[weejonnie]

'Illogical, logical alternatives' have been discussed before - with the note that, in this case, SB belongs to the class of players who would consider 6NT as a logical alternative, given the bidding sequence.

HOWEVER - the laws of bridge have now changed and any call made need NOT be a logical alternative

2017 laws (16B1a)

"(a) A player may not choose a call or play that is demonstrably suggested over another by unauthorized information if the other call or play is a logical alternative."

2007 laws (16B1a)

(a) After a player makes available to his partner extraneous information that may suggest a call or play, as for example by a remark, a question, a reply to a question, an unexpected2 alert or failure to alert, or by unmistakable hesitation, unwonted speed, special emphasis, tone, gesture, movement, or mannerism, the partner may not choose from among logical alternatives one that could demonstrably have been suggested over another by the extraneous information.

This definition does NOT say that the call or play actually made need be a logical alternative, just that there are logical alternatives to it. In this case of course the 6NT isn't demonstrably suggested by the unauthorised information and so is allowed.

[gordontd]

weejonnie, on 2017-April-06, 04:22, said:

This definition does NOT say that the call or play actually made need be a logical alternative, just that there are logical alternatives to it.

I believe that was the reason for the new wording.

[Manastorm]

It seems that 6NT is more likely to make with the remark than without. Surely the law doesn't allow it.

[lamford]

Manastorm, on 2017-April-06, 06:29, said:

It seems that 6NT is more likely to make with the remark than without. Surely the law doesn't allow it.

Drivel. You could say the same thing about 7NT.

[weejonnie]

Manastorm, on 2017-April-06, 06:29, said:

It seems that 6NT is more likely to make with the remark than without. Surely the law doesn't allow it.

The law allows it - what the law doesn't allow is for someone to base their call or play on the remark, if there are 'logical alternatives'.

In this case RRs comment* suggests that he has a 'good' invitational raise of 1NT. This means that 3NT is more likely to make than if there had been no comment. If SB hasn't an 'obvious' raise of 2NT to 3NT (i.e. one that 75%+** of his peers would make) then he has to pass. If he has an 'obvious' raise then he is allowed to make it. (In teams for instance it is more likely that 3NT would be called than in pairs.)

* RR is a scrupulously ethical player and wouldn't make such a comment; if it had been the Chimp, however, ...
** or however your RA defines a position where there is no Logical Alternative.

[lamford]

weejonnie, on 2017-April-06, 08:27, said:

* RR is a scrupulously ethical player and wouldn't make such a comment

He is scrupulously scatty and would not even realise it gave UI

[barmar]

axman, on 2017-April-05, 10:06, said:

What is the antecedent to 'the other call'?

"another" in the phrase "over another".

[barmar]

weejonnie, on 2017-April-06, 04:22, said:

'Illogical, logical alternatives' have been discussed before - with the note that, in this case, SB belongs to the class of players who would consider 6NT as a logical alternative, given the bidding sequence.

That's circular reasoning. You can't determine the class of player from the call that's made, because we're trying to determine which calls might be chosen from the class of player.

Quote

HOWEVER - the laws of bridge have now changed and any call made need NOT be a logical alternative

...

This definition does NOT say that the call or play actually made need be a logical alternative, just that there are logical alternatives to it. In this case of course the 6NT isn't demonstrably suggested by the unauthorised information and so is allowed.

The old law didn't say that the call or play actually made needs to be a LA, either. It says you can't "choose from among LAs", but doesn't say that you can't choose from among non-LAs.

However, as has been mentioned, EBU decided to interpret the definition of LA as always including the chosen action, even if no peers would consider it, just to avoid this logical conundrum. If EBU sticks with this interpretation, then the wording change to "if the other call or play is a logical alternative" doesn't really have any effect -- SB's 6NT call is still considered an LA.

But we still have the fact that the UI doesn't demonstrably suggest it over other LAs such as Pass and 3NT. So he's free to make the gambling bid in either version of the Laws.

[mycroft]

The problem is that at matchpoints, if we know in a real auction without UI that 2NT+however_many is going to be effectively a zero, and where we know that 3NT (the normal spot) will be rolled back on *this* level of UI, 6NT, if it makes (or if SB can make it) gains almost 100% of the matchpoints. And 6NT, if it goes down, loses almost 0% of the matchpoints. So, 6NT - knowing that it's this or 2NT - need only be a 3% or so chance for it to be the right action.

