Team trials bidding problem

My partner held this hand in the team trials, favorable 4th seat:

♠	Q J 6
♥	Q 8
♦	J 8 7 4 2
♣	A 10 4

South	West	North	East
			1♠
2♣	4♠	Pass	Pass
Dbl	Pass	?

What's your call?


Read more!

More on squeezes

Yesterday I talked a lot about a couple of triple squeezes, winding up with some thoughts on working out squeezes through a process of compensating for flaws.

One of the most powerful compensating techniques is a guard menace:

		♠	K
		♥	K
		♦	2
		♣	K 2
♠	A			♠	Q
♥	A			♥	Q
♦	7			♦
♣	Q J			♣	10 8 3
		♠
		♥
		♦	A K
		♣	A 9 6

You've got two isolated menaces over West (in the majors) and 1 loser, but no entries (in Love's terminology, you have BLU but no E). But, the guard menace in clubs means West is squeezed sooner -- cash 2 diamonds and if West pitches a club, cash CK and hook the club on the way back for all the tricks. (This would also work if West had the CT, too, a 1-trick triple squeeze, another form of compensation for a simple squeeze lacking the normal entries.)

In fact, though, this is so powerful that you could compensate for missing Love's B and U, too: switch the major suit Ks and Qs, giving East protection of those suits, too. Now your major suit menaces are entry-less, not-isolated, and under one opponent, yet the guard menace still compensates. Cash 2 diamonds and as before pitching a club fails, so West must pitch a major. Pitch the other major from dummy and East is squeezed between clubs and the major West pitched.

Note that one key to this hand is that West not be able to guard the long club. I don't think it's possible to have a non-strip squeeze without at least one menace isolated against one opponent and over that opponent.


Read more!

Progressive squeezes made "easy"

You may have noticed I like squeezes. Recently there have been 2 unusual progressive squeezes, here and [add link].

The end positions are repeated here with slight modifications (in both cases, East's hand is immaterial except where noted):

		♠	A 10 8
		♥	2
		♦	7 4
		♣	2
♠	K J 9
♥	J 9 8
♦	10 9
♣
		♠	Q 2
		♥	K 10 7
		♦	A
		♣	K

In this position, West has already come under pressure and still needs to discard. Declarer has 4 top winners, but can emerge with 6 of the last 7 tricks.

If you switched the SK and SQ, this could be played as an ordinary repeating squeeze: 3 threats, 2 of them over the defender, and enough entries that whatever is unguarded can be cashed to effect another squeeze (ignoring, for the moment, that there are immediate 2-trick threats in both majors). Just cash DA (optionally can be delayed) and CK: if hearts are unguarded, HK, HT effects a diamond-spade squeeze. If spades are unguarded, SK, SA, ST effects a red suit squeeze. If diamonds are unguarded, SK, SA, D7 effects a major suit squeeze.

This might also work if you could duck a heart and a diamond came back -- now the count would be rectified and all the necessary entries are in place. But, a savvy West knows that the key to progressive squeeze defense is to give up the trick that doesn't allow a subsequent squeeze. In this case, if in with H8, West can exit SK, giving up a spade trick, but preventing the red suit squeeze.

This is similar to the reason why a normal progressive squeeze needs 2 menaces over the defender. If there's only one, they can normally unguard that menace and then there's no 2nd squeeze since both remaining threats are under the defender.

If that's the key to progressive squeeze defense, then the key to setting up more complex progressive squeezes is to find enough compensation to prevent defender from breaking up the 2nd squeeze. One standard technique is a 2 trick menace, such as the heart suit above. This is "commonly" used when there are 2 menaces under the defender, a 2-trick squeeze will still succeed if the sole "over" menace is a 2-trick threat. The defense may prevent a 2nd squeeze by unguarding that menace, but it's small consolation when that immediately costs 2 tricks.

In the position above there's all sorts of compensation that's necessary:

a) the heart tenace: if West unguards diamonds, DA, SA, D7 effects a major suit strip squeeze. This would not work if the HT and H9 were switched.

b) the heart establishment threat: if West comes down to 2 hearts, we can duck a heart to set up 1 extra trick, then win the (forced, see below) red suit return for a diamond-spade squeeze. Without the H7, the squeeze fails.

c) the split tenace in spades: this is necessary to prevent a spade return in (b), this breaks up the 2nd squeeze but at the expense of conceding an extra spade trick.

d) the 2-trick threat in spades: this is not strictly necessary, but it does mean that if West errs by blanking his SK he won't take any tricks (DA, 3 spades for a red suit squeeze). The 2-trick threat in hearts is similar.

