[shevek]

---

NS a decent pair. This was their first serious game together.
EW indecent.

South assumed system off after DONT double, North guessed otherwise.
Silly-looking pass of 2♦ by West.
South explained the possible misunderstanding before East led.
East had a problem.
♥Q lead for -520.

What score or scores do you award?

[RMB1]

shevek, on 2011-October-30, 03:57, said:

NS a decent pair. This was their first serious game together.

South assumed system off after DONT double, North guessed otherwise.

Why oh why did North guess, or not say he was guessing. "Undiscussed: could be system on or system off", and there would be no ruling.

In my world, South has not used UI. South knew it was undiscussed whether 2♦ was natural, 2♥ suggests partner has guessed otherwise, so the alert does not tell South anything useful.

Is East claiming to be misinformed: how many hearts does he think there are?

Score stands.

[shevek]

RMB1, on 2011-October-30, 04:24, said:

Why oh why did North guess, or not say he was guessing. "Undiscussed: could be system on or system off", and there would be no ruling.

In my world, South has not used UI. South knew it was undiscussed whether 2♦ was natural, 2♥ suggests partner has guessed otherwise, so the alert does not tell South anything useful.

Is East claiming to be misinformed: how many hearts does he think there are?

Score stands.

In fairness to East, bidding 3♥ over North's 2♥ might be read as a stopper-ask or a big Michaels hand. East is not a strong player. She might even have imagined that NS's hearts were 5-1. The transfer created a problem she was not competent to solve.

[RMB1]

shevek, on 2011-October-30, 16:02, said:

She might even have imagined that NS's hearts were 5-1.

I count eight hearts in East's hand.

shevek, on 2011-October-30, 16:02, said:

The transfer created a problem she was not competent to solve.

True. She may have been unable to solve them if North bid 2♥ after 2♦ was explained as "undiscussed".

[iviehoff]

RMB1, on 2011-October-30, 04:24, said:

In my world, South has not used UI. South knew it was undiscussed whether 2♦ was natural, 2♥ suggests partner has guessed otherwise, so the alert does not tell South anything useful.

I don't like this argument. The alert tells South exactly what is going on; without the alert South is less sure and might get it wrong. Maybe partner has an off-shape NT with long hearts. Maybe partner has just discovered that both the red suits in the hand are hearts. I don't you can be permitted to use information of a bidding misunderstanding revealed by UI unless there really is no plausible alternative.

Whatever. In an auction affected by neither UI nor MI, EW will declare a contract on a substantial number of occasions and make at least 10 tricks; they won't always bid game.

[RMB1]

iviehoff, on 2011-October-31, 07:47, said:

I don't like this argument. The alert tells South exactly what is going on; without the alert South is less sure and might get it wrong. Maybe partner has an off-shape NT with long hearts. Maybe partner has just discovered that both the red suits in the hand are hearts. I don't you can be permitted to use information of a bidding misunderstanding revealed by UI unless there really is no plausible alternative.

For me it just a question of odds/frequency. At the table and on-line, when I see the auction 1NT-(not pass)-2♦-(Pass)-2♥ opener bid 2♥ because he thought 2♦ was a transfer. On-line, where there are partnerships who have not discussed these sequences and there is no UI, these auctions occur all the time.

As responder, I would bid on that assumption because (for me) it is overwhelming likely that is what is going on, and it is wrong to buck those odds. If others have different experience, they will make different assumptions and bid differently. If those with different experience are peers of South in the OP then South will be constrained by the UI.

[iviehoff]

RMB1, on 2011-October-31, 08:14, said:

As responder, I would bid on that assumption because (for me) it is overwhelming likely that is what is going on, and it is wrong to buck those odds. If others have different experience, they will make different assumptions and bid differently. If those with different experience are peers of South in the OP then South will be constrained by the UI.

It is all a question of when the misbid is overwhelmingly likely to have been revealed. It is like the discussion we had on a 3C response to checkback/Crowhurst. Is this an off-spec bid people plausibly make on off-spec hands that turn up from time to time, or is it overwhelmingly likely to be a misunderstanding? Plenty of people were prepared to argue that they would "make up" a 3C response to checkback from time to time. To my mind, strong NT openers overrule their partner's weak bids from time to time, because they sometimes have a strong suit of their own (and in fact he does have one, albeit not hearts). In the present case, you aren't playing with RMB's regular partner who has never been known to do that, you are playing with someone for about the second time.