Therefore, it's a logical alternative, and I've just demonstrated how it was suggested over passing by the UI.

Sure, in vacuo, it's highly illogical. But it's not in vacuo; it's in context of the known requirements of the Laws.

Do I think it's obvious? No. Do I think it could be written better, to avoid this argument? Yes. Does this corner case need to be addressed, the two times a year it's possible? Yes, because there are people in the real world who will SB with the best of a certain North London Club, if they think there's even a 3% chance of getting away with it.

Do I think that the EBU regulation will stand, and will still apply come 2018? Of course. It's a real problem, it's the correct solution, and if it requires some careful translating to be legal, well, it certainly won't be the first time here or elsewhere.

[barmar]

mycroft, on 2017-April-06, 14:33, said:

Therefore, it's a logical alternative, and I've just demonstrated how it was suggested over passing by the UI.

Is it really suggested by the UI? Or is it suggested by the fact that the UI has constrained your actions?

[barmar]

The editorial in the March 2017 Bridge World talks about a related issue.

Quote

Law 16B1a cannot be an instruction to a player to take an abnormal action, because that would violate 73C; ...

Misinterpreting 16B1A in this way turns the method of adjustinc scores into a criminal procedure, one that induces wrongdoing. ... In an uncontested auction, N-S agree spades, and N, who is unlimited, makes a slam-try. Control-biding follows, and at one point N bids 5♠ after a noticeable change in tempo. S, who has learned for the first time that N has no grand-slam ambitions, must decide whether to pass or to bid 6. A player familiar with the adjustment procedure might think as follows: "I have received UI. If I bid 6 (an action suggested by the exdtra clue), and it makes, the Director, and eventually the AC, may deem that passing 5♠ was a sensible alternative and put the contract back to 5; if I bid 6 and it goes down, that score will stand. Unless bidding 6♠ is so clear-cut that everyone in the building would bid it absent the BIT, I am better off passing 5, whether 6 makes or not."

A S who followed this train of thought and therefore passed 5♠ when he believed that his normal action would have been to bid 6 would be in violation of 73C, because he will have taken advantage of the UI by allowing it to change his call in a quest for a higher score. Some people would call that a just desert for the legal system; others would call it cheating.

It seems like SB's thinking is essentially the same as the above declarer. But rather than pass and take his average minus, he threw caution to the wind and used the UI to try for a top, although a bottom was more likely.

[weejonnie]

barmar, on 2017-April-06, 15:39, said:

The editorial in the March 2017 Bridge World talks about a related issue.


It seems like SB's thinking is essentially the same as the above declarer. But rather than pass and take his average minus, he threw caution to the wind and used the UI to try for a top, although a bottom was more likely.

The logic is flawed -the thought should be "Unless most people in the room who uses the same bidding sequences as us and is like us in assessing the hands, would bid 6♠" - you can ignore people who would never bid beyond 4♠ or who would never use queue bids as they are not to be included. (Which is why you can't just look at the score sheet to decide what is a LA).

[mycroft]

Yes, barmar. Exactly - and I do understand your point. I stand corrected; what I said is not exactly correct, and the difference is material.

It's the fact that a) the UI tells me both that 3NT is going to make and that 3NT is going to be (almost universally) bid, and that b) the law tells me that if I bid 3NT, and it doesn't go down, the TD will roll the score back to 2NT making the same number of tricks; that c) 6NT (being almost never the right call) becomes suggested over pass.

I would argue that if the UI and the auction tell you that 2NT is likely to be A-, that 6NT, going for top (almost never) or bottom (almost always) may not be logical. But when we're trading an expected 2/24 top for either 24 or zero, we can afford to take a 5 or less% chance.

[lamford]

mycroft, on 2017-April-06, 16:40, said:

Yes, barmar. Exactly - and I do understand your point. I stand corrected; what I said is not exactly correct, and the difference is material.

It's the fact that a) the UI tells me both that 3NT is going to make and that 3NT is going to be (almost universally) bid, and that b) the law tells me that if I bid 3NT, and it doesn't go down, the TD will roll the score back to 2NT making the same number of tricks; that c) 6NT (being almost never the right call) becomes suggested over pass.