More recently I posted this club grand slam:

		♠	Q
		♥	K
		♦	A Q 7 4 2
		♣
♠	A
♥	Q J
♦	K 10 8 6			♦	J 9 5
♣
		♠
		♥	10 9 8 7
		♦	3
		♣	5 2

(Note East has 3 diamonds here.)

Declarer has 5 tricks (including the diamond finesse), but by cashing a club can again emerge with 2 extra tricks.

Here, declarer appears to have a useful entry in the diamond suit, but needs to ruff a diamond to isolate the menace against West, so that's somewhat of an illusion. There's an entry in hearts, but it's blocked, and there's no entry in spades. But, the ruffing threat in diamonds and the blocked heart position looks like a trump squeeze. In fact, if West pitches the SA on the next club, then DQ, SQ does subject him to a red suit ruffing squeeze (if heart, HK, ruff, hand good; if diamond, ruff diamond, HK, dummy good).

So, a spade pitch is out. What about a heart? Declarer can unblock the HK, get back to hand with a ruff, but then will lack either an entry or a menace for the spade-diamond squeeze. But the 2-trick heart menace bails him out, and is in fact necessary.

What about a diamond pitch? There's no major suit ruffing squeeze since the spade length for a ruffing threat isn't present. Even if it were (e.g. make a diamond a spade), establishing a diamond requires a ruff, and then only the HK is left for an entry to cash the established diamond, but using the HK would break the intended major suit ruffing squeeze. Luckily (for declarer) there's a 2-trick threat in diamonds, too. (Earlier in the play, declarer had to be careful to pitch a spade and keep all 5 diamonds.)

Note that if East didn't have a 3rd round diamond guard, then declarer could have won the opening heart lead in dummy (preserving HA in hand) and would again have a more typical progressive squeeze:

		♠	Q
		♥	3
		♦	A Q 7 4 2
		♣
♠	A
♥	Q J
♦	K 10 8 6			♦	J 9
♣
		♠
		♥	A 10 9 8
		♦	3
		♣	5 2

The red suit length looks nice, but isn't really necessary: cash 2 clubs and then whatever winner LHO unguards for a repeating squeeze. So that 3rd diamond with East really messed things up and required 2 2-trick menaces and a trump squeeze to compensate.

The general point I wanted to make here (not sure if I was really successful) is that I find most squeeze books misguided. Perhaps BLUE or whatever is useful when first learning a simple squeeze (B = Both guards in one hand, L = right number of Losers, U = at least one threat in the Upper hand, what I call "over", E = Entry to established winner available), but in the long run enumerating requirements without motivations is not helpful.

For example, to me, the requirements for, say, a compound squeeze are just: a triple squeeze against one opponent such that whatever they do you can effect a double squeeze -- reading some crazy list of possible entry configurations that amounts to the same thing does not do much to provide insight. If there's some defense, or lack of entry, or whatever, that wouldn't allow a double squeeze to develop, then you need some compensation that prohibits that defense. In the first deal above, you can't rectify the count, LHO might attack your entries, and so forth, but once you start bringing the HT, SQ, and long heart into play you have a counter for all these things. It would be impossible to memorize something like "3 losers, a split tenace, a tenace and establishment threat under the defender, control but not necessarily a useful entry in the 3rd threat over the defender, etc" and have that be useful. (Not that any of this is that useful, of course.) I find it much more productive to start imagining it as a "normal" squeeze, then as flaws are identified look for compensation.

That said, enumerating squeezes can be very useful as an exercise -- I think I have a very firm grasp of double squeezes (and how to defend against them) as a result of enumerating all the compound squeezes, but I haven't bothered to memorize my list of compounds.

Read more!

Inferences from declarer's play

Matchpoints. South opens 4♥ and buys it.

