worse, so -140 is a gain and result stands. But if they could have made the contract, you need to determine if pulling the double was suggested by the break in tempo.

"Could" make 2♠, or "would" make 2♠? This is tough to determine in itself, and definitely not based on double dummy.

[blackshoe]

smerriman, on 2025-January-08, 17:57, said:

"Could" make 2♠, or "would" make 2♠? This is tough to determine in itself, and definitely not based on double dummy.

Law 12C1(b) said:

The Director in awarding an assigned adjusted score should seek to recover as nearly as possible the probable outcome of
the board had the infraction not occurred.

"Probable" implies "could", not "would".

[Cyberyeti]

blackshoe, on 2025-January-09, 11:34, said:

"Probable" implies "could", not "would".

Not sure that's true, if you have to play a 5% line not a 60% line to make it you don't get ALL of the score for making it.

[smerriman]

Other way around, probable means would (what was most likely to happen had there been no infraction), not what technically could have happened.

[pescetom]

 barmar, on 2025-January-08, 17:12, said:

You find peers of West to poll. First ask them what they would consider and do after the double without telling them that it was slow; this provides you with the logical alternatives. Then tell them that the double was slow, and ask which of these actions they think are more likely to be successful given what it suggests about doubler's hand; these actions violate Law 16.

We may well go there (and perhaps more), but for me you are jumping the gun: I would first ask about 1NT and Double. If 1NT denies support and they can prove that Double is takeout, then I would just remind East about tempo and leave the table.

And if they cannot prove that Double is takeout, then I guess we do need to poll but I have little doubt about the outcome, except perhaps whether 3♥ was even an LA to the obvious Pass.

But all this begs the questions for the more experienced players reading here: what should Double be and what does the extreme hesitation suggest ? They won the tournament, if that is relevant.

[sanst]

FWIIW: E has given a pretty good description of his hand: no hearts support, no MF, possibly a stopper in spades. After west's 2♡ E could raise with two hearts, but doesn't, instead doubles. If that's take out, I'm wondering what he would have done after 3♧ from W. E could even put 2NT or 3◇ on the table. No, I don't believe it's take out but penalties. W has made clear what his hand is and should have passed. The hesitation makes clear that E isn't sure about the double and W is making use of the UI. So 2♤x making for NS.

[pescetom]

 sanst, on 2025-January-10, 04:00, said:

FWIIW: E has given a pretty good description of his hand: no hearts support, no MF, possibly a stopper in spades. After west's 2♡ E could raise with two hearts, but doesn't, instead doubles. If that's take out, I'm wondering what he would have done after 3♧ from W. E could even put 2NT or 3◇ on the table. No, I don't believe it's take out but penalties. W has made clear what his hand is and should have passed. The hesitation makes clear that E isn't sure about the double and W is making use of the UI. So 2♤x making for NS.

For once we almost agree
So long as they do not have an agreement that this is takeout, however odd it looks.
If 1NT does not promise a spades stop then I would want to be told that (and why it was not alerted, if under our regulations). I would want to know if it denies fit too, and what 3H instead of double would be. But ultimately it makes little difference, at least this time round.

[blackshoe]

smerriman, on 2025-January-09, 13:27, said:

Other way around, probable means would (what was most likely to happen had there been no infraction), not what technically could have happened.

Okay, so the director has to be certain that the offender would make 2♠?

[mycroft]

As quoted, 12C1b:The Director in awarding an assigned adjusted score should seek to recover as nearly as possible the probable outcome of the board had the infraction not occurred.

How does the Director do that? Well, in most of the world anyway:

12C1c said:

An assigned adjusted score may be weighted to reflect the probabilities of a number of potential results, but only outcomes that could have been achieved in a legal manner may be included.

Weighted assigned score - designed specifically to solve this argument.

Absolutely, we look at "could make", and then we decide how likely it is that it "would make". We add a little bit of vigorish for the NOS, and then we assign the weights to 2♠x= and 2♠x-1 (and any other possibilities, I guess).

