Join Bridge Winners
All comments by Franco Baseggio
1 2 3 4
You are ignoring the author of this comment. Click to temporarily show the comment.
Pitch a club. A, Q. Maybe K will be stiff. Maybe RHO will win and be forced to help. If LHO wins, or RHO wins and has a 3rd spade to play (or LHO plays a club through dummy), I'm reduced to playing for Kx of hearts onside.

If the K does fall, play Q, J and take the heart finesse. I don't think the minor heart blockage poses any real difficulty. Do play a low heart next if LHO has all 4, and hope they have K too.

I feel like this can be improved.
Jan. 16
You are ignoring the author of this comment. Click to temporarily show the comment.
You are ignoring the author of this comment. Click to temporarily show the comment.
Thanks. Fixed “4 clubs”, though weirdly there was a display issue: the source already had a space.

The 2nd diagram I need to hack the auction to have South on lead. I wish there were better ways (known to me) to control this for end positions. I chose 5NT because south needed 5 tricks with no trumps.
Jan. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
Can you give another example where an extended menace eliminates the need for a positional menace? I know of exactly two ways to do that, both very unusual (this article covering one of them). I think you’ll find any other (primary) compound still has a positional menace (by which I mean ‘stopped by equal length under the menace, with any defensive assets over the menace insufficient without equal length under’), even if you’re not required to make a positional discard (that’s what the recessing/extension accomplishes).

If not, I’m fascinated by the exceptions and would love to see more.

I agree precise names of obscure squeezes tend to be more obfuscating than helpful. I can come up with a clash squeeze position that has 4 or 5 different names yet with identical N/S holdings and play. I still think the key menace is worth naming (eg “guard squeeze”), and here the key menace is actually 2: tandem (conjoined?) ruffing guards. Odd transportation solutions usually make the name, too, like “criss cross” or “entry shifting”.

I totally get it if my sensibilities differ from others’ here. I’m used to that :)
Jan. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
I find most complex squeezes can usefully be looked at as compound squeezes with some flaw, addressed with compensation. I find this least helpful with ruffing guard squeezes, though. As Len says, a key feature of compounds is that there is a basic / positional menace.

Also, East isn’t squeezed. Though on another layout they could be.
Jan. 11
You are ignoring the author of this comment. Click to temporarily show the comment.
A good aspiration for composed double dummy problems is “no red herrings”.
Jan. 7
You are ignoring the author of this comment. Click to temporarily show the comment.
Totally right. I never noticed that.
Jan. 4
You are ignoring the author of this comment. Click to temporarily show the comment.
What if I play hq at trick 3? Say declarer crosses in clubs to play another spade, now I can play another heart and it transposes. So, declarer plays a diamond (which?) instead, ducked, and plays a spade. No time to work it out now, alas…
Jan. 1
You are ignoring the author of this comment. Click to temporarily show the comment.
I will think on that. I don’t immediately see any flaw.
Jan. 1
You are ignoring the author of this comment. Click to temporarily show the comment.
If South pitches another heart, East holds SQ and plays a heart, establishing a heart winner with diamonds still controlled. West can spare a long spade on this trick.
Dec. 31, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
Changed my mind yet again.

Club lead as before: CA, SA, CK, S9 (heart pitch), hq-HA. When south pitches last club under sq, west overtakes to lead a diamond through. East covers dummy’s card. If it’s a spot, it’s trivial to kill dummy and get an eventual heart. If it’s dq-DA then south is now one suit squeezed: if an honor is unblocked, East plays a heart and eventually scores d9.

Fantastic deal. Even more fantastic if I missed a counter.
Dec. 31, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
Run diamonds in case there’s a surprise. Top hearts and a 3rd (unless someone shows out), spade hook. JX of spades seems less likely than split heart honors once we draw 4 low hearts.
Dec. 31, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
I think the most interesting defense might be a club lead.

CA, SA, CK, S9 (pitching heart), hq-HA.

Now HJ, SQ (pitching last club). East can’t lead into HK7, so plays a diamond. Declarer can win in dummy, and now cash CQ pitching a heart.

I continue to believe this is a make. If not, I’d love to learn only that and consider the winning defense.