♠	A J 8 3
♥	8 3
♦	Q 10 9 8
♣	K Q 10
		♠	K 10 4 2
		♥	K 3
		♦	K 10 7
		♣	A 8 5 2

Partner leads a 3rd/low ♠7, possibly consistent with Q-9-7, Q-9-7-x, or 7-x. Declarer wins ♠A and plays ♣K, how do you defend?



A friend sent me this hand, this was his analysis (which I agree with):

At the table, my reaction was to play the ♣A, since declarer could easily have a singleton or even a void. I won and played ♠K, which was not a success as declarer ruffed, overtook the ♣J with the ♣Q, and pitched a diamond loser on the ♣10.

In retrospect, I think there is a clear inference that suggests ducking the club. It's certainly not impossible for declaer to play the club suit on its own this way holding a singleton. However, you are looking at Kx in hearts. It is extremely likely that if declarer had no further entries to dummy he would take a heart finesse while he could.

If declarer were 2=8=2=1, he might have ducked the first trick. With 1=8=3=1, he might fear a diamond ruff if he didn't draw trumps. He still could possibly have x/AJTxxxxx/Axx/x though and play this way.

Winning the club is still ok if you cash your two diamond tricks immediately thereafter, but that seems like a big play.

Read more!

More fun with K-10 doubleton

A while back I read this post by Jonathan (don't miss the story about Kantar in the comments). So when this hand rolled around (LM Pairs Fri eve #5), I was ready:

		♠	8 7 2
		♥	A 10
		♦	A K 5
		♣	A Q 9 7 2
♠	Q J 10 9 5
♥	Q 9 8 6
♦	7 4
♣	K 10

I led the ♠Q against 3NT. Partner won ♠A and continued the suit, declarer winning the 3rd round with the king. Hoping he had ♣J-8-x I was ready with the ♣K should he lead one. Sure enough he had that holding, and sure enough he led a club... the jack!

Later in the tournament in a team match our opponent found himself facing this club position against a high-level doubled contract:

	J x
A Q x x x x x		K 10
	x x

A club spot was led, he inserted the 10 and held the trick. Working out what happened, he found partner's void and it was automatic for another club to come back to the king for a 2nd ruff and 500.

This is only a small gain since a suit preference signal should solve the defenders' problem, but declarer could have made it harder by playing the jack from dummy at trick 1.

So if you're worried about K-10 doubleton, play the jack.




Read more!

Another bidding idea

Earlier I described a twist to unusual vs unusual that I like. I think this is a materially more useful idea:

In 2/1 auctions such as 1M-2C and 1S-2D, switch the meanings of opener's cheapest rebid and the rebid of his suit (which, in my style, is a catch-all that could be just a 5 card suit with no other suitable bid).

For example:

1S-2D

2S: shows 4 hearts
2H: catchall bid.

The point is that opener is much more likely to have a nebulous hand, so when he does you want to preserve space.

After that, we don't have or need any fancy agreements: responder's rebid of opener's suit is now available to be a catchall -- usually 2 card support, unwillingness to bid 2N or anything else. It might rarely be 3 card support on a hand that wants to emphasize keeping other strains in the picture (e.g. 3 small, or slammish if opener can summon a raise of your minor). With genuine support, opener makes the same 3 level raise they would have if the auction had gone standardly (e.g. after 1H-2C-2H, responder must raise to 3H, so just make the same bid after 1H-2C-2D(catchall)).

After 1S-2C, we have a slightly more detailed agreement:

a) 2D is the catchall, but might still include bad hands with 4D. 1S-2C-2S showing diamonds is a bit preemptive, so it has close to the same requirements as a high reverse like 1S-2D-3C.

b) Over 2D (which denies 4 hearts since otherwise just rebid 2H), we the switch 2H (shows diamonds) and 3D (shows hearts).

c) Over 2D we have an extra bid available: 1S-2C-2D-3H shows a 3 card raise with extras, and 3S shows a 4 card raise.

I think this is very simple, useful when it comes up, and that it comes up frequently enough to not pose a material memory strain.

Read more!

Failed beer squeeze, solution

From yesterday's post:

♠	A 10 8 5
♥	A 2
♦	7 4 3 2
♣	A J 3
♠	Q 2
♥	K 10 7 4
♦	A 6 5
♣	K Q 5 4

South	West	North	East
1 ♣	Dbl	Rdbl	Pass
Pass	1 ♦	1 ♠	Pass
1 NT	Pass	3 NT	All Pass

♦Q led, RHO plays the K and another diamond.
Say you duck twice and LHO shifts to ♥Q. How do you play? Clubs are 1=5.


