[mikeh]

I’ve been having issues with posting edits. In particular I can’t seem to embed hand diagrams, which makes posting a bridge movie difficult. However, let me give it a try.

This past weekend saw the playoffs for the Canadian Senior Team Championship, vying for the right to represent Canada in Buenos Aires in the fall.

We played the semi final on Saturday, winning comfortably. One hand that contributed greatly arose in the third of four 14 board segment.

All vulnerable, you pick up as dealer x KQxx AK AKQxxx.

The first issue is what to open. We definitely have enough playing strength for a forcing auction, and a control-rich 21 count to suggest a 2C opening. But experience suggests that opening 2C with a long minor and a four card major can cause issues that n terms of finding the major fit,

Add to that the fact that we’re missing 12 spades and 19 hcp, suggesting that there is almost no risk of 1C being passed out, and 1C seems a better choice. Our system also helps, but I think 1C is correct in almost any standard type method. We play a method in which we transfer in response to 1C, and after 1C 1R, 1N by opener shows 17-19, eliminating the need to jump to 2N. This, in turn, allows us to respond very lightly to 1C.

Ok, we decide to open 1C. The opps remain silent throughout the action….which sees partner responds 1H, showing 4+ spades.

We have an easy reverse…even though it looks odd to people unfamiliar with transfer responses to 1C. We bid 2H, showing 4+ hearts and longer clubs and a good hand. It’s forcing and promises a rebid.

Partner surprises us by bidding 3C. This shows 3+ clubs, denies 5+ spades or 4 hearts and, most importantly, shows values…it establishes a game force, which the reverse did not.

We are surely in slam territory and really don’t need anything more than how many aces he has.we play that 4D is a kickback ask (know ckback uses a bid of one over the trump suit, at the 4 level, as keycard). Partner continues to pleasantly surprised us by bidding 4N, showing both missing aces!

Ok….what next? Axxx Ax xxx xxxx makes 7C laydown. But the same hand strengthened by any of the spade King, the heart jack or the diamond queen gives us 13 top tricks in notrump. We can’t find out about the red honours, but it’s say to ask for kings….we play specific kings but even a simple ‘how many do you have’ would work, since we’re only missing one.

The sensible course, then, is 5D, king ask.

I had a senior’s moment and just bid 7C. Oh well.

As it happens, dummy hit with AQxx Ax Jxxx xxx, so my error didn’t cost anything.

The opps led the usual trump lead (it’s expert standard to lead trump against suit grands, because good opps won’t be missing an important trump card so one isn’t giving anything away…of course one has to listen to the auction and one rarely, if ever, would lead a stiff trump lest partner have Jxxx.

We pass the first hurdle when RHO follows…a 4-0 split would likely have gotten us doubled by it’s still good to see both follow.

If trump are 2-2, we just pull trump and claim, ruffing our heart loser in dummy. But RHO shows out on the second club.

What now?

The simple line is to ruff the third club in dummy. It would be nice if dummy had a high club, but it doesn’t. So if the def3nder with the outstanding trump has two or fewer hearts, down we go.

Is there an alternative?

We have 12 tricks on top and need one more.

We could finesse the spade…basically a tiny bit less than 50%…because LHO has 20 vacant spaces outside of clubs, while RHO has 12….making it 12-10 that RHO has the king.

We could try to ruff it out, but we don’t have any dummy entries so that won’t work unless it’s Kx, vanishingly unlikely.

But….only one defender can hold as many as four hearts and we can exert pressure n him if he holds either the spade king or the diamond queen. We cash all our minor winners, reducing to a five card ending. I apologize for the lack of a diagram, but here’s the situation at trick 9:

AQ Ax J void n dummy with x KQxx void void in hand.

If the hand with long hearts has the spade king, it’s either surrenders the heart stop or stiffed the spade king…n a five card ending, one can’t hold 4 heaets and 2 spades.