I would argue that if the UI and the auction tell you that 2NT is likely to be A-, that 6NT, going for top (almost never) or bottom (almost always) may not be logical. But when we're trading an expected 2/24 top for either 24 or zero, we can afford to take a 5 or less% chance.

The contract was far less than 5%. It was an inferential problem I composed (with the exact layout being necessary) and the chance of success of the contract is exactly COMBIN(26,13):1 which is 10,400,600:1. So it can hardly be demonstrably suggested.

[barmar]

lamford, on 2017-April-07, 08:22, said:

The contract was far less than 5%. It was an inferential problem I composed (with the exact layout being necessary) and the chance of success of the contract is exactly COMBIN(26,13):1 which is 10,400,600:1. So it can hardly be demonstrably suggested.

Exactly, it was a total flier.

He probably could have been safe by bidding 4NT or 5NT. These also aren't likely to make if RR just has a minimum GF, so not suggested by the UI either, but they're not million-to-one shots, either.

[barmar]

mycroft, on 2017-April-06, 16:40, said:

I would argue that if the UI and the auction tell you that 2NT is likely to be A-, that 6NT, going for top (almost never) or bottom (almost always) may not be logical. But when we're trading an expected 2/24 top for either 24 or zero, we can afford to take a 5 or less% chance.

If it's A-, that's something like 30-40%. So for the gamble to be reasonable it should be far better than 5%.

[mycroft]

Which is what I said, no?

That if 2NT is likely to be A- on the field (not *everybody* will be in game; i.e. the auction we have is reasonable, and it's only the tank-invite that tells me that game is actually on), then it needs to be a "reasonable bid" to make any sense.

However, if "everybody" is going to be in game, but because of CHO and his "why don't you just blast it rather than tank-and-pass/tank-and-invite and pooch me?" games, the TD is going to roll *my* raise back - then even a 2% chance at a better score - especially if it's a clear top - is worth it.

This auction - yeah, okay. But I certainly know of the "Hesitation Blackwood" cases where "he doesn't know whether to bid 6, so he tanks-and-bails. *I* know we should be in 6, but 6 will be rolled back. But I can make 7 on the wrong lead..." "But TD, it's totally illogical to bid 7 off a cashing Ace!"

[barmar]

mycroft, on 2017-April-07, 10:10, said:

Which is what I said, no?

Yes, I didn't notice an "if" in your post, I thought you said that the UI told you that 2NT would be A-.

Now, I'm not so sure that the UI actually tells you that everyone will be in 3NT and also that it's a likely make. The UI suggests that many players will just bid game instead of inviting, but does it really suggest that RR is the only one who would go low? To me it sounds like he has a maximum invitation, not a minimum GF, and he's apologizing for not upgrading because of SB's declarer play (not to mention his ability to find some legal trick in case something goes wrong). And would everyone accept the invitation with South's hand? It's a 16 count, but the spot cards are horrible and no long suit.

[Vampyr]

weejonnie, on 2017-April-06, 16:00, said:

The logic is flawed -the thought should be "Unless most people in the room who uses the same bidding sequences as us and is like us in assessing the hands, would bid 6♠" - you can ignore people who would never bid beyond 4♠ or who would never use queue bids as they are not to be included. (Which is why you can't just look at the score sheet to decide what is a LA).

Do you bid queue bids up the line?

[mycroft]

Barmar, I think I'm arguing the point of the OP, not the actual hand. I may be confusing people with that, and if it's you, I apologize.

The reason for the rule is that there *are* cases, two or three a year, where the Expert knows that the "only legal Logical Alternative" is going to get an almost bottom, and the "illogical Alternative" has non-zero probability of coming in. And they *did* bid the !LA, and when it made, and it had to be allowed because it wasn't a Logical Alternative, the roof fell in (reasonably so, in my opinion).

I could see an argument where if they're trying to trade 30% for top (almost never) or bottom, it's not logical - but then what about the Barometer (or the player who is very very good at score estimating) where "I need a 70% round to win, I don't care where I end up if it's not 1st" (say I have my 10K, just need an open NABC+ win)?

Having said that, the cases like the OP (including the auction, not just the magic make hand) where it is likely just trading something for nothing, but magic happens and it makes, are so far between (even compared to the ones that generated the regulation) that if it gets caught in the backwash, I'm not going to get too upset. It's a L73 violation anyway :-).