Assuming LHO has KJ9x/QJxx/QJTx/x, you can guarantee 10 tricks by winning ♥A and running clubs to come down to:

♠	A 10 8
♥	2
♦	7 4
♣
♠	Q 2
♥	K 10 7
♦	A
♣

You have 5 in the bank and want 5 of the last 6 with only 3 tops. Luckily, the 3-loser progressive squeeze comes to the rescue.

LHO needs to pitch from KJ/J9x/Tx/- in this position. If he pitches:

a) A spade? This give up 2 spades immediately, and if you cash ♦A first you'll also have a red suit squeeze to take the rest. Ouch!

b) A heart? Play ♥10 to set up a trick there. LHO can't lead spades without giving up a 2nd trick, so must put you back in hand with a red card, now cashing all your red winners effects a spade-diamond squeeze.

c) A diamond? That gives you an extra diamond, then ♦A, ♠A, ♦7 spade exit end-plays LHO for an extra heart trick.

I guess this is my 2nd non-standard progressive squeeze in the last 2 weeks, so I'll try to write about some generalizations next week.

Read more!

A failed beer squeeze

A few years ago I tried to construct a beer squeeze: a squeeze where the opponents can successfully prevent you from gaining a trick only at the "cost" of allowing you to win the last trick with the ♦7. But, there was a cook: declarer can in fact always succeed in gaining the extra trick. Give it a try:

♠	A 10 8 5
♥	A 2
♦	7 4 3 2
♣	A J 3
♠	Q 2
♥	K 10 7 4
♦	A 6 5
♣	K Q 5 4

South	West	North	East
1 ♣	Dbl	Rdbl	Pass
Pass	1 ♦	1 ♠	Pass
1 NT	Pass	3 NT	All Pass

♦Q led, RHO plays the K and another diamond.
Say you duck twice and LHO shifts to ♥Q. How do you play? Clubs are 1=5.

A little color below, full answer tomorrow.



The auction and early defense were constructed to make this a double dummy problem -- LHO must be 4441 with all the remaining missing honors. My original intent was that you duck the ♥Q (to rectify the count), and then LHO would continue a diamond. You could cash ♥A and run 4 clubs to reach this position:

♠	A 10 8
♥
♦	7
♣
♠	Q 2
♥	K 10
♦
♣

If LHO started with KJ9x/QJxx/QJTx/x this is a non-progressive triple squeeze: unguarding hearts would allow the squeeze to repeat, and unguarding spades would give up 2 tricks instantly, but ungaurding diamonds just gives up one trick and the squeeze doesn't repeat. But it would allow declarer to, say, play ♥K, heart pitching 2 spades, then win ♠A and ♦7 and earn a beer.

But, rolling back to the ♥Q play at trick 3, declarer has a line for 10 tricks against the same layout. Stay tuned...
Read more!

Team trials bidding problem

I held this hand in the team trials, favorable 2nd seat:

♠
♥	J 5 4
♦	A K 6 5
♣	K Q J 9 5 2

South	West	North	East
			1♠
2♣	4♠	Pass	Pass
?

How would you rate Pass, Double, 4NT, and 5♣?



At the table, I doubled; I think now that it's right to bid 4NT.


Read more!

GIB declares a hand well, part II

From yesterday:

♠	Q 8 3
♥	3
♦	Q 5 2
♣	A K J 10 5 2
♠	A K 6 5 4
♥	A 9 2
♦	A 10 7
♣	6 4

Declare 6♠ after a heart lead.


GIB played ♠A, ♣A, ♣K, and ruffed a club high (when RHO followed with ♣Q), then ♠Q and running clubs left the defense with nothing to win but a top trump.

Doesn't seem that complicated or hard to find, it was just something that wouldn't have occurred to me promptly. Like I said yesterday, the technique seems more common in tricky partials than in slams.
Read more!

GIB declares a hand well

♠	Q 8 3
♥	3
♦	Q 5 2
♣	A K J 10 5 2
♠	A K 6 5 4
♥	A 9 2
♦	A 10 7
♣	6 4

The opponents are bidding hearts and you wind up in 6♠. They lead a heart, how do you play?