[barmar]

Since North alerted the 2♦ bid, the UI trumps "odds/frequency". Without the alert, you have to guess what partner's 2♥ is, and you're allowed to guess right. But the alert tells you for sure what he did, and now you're prohibited from guessing that if there's another logical alternative. So unless you claim that partner would NEVER bid 2♥ unless he thought 2♦ was a transfer, you have an alternative and you must choose it.

[barmar]

RMB1, on 2011-October-30, 04:24, said:

Why oh why did North guess, or not say he was guessing. "Undiscussed: could be system on or system off", and there would be no ruling.

In some areas, most decent strong NT players play systems on over X and 2♣, so it's not an unreasonable assumption. If it were me playing with another player of similar level, and for some reason we hadn't made it to the 5th line of the ACBL CC when we were discussing system, I'd probably say something like "Undiscussed, but probably systems on, so a transfer".

This is kind of trick in ACBL territory, where you announce transfers. But what do you do with a "probably transfer"?

[mycroft]

Having said all of that, assuming UI, what happens?

South bids 2♦ to play.
North accepts the transfer that she announces.
South looks at the 2♥ correction and with 1=6 in the red suits says "no, partner, it won't play better in hearts."
North sees the "both reds, GF", and bids the obvious game.
South passes before the doubling starts - after all, some would have bid 3NT directly over 1NT-X.

Now, was the MI corrected before the opening lead - and if so, how? And having 7 tricks and an entry, knowing dummy wanted to bail rather than GF, are we going to look for partner's entry, or lead the ♥Q? If we look for partner's entry, are we going to underlead the spade ace to find it (yes, I know, it turns out that Ace-and-out works better. Are you going to give up your side entry on the hope?)

This sure looks like a poll question. Like barmar, this is something that anyone with 5 minutes to go over a card would have had an agreement on in the ACBL (as the "System on over..." line is immediately below the NT range) - whether they *remember* what they agreed, I don't know. I would also expect that a pickup pair of any sort in the ACBL would be playing "system on over artificial doubles" - likely "system on over any doubles of a strong NT", unless they had specifically discussed something else. That doesn't mean that they won't forget in the heat.

But ask people of East's calibre what they lead to 3NT with the correct explanation ("we've never played before and didn't discuss this.") I note especially that that should get them the right answer to what South has, given that if it is a transfer, East can see 15 (or 14, I guess it could be stiff K) hearts. If enough people find the spade lead then either -6 or -9, depending on which spade got led (I'll give them the heart switch even on a spade underlead). If nobody does, or almost nobody, then either -3 (heart A, spade switch), or making 6 (heart A and heart, or minor) or 7 (heart not-A).

How the poll will guide me: in the ACBL, we're in L12C3e territory, so if "anybody" finds the spade lead, then it's at all possible and E-W get -lots; if a decent fraction find it then it's "likely" and so do N-S (I'm tempted to give them -3 anyway, as HA and spade switch seem reasonable). In the rest of the world, it's a matter of what percentages I give the various results.

[c_corgi]

I don't think 3NT is the obvious game from North's point of view. If a forcing 3H was available it would be automatic. It may be that he was catering for South to hold D rather than H+D, and it would be interesting to know if there was any UI which suggested this.

[barmar]

I think most would play 3♥ as confirming a ♥ fit (and possibly showing a max), not temporizing. The normal bid when partner shows two suits and forces to game, but you don't have fits for either of them, is 3NT. Although I suppose with no real stopper in one of the side suits, he might prefer to play in the 5-2 heart fit, but then he should probably jump to 4 -- there's no way responder would know that opener is looking for a spade control for 3NT.

[campboy]

barmar, on 2011-October-31, 10:18, said:

Since North alerted the 2♦ bid, the UI trumps "odds/frequency". Without the alert, you have to guess what partner's 2♥ is, and you're allowed to guess right. But the alert tells you for sure what he did, and now you're prohibited from guessing that if there's another logical alternative. So unless you claim that partner would NEVER bid 2♥ unless he thought 2♦ was a transfer, you have an alternative and you must choose it.

You're all right until the last sentence. But you certainly do not need to claim that it is impossible for partner to have done this thinking 2♦ was natural, merely that it is so much less likely that no-one in their right mind would guess that way. If you have a choice between action A, which is right 90% of the time, and action B which is right the remaining 10% of the time, and the costs of being wrong are similar, then it just isn't a logical alternative to choose B, even though it could work out better. For the situation in question 10% is a massive overestimate IMO.

