October 15, 2013 – Is the Best Percentage Play the Best Choice?

Let’s start this off with a simple play problem:

  
You open one notrump, raised to three, and West leads the spade seven.  How would you play this hand?

My title question may seem quite strange.  How could the best percentage play not be the best choice?  That depends, I suppose, on how you calculate your odds.  To take a simple example:  Suppose your contract depends on one of two finesses – 75%.  Instead, you could try a complete swindle play, which will never land the contract unless you dupe your opponents, but these particular players will fall for that ruse every single time.  So?  Is 75% better than a play that can’t possibly work, yet always will? 

I, personally, almost never play for miracles.  If I need three finesses, a break, and a squeeze for my contract, I will always look instead for a con-game.  Defense is hard, and playing for an error is usually better than playing for a miracle.  Such considerations are really also percentages.  How likely is a fatal defensive error?  Is that more likely than your other line?  These are questions you have to ask yourself as you map out a declarer strategy.

I’ll return to this, and my play problem, after a few bidding comments.  Sunday night saw several bidding fiascos.  I’ve said this before, but it is worth reiterating:

     You don’t have to play the best system.  But you and your partner must play the same system!

On one hand, South bid four clubs, Gerber, and signed off in four notrump.  North took four clubs to be natural, bid a natural four diamonds, and took four notrump for key-card in diamonds.  The final contract was not elegant!  Who was right?  Who cares!  Just be on the same page.

The other disaster was much smaller – North held

The auction started out, with North-South vulnerable:

 

This is a pretty good hand.  What are you worth, and what do you bid?

To me, this looks like a five-and-a-half club hand.  I could live with five clubs, I could live with six clubs.  In my partnership, I would bid 4NT, showing a hand too good to bid five clubs.  At the table, North found another way to express a hand too good to bid only five clubs – North passed, intending to pull a double to five clubs.  This was excellent evaluation – a forcing pass seems just right.  Trouble is, everyone else passed too.  To North, this pass was forcing, but not to South. 

     You don’t have to play the best system.  But you and your partner must play the same system!

Should this pass be forcing?  I don’t know.  I use very simple forcing pass rules:

 

1.  If we are in a game force, and the opponents compete, passes are always forcing.
2.  If we have shown game-invitational values, then passes are forcing at low levels and at high levels, but not at the three or four level. 

Two simple illustrations:

            1C – (1H overcall) -2H – (2S) – pass is forcing.

            1C  – (1H) – 2H – (3H) – pass is not forcing.

As for the auction above, perhaps the three club call, under pressure, shouldn’t promise game-forcing values.  If so, North’s pass isn’t forcing.  Or, perhaps three clubs commits us to game, and the pass was forcing.  I don’t feel strongly either way.  Just be sure you and your partner agree!

Let’s turn now to the play problem, from hand four.  At the other table, South upgraded the hand, and got even higher –

 

This table also avoided the killing club lead.  How should four notrump be played on the spade seven lead?  

There are eight tricks off the top.  In three notrump, you have chances for a ninth trick in either hearts or diamonds.  In four notrump, you will need diamonds.  There are multiple ways to tackle the diamond suit.  These are:

 

1.  Bang out the two top honors – the best percentage play in the suit.
2.  Start with the jack from dummy, trying to coax a cover.  If East does not play the queen, win the king and finesse West for the queen.
3.  Start with the ten from your hand, again trying to coax a cover.  If West plays low, put up the ace and finesse coming back.

Then there is a vastly inferior approach that may still be best – win the spade in dummy and lead a diamond to the ten.  Why in the world might that be a good play?

Simple.  Half the time that finesse will win.  If it loses, West has to find the club shift, without a lot of information.  Since West didn’t think a club opening lead was attractive, maybe …  I have a feeling that this might well be the best line of play in four notrump.

Of the two “let’s tempt a cover” lines, neither has much chance to work at the table where South opened with one diamond, but they might succeed after a one notrump opening.  Which of those is better?

If you start with the jack from dummy, you are screaming that you hold strong diamonds, and very few defenders will cover.  The ten from hand is much more likely to be covered – isn’t that how you would tackle a suit like:

 

If West does not cover from Qx or Qxx, the defenders lose a trick. 