To me it looks like an immediate club to the 10 will work when both black suits are 3-2. GIB found a different line with the same property, though, using a technique that seems more common in a moysian partscore. Answer tomorrow...

Read more!

Novel card combination

Jeff showed me two pretty cool things at nationals. I see he's written about them, so I'll just send you there:

Novel card combination

Five or Four

Perhaps I should also check out his typesetting tools...


Read more!

Pointless squeeze (from Nat'l Swiss)

For the first time in a long time I went to basically all of nationals, playing about 17 sessions. In the final session, I mis-sorted a hand for the first time I can remember (it was really freaky seeing diamonds turn into hearts, particularly when my hand went from 5323 to 5503 after I had mis-bid and mis-defended). And there was also this example:

♠	J 9 8 3
♥	Q 10 8
♦	K Q 5 2
♣	6 4
♠	5 4
♥	A K 9 3 2
♦	10 7
♣	A J 10 2

Me	LHO	CHO	RHO
1♥	2♠	3♥	All Pass

2♠ was 4 spades and a 5 card minor, 11-15 points.

The defense led a low diamond to the K and Ace. A diamond was ruffed, then a high spade and a spade to the 9 and K. Another diamond was played: I ruffed 9, which held, then played ♣10 to LHO who won and played a trump, ♥10 holding in dummy.

Anyway, after winning the ♥10 I was in this position:

♠	J 9
♥	Q 8
♦	Q
♣	6
♠	-
♥	A K 3
♦	-
♣	A J 2

I'm already down 1, but want the rest of the tricks. So I ruffed a spade high, and ♥K, ♥Q, ♦Q squeezed LHO in the black suits.

Of course, ♥K successfully drew all outstanding trumps, so for this squeeze I had to "rectify" the count by compressing my trump winners. I could have just ruffed a club.

So after 500 hands I can still spot a squeeze, but ruffing in dummy is beyond me.

I'm not sure if the ♣10 was a good play or not. I could have always drawn trumps in 2 rounds and set about losing and ruffing and pitching my 3 club losers. But, on that layout ♣10 was safe, and if LHO had been 4216 and defended the same way I would still have made. I don't see a way to cater to both possibilities, so maybe ♣10 was best (other than leading ♥10 from dummy, except you're not in dummy).

Read more!

LM Pairs declarer play, part II

This was the end position from yesterday's post:

Q9x
-
Qx
Jx
 
KTxx
QT
x
-

You're about to win HQ, at which point you'll have 4 hearts, 1 club, and 2 diamonds won or established. The opponents have 2 tricks in the bank, plus CQ and SA when they want them. It appears LHO is 4=3=2=4. What now?

I decided to play LHO for SA and CQ and avoid guessing the SJ. I cashed 2 hearts pitching a spade and a diamond, then crossed to DQ leaving this position:

Q9
-
-
Jx

KTxx
-
-
-

If LHO pitches 2 spades, I can play SQ to establish that suit. LHO can kill it by ducking SA, but then he gets thrown in to allow the CJ to score. If he holds onto 3+ spades, I can instead duck a club (before or after playing SQ).

In practice, LHO pitched 2 spades, I played SQ, and RHO won SA, cashed the long diamond, and played a club to partner's Q for down 1.

Read more!

LM Pairs declarer play

I declared this hand in the LM pairs:

Q9x
Kx
KQxx
Jxxx
 
KTxx
AQT8
xxx
Kx
 
1D (X) XX (1H)
 X (P) 2D (P)
2N (P) 3N

Not really sure how I feel about our choices in this auction; we definitely had some disagreement about whether 2N should be forcing.

Anyway, in 3N, the defense started CA, club to my K and I played a diamond to the K which held. Not sure what's best here, probably just ducking a diamond, but I elected HK, HA, diamond and LHO won and exited HJ in this position:

Q9x
-
Qx
Jx
 
KTxx
QT
x
-

You're about to win HQ, at which point you'll have 4 hearts, 1 club, and 2 diamonds won or established. The opponents have 2 tricks in the bank, plus CQ and SA when they want them. It appears LHO is 4=3=2=4. What now?

More tomorrow.

Read more!

