Wednesday, November 4, 2009

Board 36

If you click on the link "Board 36" below, you will be able to download a file called "Board 36.PBN".  If you own a copy of Jack or any other bridge-playing software that can read a PBN file, you can then play the deal yourself before reading this post.  You can bid it as South if you wish, but switch to the West seat for the play.

Board 36
Both sides vulnerable

♠ Q 10 A Q 6 5 4 5 ♣ 10 7 5 3 2

Partner opens one spade in second seat. I bid one notrump, partner bids two spades, and I pass.

At the other table, Jack raises to three spades in this position, and opener passes. I didn't even consider three spades, at least not for very long. This hand isn't going to play very well unless partner has a heart fit. Even three may be too high. As a general rule, you should be conservative with invitations that take you from the two-level to the three-level. When you make such an invitation, there are two bad things that can happen: partner can refuse the invitation and go down one or partner can bid game and go down (sometimes even if you belong there). Once you're already at the three-level, the odds are different. You can be more aggressive about bidding on.

There wasn't much to this deal from declarer's point of view. So I'm going to move you over to my teammate's seat at the other table and have you defend three spades after the lead of the four of clubs:


NORTH
♠ Q 10
A Q 6 5 4
5
♣ 10 7 5 3 2




EAST
♠ 9 6
J 10 8 2
10 9 8 3
♣ A K Q

West
North
East
South
Pass
1 ♠
Pass
1 NT
Pass
2 ♠
Pass
3 ♠
(All pass)

Partner and declarer have 22 high-card points between them. If you're going to beat this, partner needs his fair share. That means he can't have a lot of shape, since he didn't bid over one spade. A six-card diamond suit is probably out of the question. This makes declarer either 6-2-3-2 or 6-1-4-2. Declarer should have a high diamond honor, since partner didn't lead the ace of diamonds. And if that honor is the ace, declarer should have the queen as well, since partner didn't lead the king.

You have two club tricks and no heart tricks, so you need to find three tricks in spades and diamonds. Two possible defenses spring to mind: (1) win the first trick with the club king and switch to a trump, or (2) win the first trick with the club king and switch to the ten of diamonds. One thing you shouldn't do is win the first trick with the club queen. Since you're a passed hand, you might as well just show declarer your hand if you're going to do that. You want declarer to worry about your holding the heart king--or maybe the spade king or the diamond king. Partner will be deceived, too, but the deceit isn't apt to be material to partner. As long as he knows you have two club tricks, he doesn't really care whether you or declarer has the queen. In fact, the falsecard may actually help him. If you win with the queen, it may be possible from his point of view for declarer to have a singleton. If you win with the king, he knows you have a second club trick.

How do you decide between defenses (1) and (2)? You begin by deciding, in general terms, what you are hoping for with each defense. Then you construct some specific layouts where one defense works and the other one doesn't. This last step is important. It's surprising how often you will discover that, for one defense or the other, you can't construct a winning layout that's consistent with what you know about the deal. Failure to construct a specific deal where your defense is necessary is probably the single most frequent cause of defensive errors.

First the generalities.  There are three scenarios we are concerned with:

(A) Declarer needs to ruff diamond losers in dummy. (In other words, he can't dispose of his losers in some other way, such as establishing and running clubs.) In this case, you want lead trumps. (In the interest of completeness, let me point out that this isn't always true. Although it doesn't apply on this deal, sometimes the best way to stop ruffs is not to lead trumps but to attack declarer's entries to his hand.)

(B) Declarer does have another way to dispose of his losers, namely by pitching them on dummy's clubs. But you have diamond winners to cash off the top. The winning defense is to draw dummy's trumps so you can cash them before declarer can take his discards.

(C) As in (B), declarer can pitch his diamond losers on dummy's clubs.  This time, however, you don't have the tempi to draw trumps and cash diamond tricks. Instead, you need to lead diamonds to force declarer to ruff them (before he can discard them) in the hope of establishing trump tricks for the defense.

Now let's generate a specific hand for each scenario. We will get some idea of how likely each scenario is by how hard we have to work to construct an example. I'm going to warn you ahead of time that this is not going to be a complete and thorough analysis. When I consider a possible hand for declarer and say that a certain defense beats it, it may well be the case that declarer has some alternative line of play that enables him to make his contract. That's not important. What we're trying to do here is not replicate the detailed discussion you might have at Wo Hop after the session. It's to replicate your thought processes as you sit at the table with everyone glaring at you, trying to decide how to defend. Your goal isn't to figure out every variation to the last detail. It's to figure out, in broad strokes, which defense looks stronger.

Scenario (A)

♠ K J x x x x x A Q x x ♣ J x

With only one heart entry, declarer can't establish and run clubs. He needs to ruff diamonds in dummy, and a trump shift puts a stop to that. After two rounds of spades, declarer can come to eight tricks with a heart finesse, but he can't find a ninth. Note that you can't afford to cash a second club before playing trumps. In constructing this deal, observe that I gave declarer a singleton heart so that he had only one dummy entry. A singleton heart might well be a prerequisite for scenario (A).