We then cash the spade ace. The king, as it happens, won’t appear but we aren’t that worried….that parlay didn’t work but if the long hearts also have the diamond queen, he’s already thrown all his spades before we cash that ace…and that ace crushes him….he can’t keep 4 heaets as well as the diamond queen.

Thus, the squeeze only needs the hand with 4+ hearts to hold one of the two relevant honours, and the odds of that seem (to me) to be greater than the chances of a 4=3 heart break.

Does that mean the squeeze is better?

No….what this comparison ignores is that we don’t care about where those honours are if hearts break 4=3…we make by the ruff, while the squeeze fails when the long hearts don’t hold either relevant honour.

So we’re only worried about 5=2 or worse heart breaks and the wilder the heart break, the less likely it is that either honour is in that hand.

I didn’t even try to do the math at the table…I’m not sure I could, without pen, paper and, probably, a calculator!

My sense was that the ruff as probably the best play…..since hearts broke and one of the honours was in long heart holding, both lines work. Maybe some reader can work out the odds.

We scored 2140.

Now, it wasn’t possible to know how good this grand was before seeing dummy.

Many hands make it cold. 4+ trump…claim. Jxx…claim if trump are 2=2 or unless hearts are 6=1 or 7=0. Partner has the heart jack or diamond queen or (given my senior’s moment) the spade king…claim. And so on. But 4=0 trump…you’re down, likely doubled. And f partner has Axx n hearts, you may end up needing a squeeze if hearts aren’t 3=3 (with a very good chance of winning).

Why is that important in the auction? Most experts agree that one needs to see the grand as at least a 75% favourite to justify bidding it. The straight imp odds….how many you win for the grand slam bonus, against opps in the small slam, compared to how many you lose when it fails suggests that the grand has to be a heavy favourite but not by a 3-1 ratio. So why the high standard?

Because experience suggests that even expert pairs can screw up by missing even the small slam. Ndeed that happened on this board. Our opponents played 5C….I don’t know their auction and wouldn’t give it if I did….I’m sure they’re embarrassed by it and it doesn’t reflect how they played during the match.

My point is that we’d have won 13 imps by stopping in 6 and lost 12 by failing in grand….we won 17 on the board…so in essence gained only 4 imps by bidding 7, risking 12 had we failed. My point of view, given the action, remains that bidding the grand was correct, given that dummy was actually close to as bad a hand as was possible on the auction.

[DavidKok]

Very nice bridge movie! I would have played for the squeeze, and I have a few more thoughts below.

mikeh, on 2024-June-17, 16:16, said:

All vulnerable, you pick up as dealer x KQxx AK AKQxxx.

[..]

As it happens, dummy hit with AQxx Ax Jxxx xxx, so my error didn’t cost anything.

[...]

Does that mean the squeeze is better?

No….what this comparison ignores is that we don’t care about where those honours are if hearts break 4=3…we make by the ruff, while the squeeze fails when the long hearts don’t hold either relevant honour.

Does this imply that we had some good heart spots as well? If the x-es are low ones the actual hearts held by the opponents don't matter, do they?

I've learned most of what I know of squeezes from Love, and would classify this as a "single played as double, type B2-1 squeeze" with the 'both' threat being hearts and isolated with two entries and one entry opposite, and the spade and diamond threats being the right and left threats. The treatment for this squeeze is literally "do whatever you please as long as you do not break down the entries in hearts", and I think the odds of the 4(+)-card heart suit being with either the ♠K or the ♦Q, or of course the ♦Q dropping (which has a slight though practically irrelevant risk to test early if we go for the ruffing line, since now we lose to diamonds 6-1) is probably slightly better than hearts being 4-3 (close to 60%, the a priori odds are around 62% but the vacant spaces information shifts it a bit).