EDIT: see next comment
Dec. 31, 2019
Franco Baseggio edited this comment Dec. 31, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
GC: nope :). That might work. I kinda think it does. Need to ponder more….
Dec. 30, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
SA, SQ, SK, spade to dummy. You have 1 spade, 2 hearts, 2 diamonds, and 3 clubs.
Dec. 30, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
EDIT: I changed my mind several times, see my last comment

After h2-6-10-K and playing diamonds, East ducks 2 diamonds, wins the third, and plays a heart to dummy’s Jack.
Dec. 30, 2019
Franco Baseggio edited this comment Dec. 31, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
EDIT: I changed my mind several times, see my last comment

I think this is right. Combine with ducking 2 diamonds and a later low heart. If we can limit declarer to 4 red tricks then we can allow a spade to score and wait for a club.
Dec. 30, 2019
Franco Baseggio edited this comment Dec. 31, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
Steve: Duh, if only I reread my own post that would be obvious. “Remote” is dead wrong (except maybe for the auction, where a passed hand probably holding a spade honor or two probably doesn't have KQ in both reds).

David: The links in the comment above yours (including some of the comments and related posts) may be what you're looking for. However, they also delve into another layer, which may make it a slog.

In general, both sides pick probabilities that present the other side with a pure guess, one where they're indifferent between their plays. That might not be possible if the prior odds are too remote, which is what I concluded here: West can safely always blank comfortable that declarer nevertheless should always pay off to it. If West were more likely to hold the honor a priori, then at some point they need to adopt a mixed strategy.

If you think you have a good example post to send me, I can try to do what you ask with it.
Dec. 17, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
“ It's an interesting game theory problem beyond the scope of this comment.”

… but not beyond the scope of _this_ comment :)

Actually, if KQ with West vs split honors is 1:2, the Nash equilibrium is for West to always bare them and for Declarer to always fall for it, so the “risk a trick” issue is just a psychology/imagination challenge.

GIB will never make that play, but I don't actually intend this to be a “what would you do against robots” problem even if the setup involves them.

(In truth this is still simplifying; you have to think about what the field will do. And at BAM it can get really complicated: South and West on the same team need to coordinate to reach the true optimum strategy, possibly the most esoteric bridge thing I've encountered. If this sounds interesting, see: http://jlwbridge.blogspot.com/2009/07/spillover-from-daily-trout.html and http://fmbbridge.blogspot.com/2009/07/game-theory-addendum.html)
Dec. 17, 2019
You are ignoring the author of this comment. Click to temporarily show the comment.
EDIT: I largely lost my mind here in defending “remote” from the previous comment.

Both squeezes have card-reading charges, both of which affect the decision. “Remote” may overstate the charges on the one hand (see below), and “boils down to … more or less likely than split diamond honors” definitely understates it on the other. So the bottom line is that a 50% chance of a spade shift is the lower bound on what is necessary to justify adopting my recommended line, and the right bar is certainly at least somewhat higher.

Once we have a bar, I don't consider myself well-calibrated on how likely a shift is; perhaps that's a high enough bar that ducking trick one is clearly indicated. I'm 100% certain that I shouldn't have neglected this issue.

Regarding playing for 11 after a spade shift, let's play it out: duck J, win A, run 7 clubs, keeping A and AJT unless you see a heart honor (in which case retain T and hook with it). Now play to A. If East doesn't play an honor, you settle for 10 tricks without risk. If East does drop an honor, now you have to guess whether it's from the hoped-for KQ, vs stiff or from Hx. Against reasonable opponents, both seem possible enough (particularly after a spade switch from partner), and from a priori likely enough holdings (at least 2x the both-honors holding), that it's hard to play for your a priori 25% chance of an 11th trick. A 1/3 chance of stiffing the honor (which really isn't that hard) or keeping a spot and then dropping the honor (harder, but not _that_ hard if you're alert) is enough that you can never make the play for an 11th trick. Of course, if it's < 1/3 you will play for it and gain a trick on the 25% where it's available, but lose a trick some fraction of the time.

Against all this, East may be confused about the location of the T and have more trouble (making this less “remote”)

Anyway, that's an unpacking of my intuition behind “remote”. The exercise makes me back off a little, but I don't think I'd risk that play under most circumstances.

Without doing the analysis for the other line, I'll just note that the scenario where you lose a trick only develops on a priori less likely layouts and also requires West to risk blowing a trick (and a subtle defense by East to pitch a winner and keep a loser). This is still not a trivial charge (and so I cop to exaggeration), but I think it's notably smaller. It's an interesting game theory problem beyond the scope of this comment.
Dec. 17, 2019
Franco Baseggio edited this comment Dec. 17, 2019
1 2 3 4
.

Bottom Home Top