[mycroft]

I stick with my first answer. Even if 2♥ is "5=2 in the reds" after a natural 2♦, with 1=6, "partner, hearts won't, in fact, play better than diamonds." So, I think pass fails the LA test on the "nobody's playing the 6-card fit when a good 8-card is available" whether or not it fails the LA test on "nobody's going to believe that 2♥ isn't 'partner thinks I transferred'".

[Vampyr]

campboy, on 2011-November-01, 17:18, said:

If you have a choice between action A, which is right 90% of the time, and action B which is right the remaining 10% of the time, and the costs of being wrong are similar, then it just isn't a logical alternative to choose B, even though it could work out better.

A 10% action is most definitely a logical alternative in the UK. I do not have the White Book to hand, but I believe that a logical alternative is one that 15 out of 100 peers would consider, of which one or two would actually do it.

Which raises a question -- can anyone tell me how a logical alternative works when, say, only 5% of players would consider an action, but all of them would do it?

[phil_20686]

It seems likely that east would bid 4H if not confused by the misinformation, or worried that her partner would misunderstand the misinformation. Obviously this could be established by a poll of peers, but after 1N x 2d p p or 1N x 2d 2S p I would presume a large fraction are bidding 4H to play.

Since that makes it seems like the adjustment should be to 4H =, rather than to 3N- lots.

[jallerton]

Vampyr, on 2011-November-02, 11:20, said:

A 10% action is most definitely a logical alternative in the UK. I do not have the White Book to hand, but I believe that a logical alternative is one that 15 out of 100 peers would consider, of which one or two would actually do it.

That's not correct, Stefanie. Law 16B1[b] defines the term:

Law 16B1b said:

A logical alternative action is one that, among the class of players in question and using the methods of the partnership, would be given serious consideration by a significant proportion of such players, of whom it is judged some might select it.

In England, further guidance is available from the EBU White Book:

EBU White Book 16.6.1 said:

What is a “significant proportion”? The Laws do not specify a figure, but the TD should assume that it means at least one player in five.
If fewer than about one player in five of a player’s peers would consider the action then it is not an LA.

Vampyr, on 2011-November-02, 11:20, said:

Which raises a question -- can anyone tell me how a logical alternative works when, say, only 5% of players would consider an action, but all of them would do it?

As per the above interpretation, 5% is not a significant proportion, so the 5% action is not a logical alternative.

[campboy]

Quote

If you have a choice between action A, which is right 90% of the time, and action B which is right the remaining 10% of the time, and the costs of being wrong are similar, then it just isn't a logical alternative to choose B, even though it could work out better.

Vampyr, on 2011-November-02, 11:20, said:

A 10% action is most definitely a logical alternative in the UK. I do not have the White Book to hand, but I believe that a logical alternative is one that 15 out of 100 peers would consider, of which one or two would actually do it.

Ok, suppose I found 100 people and offered each the following bet. They choose either A or B, we then observe the first ball drawn in the UK national lottery next weekend and depending on the value I either pay them ten pounds or nothing. If they chose A I will pay them if the number is in the range 1-44; if they chose B I will pay them only if it is in the range 45-49. How many of my 100 people do you think would seriously consider choosing B?

(I use this method rather than, say, a ten-sided die merely to avoid anyone basing their choice on the suspicion that I am manipulating the odds in some way.)

[Vampyr]

campboy, on 2011-November-02, 16:44, said:

Ok, suppose I found 100 people and offered each the following bet. They choose either A or B, we then observe the first ball drawn in the UK national lottery next weekend and depending on the value I either pay them ten pounds or nothing. If they chose A I will pay them if the number is in the range 1-44; if they chose B I will pay them only if it is in the range 45-49. How many of my 100 people do you think would seriously consider choosing B?

(I use this method rather than, say, a ten-sided die merely to avoid anyone basing their choice on the suspicion that I am manipulating the odds in some way.)

Apparently the standard in the UK is that B would have to be 40-49 to make it a logical alternative. I was sure that earlier it was as I wrote; but maybe I am wrong.

Anyway, it seems that the "1 in 5" player does not have to take the action, just consider it.

[campboy]

I don't think you understand. Whether an action is an LA is judged principally on the proportion of players who would seriously consider it. The chance of an action being successful is something completely different.