I like the methods up to 3♣, and definitely agree with opening 1♣. On balance I like the grand - with clubs 2-2 it is cold, and if they split 3-1 we still have very good odds with either the ruff or the squeeze. If the clubs had split 4-0 that would be unfortunate (with or without a double), but I do not think this shifts the odds by enough to make it a poor grand.
In my experience people at all levels are worse at slam bidding than most people think, so the odds of the opponents being in game definitely play a role in these estimates. Bringing back -100 for going one off in a grand and hearing the teammates call out their own -680 or -690 when scoring up is one of the infamous disasters to avoid at teams, but I would still go for the grand here.

[mikeh]

DavidKok, on 2024-June-18, 04:03, said:

Very nice bridge movie! I would have played for the squeeze, and I have a few more thoughts below.

Does this imply that we had some good heart spots as well? If the x-es are low ones the actual hearts held by the opponents don't matter, do they?

I've learned most of what I know of squeezes from Love, and would classify this as a "single played as double, type B2-1 squeeze" with the 'both' threat being hearts and isolated with two entries and one entry opposite, and the spade and diamond threats being the right and left threats. The treatment for this squeeze is literally "do whatever you please as long as you do not break down the entries in hearts", and I think the odds of the 4(+)-card heart suit being with either the ♠K or the ♦Q, or of course the ♦Q dropping (which has a slight though practically irrelevant risk to test early if we go for the ruffing line, since now we lose to diamonds 6-1) is probably slightly better than hearts being 4-3 (close to 60%, the a priori odds are around 62% but the vacant spaces information shifts it a bit).

I like the methods up to 3♣, and definitely agree with opening 1♣. On balance I like the grand - with clubs 2-2 it is cold, and if they split 3-1 we still have very good odds with either the ruff or the squeeze. If the clubs had split 4-0 that would be unfortunate (with or without a double), but I do not think this shifts the odds by enough to make it a poor grand.
In my experience people at all levels are worse at slam bidding than most people think, so the odds of the opponents being in game definitely play a role in these estimates. Bringing back -100 for going one off in a grand and hearing the teammates call out their own -680 or -690 when scoring up is one of the infamous disasters to avoid at teams, but I would still go for the grand here.

I may have misunderstood you, but I think your assessment of the squeeze chances aren’t quite accurate. If the issue were….what are the chances that the long heart suit is accompanied by either or both of the relevant honours? Then I’d agree with you. But that’s not quite the assessment. Most of the time the long hearts have a relevant honour, the ruff line works…because 4-3 is roughly 62%. So the issue is what are the odds that the long hearts hand has one of the honours when the suit is 5-2 or worse…less than 38% of the time?

Ignoring the heart suit, the chances that one opp has both honours is 25%, but give one hand 5 hearts and the other two? If the long trump hand has 5 hearts, the ruff line still works, but now that hand has only 5 spaces for the honours and the short heart hand has 10….so the squeeze odds shift dramatically (lower), but I haven’t even tried to calculate them. Of course the odds are somewhat in favour of the long hearts being with the short clubs….now the long trump hand has 8 vacant spaces and the long heart hand 7, meaning that the long heart hand has close to a 50% chance of having an honour. 5=2 is about 30% of the hands…6=1 and 7=0 account for less than 8% of hands.


Btw, the hearts were KQ54 opposite Ax, so you are correct in that the spots don’t matter….give me KQ10x and the calculations change since now we also make when either defender has Jx….again, Jxx is irrelevant since the ruff line works then as well. I’m writing this at 3:49 am due to insomnia so please forgive any error��

[DavidKok]

I don't think this is the right way to analyse two competing lines. We have two options, both have some success percentage, and we should pick the better one. We may choose to simplify the analysis by eliminating the cases where both lines work or neither line works, and compare what remains. But in that case we should be comparing "the squeeze works but the ruff fails" with "the ruff works but the squeeze fails", and I think you are comparing "the squeeze works but the ruff fails" with "the ruff works".
I went for the first approach, and attempted to estimate the odds of each of the lines working without referencing the other.