Spingold bidding problem

My partner thinks he got this one wrong in a Spingold match:

AJTx
KJTx
Kxx
xx
 
1D (1S) X (P)
3D ( P) ?

Would people bid 3N instead of double?

Having doubled, what do you do now?

Result below.



He chose 3N and the opponents ran 5 clubs. At the other table, they bid 4D and got to 5D making. One expert suggested bidding 3S, which is great if it implies club doubt, but it doesn't sound that way to me.

My hand was:

x
Axx
AQJxxxx
QJ

On the lead of SK, 5D is cold (win, draw trumps, pitch a club on SJ).

Read more!

Old rgb hand follow-up

I promised some follow-up on the unusual progressive squeeze from the end of last week. Turns out there's another one in my archives that will be appearing sometime this month, so I'll save my general comments until after that.


Read more!

Another old rgb hand -- Solution

This problem appeared in yesterday's post; the solution is below the fold.
[UPDATE: formatting issues corrected. Also, further discussion coming later.]

Qxx
Kxx
AQ7xx
K8

-
Axxxx
2
AQJ7xxx

Play 7C after a heart lead, LHO originally opened 1S.

Answer below...



Give LHO 5 spades for the opening bid, 3 hearts for you to have a chance, and K 4th or 5th of diamonds (hopefully you can figure it out). Win the lead in *hand* and run 5 rounds of trumps:

    Q
    K          
    AQ7xx      
A   -          
QJ             
Kxxx    
-   -
    xxxx
    x
    xx

On the next trump, LHO is squeezed. A heart pitchgives up 3 tricks. A diamond pitch allows you to finesse and ruff, setting up the suit. So he must pitch the SA. Dummy pitches a diamond.

Now take the diamond finesse, cash the DA and SQ and LHO is trump squeezed:

    Q    
    K              
    7x             
-   -
QJ
Kx                 
-   -              
    xxx            
    -
    x

On the SQ, a heart pitch allows HK, hand good. A diamond pitch allows diamond ruff, dummy good.

Read more!

Another old rgb hand

More mining old rgb posts for material:

Qxx
Kxx
AQ7xx
K8

-
Axxxx
2
AQJ7xxx

Play 7C after a heart lead, LHO originally opened 1S.

Answer tomorrow.

Read more!

Two previously undiscussed situations

Practicing before nationals, we weren't sure what our agreements were after:

1.

(1S)-P-(P)-1N
( X)-?

Where X is just cards.

For simplicity, we decided just to play the same system we would after 1N-(X) showing cards, too. (In our case, P is to play, XX is 1 minor or both majors, 2m is that suit and a higher, and 2M is natural.)

Does anyone have a better idea?

2.

1S-(2N)-X-(4C)

Normally after we double their 2 suited bids, subsequent doubles are penalty by either player. But then does that mean we're in a forcing auction? Seems pretty awkward to treat it that way.

Really once they jump, I think takeout by opener and cards by responder is best.


Read more!

Favorite card combination follow-up

Yesterday I posted a card combination essentially similar to this one:

KQ8x

A7xx

It appears that there's nothing to do here except hope for a 3-2 split. But say you start with an honor from dummy, what is RHO supposed to play holding J9xx? From his perspective, you'll have the ten 2/3 3/4 of the time, and if he never falsecards with the 9, you'll be able to pick up all his stiff 9s (when you have the T).

In fact, RHO should drop the 9 1/3 of the time. When you hold the T, you can do no better expectation-wise than to fall for this falsecard and he'll gain a trick.

But, if LHO has the T, you will score an "impossible" trick.

Note that leading from your hand can't work: you must force RHO to play before LHO.

But, this layout also creates a great opportunity for a defensive Grosvenor Coup: you cash the K and see the 9 on your right and ten on your left. But when you play the Q next, RHO shows out!




Read more!

My favorite card combination

How do you play Q862 in dummy opposite AK73 in hand?

Imagine (perhaps a stayman auction) that your length and top honors are known.

A hint is below the fold...



Suppose you're playing:

Q862 in dummy, opposite

AKT3 in hand.

If you start with the Ace, 5, 2, 9, then what?

Against a decent RHO, the 9 is more likely to be a mandatory falsecard from J9xx (which is a priori 3 times as likely as stiff 9), so you should play the Q next.

Is there anything you can do about this?

More in a subsequent post...


Read more!