By the way, what do you make of the spade seven lead?  Is that fourth best?

No.  That would give West 10987(x) in spades, and West would lead the ten.  So West has chosen a passive lead, and is quite unlikely to hold a five card club suit.  That suggests another approach to the hand – cash out the spades and hearts, and try to count out the hand.  In 3NT, if hearts split, you will take your nine winners.  Even if they don’t split, you will know the heart distribution, have a pretty good idea of the spade count, and you can infer at least five clubs on your right, so guessing the diamonds might be easy.

I don’t believe that there is a correct way to play this hand.  Personally, in 3NT, I would win the spade in hand and try the diamond ten, to the ace.  Then I would test hearts, finish spades ending in dummy.  If nothing good happened, I’d try a diamond to the nine. 

In four notrump, where South had bid diamonds, this “cover” ploy won’t work, so I would try to count out the hand, and make my diamond guess late in the day.  But, as I said, leading a diamond to the ten at trick two is a very practical line of play, and might well work out more often than my scientific line. 

I’ll close with board seven, an incredibly tricky three notrump contract.  Take the West seat and think about your opening lead:

What would you lead?

You can never get blamed for leading partner’s suit, but here, with a club entry, a diamond looks right.  As little as the diamond queen in partner’s hand should set the contract.  But that doesn’t answer my question.  Which diamond is best?

The jack is the normal lead, but the five could turn out to be the big winner.  Imagine that the diamond suit looks something like this:

 

Only the five lead would avoid blocking the suit.  Still, I won’t worry much about five diamonds in dummy – on many of those hands, North would rebid diamonds.  Here is another potential layout:

 

 

Here, you have to lead the diamond king to clear the suit.  Obviously, the king is also the big winner when South has the singleton queen.  That seems like the best lead to me, so I would lead the diamond king, trusting partner to overtake and return the suit on this type of layout:

 

 

At the table, West tried the diamond five, which, as I said, could certainly work out well.  It didn’t, as it ran to declarer’s doubleton eight:

South won the diamond eight and led the club queen, which held, then club ace, club ten, to West’s king, throwing a heart from dummy, while East discarded a spade and two hearts.  North won the next diamond, to leave this ending (which you can also see hitting the next button above):

  
South, with five obvious tricks, needed four more.  When the spade was led from the table, East won, and?

Obviously, the defense gives up if South can win a trick, so East must break hearts.  In practice, East played a heart now.  North won three heart tricks, the diamond ace, and lost a diamond to West.  West had another diamond to cash, but had to give South a club at the end.  It looks like it might have helped had East cashed another spade, letting West discard that club loser.  But, after three hearts and the diamond ace, dummy can exit with a heart, for South to win the last trick with a spade. 

You can follow the two lines below:

To set the contract, East had to have kept another heart, and one less spade.   Picture this ending:

 

Now East can cash both spades, and exit with a heart, and the defense must come to five tricks one way or the other.  Should East have found this defense? 

I think so.  Once South set up clubs, East should certainly see that giving South a spade trick would be disastrous, so East should start throwing spades like crazy. 

Since the contract could have been set despite the disastrous opening lead, I figured that the hand could be set easily on the diamond jack or king lead.  I tried the GIB button on BBO and found, to my shock, that, double-dummy, 3NT is cold on any lead!  Go back to the beginning of one of the BBO diagrams above and click on the GIB button.  How could that be?

Well, let’s try the diamond king lead.  North wins, and leads a spade.  East probably does best to duck, and let South win.  South leads the club queen, ducked, and (finding out the bad news), loses a spade to East.  This will leave:

 

 

  

What can the defense do?  If East plays a heart, North wins three hearts, and the diamond queen, and the puts West in to lead away from the club king.  Okay, how about spades from the top?  West can let go of a club and two diamonds.  But South wins the spade, and puts West in with a diamond.  West can exit a club and let South win the rest, or a heart and let North win the rest. Again, you can follow the play here:

 

The toughest opening lead is probably a club, attacking South’s entries to a spade winner, but the hand is still cold.  I’ll leave you with that puzzle – figure out how 3NT can be made on the lead of the club eight.