[mw64ahw]

I'm not quite sure why
1♣-1♥
2♥-3♣
establishes a GF from responders perspective when they could be as weak as KQxx, xx, xxxx, xxx and showing preference. Opener may be justified in not stopping before game with this hand, but Kickback does not look like it is based on knowing responder's strength.

What does responder do with KQxx xxx xxxx xx?

Would
1♣-1♥
3♥ be more reflective of openers strength and shape x4x6?

A 2♣ opening doesn't look too bad in this instance

[DavidKok]

I can't speak for mikeh but I play Lebensohl after reverses, so that on this auction 2NT is weaker and typically a signoff in 3♣ facing a minimum reverse, while stronger hands bid to the 3-level immediately.
Jumping with strong hands is typically not beneficial, and I would not wish to use the jump to 3♥ for this hand type. Such a jump makes it very difficult to explore whether we belong in spades, hearts, notrump or clubs and slam, game or even partscore. Keep in mind that in a T-Walsh style responder can be quite weak for their initial 1♥ response.
A weak hand with at most 3 hearts and 2 clubs indeed is a problem, especially without a fifth spade. Even worse would be something like 4=2=5=2 or 4=2=6=1 minimum. I think most bidding systems will run into this problem though - misfit hands on reverse auctions tend to struggle. In Dutch Doubleton this is less of a concern, but in standard and T-Walsh you may have to give false preference to a 5-2 minor suit fit.

[mw64ahw]

DavidKok, on 2024-June-18, 06:54, said:

I can't speak for mikeh but I play Lebensohl after reverses, so that on this auction 2NT is weaker and typically a signoff in 3♣ facing a minimum reverse, while stronger hands bid to the 3-level immediately.
Jumping with strong hands is typically not beneficial, and I would not wish to use the jump to 3♥ for this hand type. Such a jump makes it very difficult to explore whether we belong in spades, hearts, notrump or clubs and slam, game or even partscore. Keep in mind that in a T-Walsh style responder can be quite weak for their initial 1♥ response.
A weak hand with at most 3 hearts and 2 clubs indeed is a problem, especially without a fifth spade. Even worse would be something like 4=2=5=2 or 4=2=6=1 minimum. I think most bidding systems will run into this problem though - misfit hands on reverse auctions tend to struggle. In Dutch Doubleton this is less of a concern, but in standard and T-Walsh you may have to give false preference to a 5-2 minor suit fit.

Thanks
I take the point on Lebhensohl, which means you don't have the option of 2NT with your 4252 hands.

On 1♣-1♥-3♥ I don't think you have much issue with 9 tricks in hand even if responder has 5♠ and nothing and especially when the ops. aren't bidding

[mikeh]

DavidKok, on 2024-June-18, 05:03, said:

I don't think this is the right way to analyse two competing lines. We have two options, both have some success percentage, and we should pick the better one. We may choose to simplify the analysis by eliminating the cases where both lines work or neither line works, and compare what remains. But in that case we should be comparing "the squeeze works but the ruff fails" with "the ruff works but the squeeze fails", and I think you are comparing "the squeeze works but the ruff fails" with "the ruff works".
I went for the first approach, and attempted to estimate the odds of each of the lines working without referencing the other.

Test

[mikeh]

DavidKok, on 2024-June-18, 05:03, said:

I don't think this is the right way to analyse two competing lines. We have two options, both have some success percentage, and we should pick the better one. We may choose to simplify the analysis by eliminating the cases where both lines work or neither line works, and compare what remains. But in that case we should be comparing "the squeeze works but the ruff fails" with "the ruff works but the squeeze fails", and I think you are comparing "the squeeze works but the ruff fails" with "the ruff works".
I went for the first approach, and attempted to estimate the odds of each of the lines working without referencing the other.

