Defensive problem

[kenberg]

People seldom respond to my posts so I posed a question from the Common Game on Bridge Winners. Not sure I will get any replay there either, but I can try.

http://bridgewinners.com/article/view/common-game-2018-03-23-board-9/

[OliverC]

Hi Ken, playing Standard carding I would definitely be playing the 4 on trick one (ie: giving Count). Clearly North cannot have 5-card Hearts on this hand, so that places you with 3-card Hearts (if you had 5, it really doesn't matter WHAT happens on the next trick) and Declarer with K?xx.


If Declarer has KJ9x, then they'd have let the opening lead come round to their hand rather than going up with Dummy's Queen, so clearly they are likely missing the 9 or the Jack (or both). Either way, placing you with 3-card Hearts should help, because the likelyhood is that Declarer has either KJxx or K9xx in Hearts, so two leads through Declarer's hand is going to be essential to bring 3 tricks in in Hearts.


That will hopefully lead them to seek an additional entry in your hand in Spades or Diamonds ( King Lead a good shot), or to simply play passively and allow that additional entry in the West  hand to develop if it's there.

[kenberg]

I find some hands very interesting and this is one of them. I played the 7, it worked badly, I later thought I should have played the 4, I went home and got Stewart's book out and found a similar hand, and so on. I'll say more at the risk of sounding obsessive.

I'll start with a little more about Stewart's example hand. Actually Stewart borrowed the hand from Woolsey, and his comment is worth reproducing [I think I will easily be on the proper side of copyright laws here]: "This hand is drawn from Partnership  Defense in Bridge by Kit Woolsey,the best treatment of defensive signals available. When this book appeared in 1980, I thought it was the most important work in defensive play since Hugh Kelsey's Killing Defense was published in 1966."  That's quite a tribute. I have Woolsey's book but it has been a while since I have looked at it. I do like both his writing and his approach.


Anyway, here is the Woolsey/Stewart example hand:

   Dummy
  Q75
  AT94
  K82
  T73

                 Third hand
                  J96
                  K73
                  J65
                  AJ82

The auction is 1NT-3NT, the 2 is led, dummy plays the Q, and Stewart/Woolsey say, "you signal with the 9".

So I need to think more about this. A few pages back Stewart has a hand where a 6 is led, dummy plays the 6 from J92, and the recommended play for third hand, holding only spots, is a count signal.  This seems to be the dividing point: Count when dummy has and plays the J, attitude (possession of the J) when dummy has and plays the Q. But as Stewart also notes: "Not all attitude situations admit to such an easy interpretation".

In the case at hand I am confident that I will be getting the lead before partner does. Moreover, when I then lead the J he will play me for three cards since if I held four cards I would presumably have led back my original fourth best. I cannot be certain whether partner's lead is from 4 or 5 cards but 4 is more likely, giving declarer four. It' not so much that I can think out the full problem exactly, but definitely I don't want partner, after JKA3, to play the T hoping to drop my non-existent 9.

The play went to the Q, to the A, to the T, I am in. I lead the J to the K and A and partner is in. If I had played the 4 at T1 partner still would not know whether that was from J74 or J94, but at least the 4 might have slowed him down. Declarer  holds the K else she probably has the J (she has to  have something somewhere)  and would have played a small club to the J, even if it lost hearts could not be continued by E. I gave count in clubs, declarer must have been dealt a 3=4=3=3 shape since with four diamonds she would have opened 1 and with four spades she would have raised spades. But now it is looking strongly like I have the Q. Partner can reason that even if playing the T drops the 9 we get three hearts and one diamond. If we are going to beat this I need to have the Q. But also, declarer could have started on spades instead of diamonds had she held Qxx.  It's one of those choice things. If she lacks the Q she has not choice put to play on diamonds, if she has the Q she could go for her 9th trick in either spades or diamonds. But it's more than that. Imagine she has the Q.  Say she comes to hand with a club and leads the Q. If E holds the  K and covers, she could duck. She has now established 9 tricks, E cannot continue hearts, and if spades break she gets a 10th trick. So it seems likely she would play that way if she had the Q. In fact, had she held the Q she might well have led low toward it from the board at T2. Why come to hand if when she leads the Q she pans to play low whether or not the Q is covered?  Even if I have the Kxx in there is a non-trivial chance that I won't rise. Doing so establishes three spade tricks for her and so could be wrong at least at matchpoints. 

So I think pard might well have decided that there was no hurry to cash the T  had I played the 4 at T1. Playing the 7 convinced him that there was no reason to wait. he could just cash the hearts.

Coda: Amusingly, in the Woolsey/Stewart hand the original signal is in spades, but the key is in clubs. Declarer comes to hand in diamonds and leads a heart to the Q and East's K. To beat the hand, E shifts to the J, something of a super-surrounding play since W holds K94.
Having borrowed from Stewart and from Woolsey via Stewart I might mention that I recommend both of these authors.

 Added: I just dug out the Woolsey book. Yep, it's a good book. It's old, but we senior citizens do not always consider that a defect.

[OliverC]

Ken, Obviously partnership style has a lot to do with this kind of problem. In the absence of that my tendency is to give Count more often than attitude, on the basis that helping partner to count the hand is usually more important than showing attitude. If partner can work out that I have a 3-card holding when they hold A1086 and Dummy has Qx, it doesn't really matter what exact Hearts I started with: he always needs to try to engineer an opportunity for me to lead Hearts through Declarer once, if not twice, whether I hold xxx, 9xx, or Jxx.


It's only when I hold J9x and they have A108x that dropping my 9 is a worthwhile option and when encouraging with the 9 gives us any real advantage. If Partner started with A10xx or A8xx, we're back to the original point of needing the two leads through Declarer.

[kenberg]

Yep, this seems right.
I played once with a guy who favored giving nothing but count signals. Obviously a bit extreme, but in fact if you can work out the shape then often you can work out the right play even if you are not sure just who holds which spots.
He was not a regular bridge player but he was very good at problem solving in general and he felt if he knew the shape he would work out the play, and pretty often he was right.
I am not that extreme, but I do think that knowing shape, combined with a little reflection on what declarer did and did not do, can solve a lot of defensive problems.

[OliverC]

One of the reasons I prefer to play Odd/Even signals with standard Count, is that very often I can encourage/discourage and give a count signal simultaneously. I'm aware, of course, that the ACBL have a problem with o/e signalling (not that I give two hoots what the ACBL like and dislike, because I don't live in the USA and never play in anything held under their auspices on BBO).

I'm a great believer in giving Partner the signal I think they need in any given situation if I can't combine them (ie: Attitude, Count or Suit Preference) in the belief that Partner ought (hopefully) to be on the same wavelength. Obviously, their expertise and experience is a factor in my expectations. Certainly I find Count signals from Partner more helpful, because I often have a good idea about their attitude from the bidding or early play. Counting the hand as accurately as possible is a really crucial aspect of the play of the hand (whether as Declarer or in defence), and something I do religiously on every hand as much as is possible.

Counting the hand is probably the single thing that distinguishes good players from average ones, and their inability to do so consistently and accurately (not to mention the ability to take all of the inferences and assumptions that flow from counting the hand accurately) is what holds probably the majority of players from really improving their card play.

Count to the exclusion of everything else does seem a little extreme, though, I agree. There are some times (particularly at trick 1 against a suit contract when you are leading a suit Partner has not bid) when their attitude to the lead can be critical, but even then, Count is also useful if you can do both.