We've shown that a trump shift works.  But that's only half of our task.  We must now see if a diamond shift works as well. If you play a diamond, declarer can win, ruff a diamond, ruff a heart back to his hand, and ruff another diamond. Now how does he get back to his hand to draw trumps?  Unless he has the eight of spades, he can't.  Even if he's missing the seven, he has problems.  It might help to take a red-suit finesse, but which one?  And how does he manage if we modify the hand so that he's missing the jack of spades.  In short, a diamond shift might or might not beat him.  It depends on how good his spades are and on what line he chooses.

Scenario (B)

♠ A J x x x x K x K x x ♣ J x

Declarer no longer needs to ruff diamonds. He can establish and run clubs if we give him enough time. But if we clear dummy's trumps, we can cash diamonds when we win the second club. Again, cashing a second club before shifting to trumps doesn't work.

How about a diamond shift?  It isn't fatal as long as we work out to switch to trumps at trick three.  Say declarer covers the diamond ten.  Partner takes his ace and plays a spade.  Declarer crosses to the heart king and ruffs a diamond.  Now two more hearts, pitching something.  Partner ruffs, crosses to you in whichever suit declarer didn't pitch, and you lead a heart, promoting his spade king.  Declarer may have a different line that works. But that's not important.  The important thing is that, even though a trump shift is stronger, a diamond shift isn't necessarily fatal.

Scenario (C)

♠ A x x x x x x x A Q x ♣ J x

We don't have the tempi to draw trumps and take diamond tricks. Say the defense plays two rounds of spades, and declarer plays a second club. You win and switch to diamonds. Declarer rises with the ace, draws the last trump, and uses his two dummy entries to ruff out clubs and run them.

A trump shift doesn't work, but a diamond shift might. If declarer rises and plays a club, you play another diamond to tap dummy. Partner now has two trump tricks. Success for declarer hinges on whether he can dispose of his last diamond. And whether he can do that or not depends on how high partner's spade spot is. If partner has the seven or eight, declarer is in trouble. If he ruffs a club, plays a heart to dummy, then plays a winning club, you can uppercut with the nine of spades. If he ruffs a club, then ruffs a diamond, he can't get off dummy.

Can he make it if he takes a heart finesse at trick three before playing the second club? Again, that can wait for Chinatown. It's clear that a spade shift gives declarer no problems and a diamond shift does. At this point, that's all we care about.

What have we learned? One thing we've learned is there's no clear answer. Sometimes a trump is necessary, sometimes a diamond is necessary, and sometimes either will work. A trump shift would seem to do better when declarer has a singleton heart. When he has a doubleton heart, rendering the club suit more of a threat, the relevant factor seems to be how good partner's spades and diamonds are. The better his spades, the more likely the tap is to work. The better his diamonds, the more likely a trump shift is to work.

So which play is right? My impression is a diamond shift works more often. Whenever we found a layout where a trump shift worked, a diamond shift might work also, depending on a variety of factors.  In deal (C), where a diamond shift was better, the trump shift accomplished nothing.  I only looked at three hands, so that's hardly conclusive, but that's about all you have time for with the director breathing down your neck.  Remember, this blog isn't about clean solutions; it's about how to approach the problem.  And that's good, because this problem may well not have a clean solution. That's why you would never see it discussed in a bridge book even though it's the kind of problem you face all the time.

We have accomplished one thing, though. If we choose incorrectly and partner asks, "What exactly were you playing for?" we can answer him. It's embarrassing to have to think about it and then admit that your chosen defense was never superior to the alternative on any plausible layout.

Jack, incidentally, does not adopt either of the suggested defenses. At both tables, defending both two spades and three spades, he wins with the queen of clubs, cashes the ace of clubs, then shifts to a trump. It's hard to see how cashing a second club can be right. That's simply doing declarer's work for him. (What exactly were you playing for, Jack?) At both tables, declarer rode the spade around to to West's king, then easily took the rest, making four:


NORTH
♠ Q 10
A Q 6 5 4
5
♣ 10 7 5 3 2


WEST
♠ K J 8
K 9
K J 7 4 2
♣ 8 6 4


EAST
♠ 9 6
J 10 8 2
10 9 8 3
♣ A K Q


SOUTH
♠ A 7 5 4 3 2
7 3
A Q 6
♣ J 9



I was curious to see how Jack would declare after a diamond shift, so I replayed the deal with myself as East, defending three spades. I won the club king and shifted to the ten of diamonds. Declarer finessed the queen. Partner won with the king and shifted back to clubs. I won with the ace and played another diamond. Declarer took the ace, ruffed the diamond, ruffed a club, and played ace and a trump. Down one.

Me: +170
Jack +170

Score on Board 36: 0 IMPs
Total: +91 IMPs

No comments:

Post a Comment