Test

[gszes]

Good post with a lot of bridge thinking. I should probably quit now but I feel compelled to add something about the play of the hand. I like the idea of going for the squeeze mainly because I feel it is safer (and psychologically more satisfying) than going for the heart ruff. My thoughts (after playing the 2nd round of trumps) are why do we not increase our odds a bit by trying to ruff out Qxx in diamonds? It seems perfectly safe to draw the 3rd round of trumps.
trick 4 dia A
trick 5 dia k
TRICK 6 heart to the A
trick 7 ruff a dia If the Q has not fallen yet we are no worse off than we were by going for the squeeze immediately and we might gain valuable distributional information that might allow us to eliminate or favor our choices later on in the play. Who knows? maybe we will decide the spade finesse is the way to go.
trick 8 play a trump and pitch a spade
trick 9 play a trump and pitch a spade
Leaving us with
S x
H AKx
opposite
S AQ
H x
D J

If we have no (useful) distributional information available we can play a spade to the A which will either drop the now stiff Spade K OR successfully set up the dia Heart Squeeze if it was available.

[akwoo]

mw64ahw, on 2024-June-18, 06:44, said:

I'm not quite sure why
1♣-1♥
2♥-3♣
establishes a GF from responders perspective when they could be as weak as KQxx, xx, xxxx, xxx and showing preference. Opener may be justified in not stopping before game with this hand, but Kickback does not look like it is based on knowing responder's strength.

What does responder do with KQxx xxx xxxx xx?

Would
1♣-1♥
3♥ be more reflective of openers strength and shape x4x6?

A 2♣ opening doesn't look too bad in this instance

In addition to the earlier point about Lebensohl, so a weak hand that has preference for clubs bids 2N and then passes partner's expected 3♣ bid... I doubt that the 1♣-1♥-3♥ sequence is natural for them - it's probably some sort of splinter in support of spades (but there are various ways to assign meanings to 2N, 3♦ and 3♥, none of which need to be natural, so it could be something else).

[mw64ahw]

akwoo, on 2024-June-19, 14:41, said:

In addition to the earlier point about Lebensohl, so a weak hand that has preference for clubs bids 2N and then passes partner's expected 3♣ bid... I doubt that the 1♣-1♥-3♥ sequence is natural for them - it's probably some sort of splinter in support of spades (but there are various ways to assign meanings to 2N, 3♦ and 3♥, none of which need to be natural, so it could be something else).

Yep
It slightly academic for me as my bidding goes.
1♣-1♠ GI
3♥ x4x6 SI which makes life easy

[mikeh]

akwoo, on 2024-June-19, 14:41, said:

In addition to the earlier point about Lebensohl, so a weak hand that has preference for clubs bids 2N and then passes partner's expected 3♣ bid... I doubt that the 1♣-1♥-3♥ sequence is natural for them - it's probably some sort of splinter in support of spades (but there are various ways to assign meanings to 2N, 3♦ and 3♥, none of which need to be natural, so it could be something else).

We play 1C 1H 2N as a 17+ raise of spades, and it can be unbalanced but won’t be most 4=6 black hands since we can bid 4C with a ‘good 4=6’.

3H, instead of 2N, is around 15-16 with heart shortness and 4 spades.

When one plays 1C 1H 1N as 17-19….in standard it’s a 2N rebid…one doesn’t just gain bidding space nor just the ability to get out at 1N or 2 of a suit, both when responder is very weak…but also the ability to use the 2N for more esoteric purposes.

After 1C 1H 2N, responder can transfer at either the 3 level (to play or make a slam try) or tye 4 level (to play) or can relay via 3C, asking for a description of openers hand…with 4 responses. Max unbalanced (with a relay available thereafter) min unbalanced (relay if needed) min balanced (natural thereafter) and precisely 4333 max, allows for 3N.

A similar scheme applies to 1C 1D 2N, where 1D shows hearts and 2N is a strong raise.

[akwoo]

Thanks for the info, mikeh.

Also, congratulations on your victory.