Then we matchpoint both of those scores against the field, multiply by the weights and add together, and compare that to the MP score for -140. If it's better, we assign it. If it's not, we rule "no damage" and let the table score stand.

How do we work out weights for "would make"? Any possible way that the director considers reasonable. Yes, that includes DD results (and "okay, does it make sense to finesse the 8 round 1?"), table results (although who knows if those match the information (the auction, mostly) that the table director has), director's opinion, polling of players of the (adjudged) declarer's skill (do we use the "field results" when this North is known to be the best declarer in the room? How about if this declarer couldn't resolve a dummy reversal if the opponents gave it to him?), and probably more.

All this, of course, assumes that we have ruled that pass is a LA in this auction with these players and their system. Having said that, 1NT to almost all implies a spade stopper, and after a natural NT, doubles by the NTer are penalty. I would need either "local knowledge" of something different or serious evidence that this isn't the case - a slow "could be penalty" double *always* invites pulling, even if the hesitation is "searching memory for the notes that it's takeout". Now, in the EBU, with their idiosyncratic (but solid) Alerting regulations over Double, the Alert or lack thereof would be more evidence of the system agreement. Most of us don't have that, though.

[smerriman]

blackshoe, on 2025-January-10, 09:44, said:

Okay, so the director has to be certain that the offender would make 2♠?

Of course not; you quoted yourself it says probable, not certain. But the weight is based on the probability you WOULD make it - the fact the opponents made an infraction does not mean you get to cheat, look at their hands, and make the contract just because double dummy says you COULD. And likewise, if double dummy said -1, but there's a chance you would have made it, you get some proportion of that, even if double dummy says you couldn't. Either way, double dummy is irrelevant.

[blackshoe]

I didn't say anything about "double dummy".

I wouldn't give much weight, perhaps none at all, to the "field results". After all, as you say, mycroft, who knows how things went at other tables?

Perhaps I'm reading too much into it, but the implication here seems to be that we should always weight the adjusted score. I don't think that's true.

[smerriman]

blackshoe, on 2025-January-10, 15:01, said:

I didn't say anything about "double dummy".

Ah, OK - pescetom asked about double dummy, I responded about double dummy, and you quoted my post about double dummy when replying, so I assumed you were.

[mycroft]

Yes, I think we should always weight the score.

Now, it's just that most of the time, the weighting is either obviously "100% this" or close enough that "give a little vig to the NOS" makes it "100% this".

But on this hand in particular, if we believe that Pass is a LA demonstrably suggested by the slow double, the contract is going to be 2♠x N, and there are going to be a range of results in that contract. I haven't looked myself (frankly, I haven't put forward more than a cursory response to "is Pass a LA?"), but the general discussion here is that "sometimes it'll make, and sometimes it'll go down 1".

And that is *exactly* where 12C1c comes in. Work out good values for the "sometimes", and see if the adjusted score is better for the NOS than the table score.

The old 12C1e ("best result that was probable for the NOS, worst result at all possible for the OS") was deemed to be prejudicial to the OS, and frequently "a windfall" for the NOS, so even in the land of "we have pros who will play to and maybe over the line to get their clients results, so we have to nail them to the wall hard when they do go over to rein them back in", it was eventually determined that "no, what the rest of the world does is better even there, and definitely in the 90% of rulings that don't involve bridge lawyers".

And while the glorious (not really, but the odd one was good for the cynical soul) days of "double bad" 12C1e rulings is gone, it really is easier to sell the weighted ruling than even the "double bad"s.

[sanst]

 blackshoe, on 2025-January-10, 09:44, said:

Okay, so the director has to be certain that the offender would make 2♠?

It might be made if N plays E for his actual hand, minus one is rather likely.

[pescetom]

What attracted my attention to this sequence (apart from the fact that it cost me as South a bottom, with no redress from Director) was that 3♥ is not even an LA. Such was my judgement at the table at any rate and so said a poll on TOS too: 100% would pass the double (having received no mention of delay).

[blackshoe]

It would be nice if every ruling made were to be the exactly correct ruling for the circumstances.

It would be nice, but it ain't gonna happen.