September 23, 2013

We had a wild set of hands Sunday night.  On the first seven boards, there were five slam decisions.  The other two hands were played in three spades doubled (down one on one, making on the other)!  I have expressed my opinion before about doubling part-scores.  Mostly, it is don’t!  These two hands weren’t strong arguments against my view, and the combined doubles cost a net seven IMPs. 
I will focus on two of the interesting play problems from the slam hands.  Both tables bid well to six spades on board seven, with

   
The auction at our table, with both sides vulnerable, was

I don’t know what North meant by five notrump.  In the old days, this would have been a grand slam force.  Today, most players use Key Card Blackwood instead, and reserve 5NT as a pick-a-slam bid.  I don’t see the point of either.  You want to play in spades, and seven rates to be too much – AKJ10-eighth of spades is not enough to ensure 12 tricks, let alone thirteen.  But, the play is my concern.  First off, how would you play six spades on a club lead? 

With the ace of trumps to lose, and two potential club losers, you won’t have time to establish the hearts.  On a club lead, you will need a finesse.  You can take the club hook, or rise with the ace, finesse in hearts, and hope to throw clubs on the diamond ace and heart ace.  Since finessing in hearts needs lucky splits as well as a finesse, you should take the club finesse at trick one.

Okay, now consider how to play the hand on a diamond lead.  In particular, do you win the diamond ace early or save it for later? 

Let’s project out the play.  There are eleven obvious tricks, seven spades, two hearts, and the minor-suit aces.  The best chance for trick twelve is the heart suit.  So, first question:  Do you start on hearts right away, or go after trumps?

You may well need the trump queen and club ace as entries to establish hearts, so trumps must wait.  You certainly don’t want to lead trumps and have the defenders knock out the club entry.  So, we will win the diamond, or trump the diamond in hand, and play heart king, heart ace, and trump a heart.  Suppose hearts split out.  What next? 

That’s easy.  Lead a high trump from hand.  Unless trumps are three-zero, we can claim. 
What if hearts were four-two, with the queen still out? 

Now we need an extra entry to dummy, so it looks best to lead a trump to the queen and hope West has the trump ace. 

So, back to my original question.  Should you keep the diamond ace in dummy?  Suppose this is the full hand:

  
Following our projected line of play, we trump the third heart high, and find out the bad news there.  Then try a spade to the queen – still no luck, as East wins.  However, East has no more trumps, and might be endplayed, provided we kept the diamond ace on the table.  A club exit gives us a trick, and a diamond or heart gives us the extra entry. 

Now consider a hand like this:

  
This time, we can’t endplay East, and everything seems to fail, but, in the end, when we run our trumps, we can squeeze West in hearts and clubs, provided, of course, we cashed the diamond ace along the way.  You can follow the squeeze via the next button above.

Neither construction seems very likely, and, if I had to guess, I’d say the first will occur more often than the squeeze.  Now, to get really tricky …  Is there a way to handle both of these positions? 

Yes.  If we take the diamond ace early, and also trump a diamond, then, in the first construction, East is left with a third losing option – giving us a ruff-sluff. 

The best line of play is this:  Win the diamond ace, throwing a club and trump a diamond.  Then heart king, heart ace, heart ruff high, and, if hearts aren’t set up, spade to the queen. 

At the table, West lead a heart.  This doesn’t do much to change the basic line of play, but you might get lucky in hearts at trick one.  I’d try the heart nine or the heart jack, and hope to build up a third heart winner immediately.  Say you try the heart jack, but East plays the queen.  Win the king and return a heart, planning to play as above.  Here, though, you won’t be able to ruff a diamond, so you can keep the diamond ace on the table, to cater to the first layout, or cash it, to set up a possible squeeze. 

Unfortunately, the heart was a singleton, and no line had any chance.  Both declarers lost a quick heart ruff, and finished one off. 

My other exhibit was board three, where slam was reached at only one table:

Our auction was, with East-West vulnerable, 

  
West leads the nine of clubs.  Plan the play.

If trumps split, we can count five trumps, three clubs, the diamond ace and the spade ace.  That’s ten tricks, and we can trump a diamond in dummy for eleven.  Trick twelve looks like it must come from a three-three spade split or a three-three club split.  Maybe this isn’t such a great slam.  Is there a way to combine those three-three chances?

Yes.  Win the club in hand, draw two rounds of trumps with the ace and queen, trump a spade, trump a diamond, trump another spade, and draw the last trump.  Then we can go over to the club queen to try spades, with clubs in reserve.  This line actual has some extra chances.  If West is, say, 2-2-5-4, we will lose the fourth club to West, who must lead into our diamond ace-queen.  Suppose, instead, the hand looks like this:

  

Now, neither suit splits, but, after the ace of spades, we simply play another spade, discarding a club and a diamond.  West, down to nothing but diamonds, will have to lead into our ace-queen (you can follow the play trick-by-trick with the next button). 

This is not a bad plan, but, given the overcall and the opening lead, clubs aren’t splitting, and we shouldn’t bother hoping for a three-three club split.  Instead, we need to set up the spades.  With a little care, we can establish spades even when they are four-two, so long as West doesn’t hold three trumps and two spades.  I think declarer should win the club in hand, play a trump to dummy, ruff a spade, and trump to the dummy, and play spade ace, spade, in this position:

  
Then, if East follows, trump with the nine.  If spades are three-three, draw the last trump and claim.  If West shows out (but can’t overruff), trump a diamond, and play a fourth spade, discarding.  We’ll win any return in hand, draw the last trump, and get to dummy with the club queen. 

If, instead, East shows out, discarding a diamond, things are a bit trickier.  One possibility is to discard on this trick, which will net twelve tricks unless West wins and leads the last trump.  Maybe it is safer to trump this trick, trump a diamond, and discard on the fourth spade.  Could anything go wrong there?

Well, again this might be the full hand:

  
Follow the play above to see the looming disaster.

So, is West more likely to be 4-2-6-1 or 4-3-5-1?  I think, at the vulnerability, the six card suit is much more likely for the overcall, so I would discard on the third spade. 

Can the hand be made if trumps don’t split?  Yes, so long as spades are friendly.  Suppose this is the full hand:

 
Start just as before, but when West shows out on the second trump, win, set up spades, trump a diamond, and run spades.  Whenever East ruffs, discard, win the return in hand, draw the last trump, and go to dummy with the club queen. 

At the table, South started correctly, winning the club in hand, trump to dummy, spade ruff.  But now, declarer led the heart king.  This play would cost the contract if trumps were four-one, or spades four-two, but both majors split, so we ended up +1010 anyway. 

Our declarer played the first hand better than his counterpart at the other table, going after hearts immediately, but the bad heart split made the play irrelevant.  South mis-played the second hand, but the cards were very friendly this time, and the bad play won eleven IMPs.  Sometimes, justice has to wait until the mentors write up the hands.