Wednesday, March 30, 2011

A difficult hand

At the club the other evening I held this hand ♠64 J983A82 ♣JT94 and heard the following auction, starting on my right (we were silent throughout): 1 - 2♣ - 2NT - 3♠ - 4♣ - 4NT - 5 - 6♣.  I was in the passout seat and considered briefly whether to double.  Clearly this contract was going down unless somehow declarer had a void in diamonds or could quickly pitch a losing diamond.  The Roman keycard blackwood sequence suggested no voids, however.  So it seemed prudent to pass and await developments.  There was another very good reason not to double and that was because double would ask for an unusual lead.  Since I could have doubled 5 for a heart lead, it wouldn't be clear what an unusual lead would get.  A heart lead could easily be disastrous if partner had say the K (he did).  Finally, there was the fact that I knew that they were in a bad contract and so I was happy.  Why tip them off?  Stay with happiness.

Somewhat surprisingly, partner led the K and we took the first two tricks.  I subsequently got my club trick and so we were +200.  Obviously we were getting a top.

But wait!  The hand proved to be difficult for our "team-mates" too and, believe it or not, +200 was a common result for our direction.  There was a 500, three 300s, and four other 200s.  So our score was only 15/21.  No other pair was in 6♣ (two were in 5♣ down 1) and the majority of the blood was shed in spade contracts.  Our side doubled only once: a 4 contract (that accounts for the 500).

So let's see how this should have been bid.  Let's put ourselves in the dealer's seat with: ♠K93 AQT62973 ♣K5.  After the initial 1 call and the game-forcing 2♣ response, what should the rebid be?  In my opinion, 2 is automatic.  2NT would be fine if we had some sort of honor in diamonds, but we have nothing.  Of course, we hate to bid 2 with only five pieces, but partner knows it's a possibility so should take that into account.

Partner will now bid 2♠, completing the picture of his hand as being at least 5-4 in the black suits.  Now what?  Again, I think the correct bid is automatic: 3.  This, the fourth suit, is artificial and says I have no obvious bid -- I can't support either of your suits and I can't bid notrump because I don't have anything in diamonds.  Partner will bid 3NT with a diamond stopper and all will hopefully be well.  If he happens to have 4045 shape, he can "raise" diamonds.

In this case, partner doesn't have a diamond stopper and will now bid 4♣, hopefully showing a sixth club.

At this point, I think pass stands out but I wouldn't call it automatic.  5♣ might be right.  Most partnerships don't consider a 2/1 game force to be forcing to a minor-suit game when there's no fit and we can't play notrump.  We do have quite a nice club holding, but there again, we have a bare minimum hand.  In this case, 4♣ is the last making contract.  My LHO's hand, by the way, was ♠AQ72 5T4 ♣AQ8732.

The key to bidding this hand, I believe, is our old friend the fourth suit.  It's amazing how often people fail to use it when it's the right bid.  I've even heard it said that since we're already in a game force, how can it be artificial?  That is nonsense of course.  Like "double", the fourth suit is an under-utilized bid that should be our friend.

Thursday, March 10, 2011

To alpha or not to alpha, that is the question

Whether 'tis nobler to play precision at all is another question.  Many pairs play what I think of as semi-precision.  The 1♣ opening in such systems essentially is used to divide the bidding space into good opening hands and, by elimination, bad opening hands.  After the first round, most of the calls are natural.  These pairs seem to suffer from all of the disadvantages (particularly the 1 opening) while enjoying none of the advantages.

I'm in the more traditional camp.  Memorize and use all of the asking bids (alpha, beta, etc. – although they also go by other more prosaic names) and get the most out of the system.

The question that I set out to consider is whether or not to use alpha and, in particular, "second" alpha.  It seems to me that there are hands that are suited to alpha bids and hands which are not.  It's a question of figuring out the relevant frequencies.

Here's how alpha is supposed to work: you pick up ♠AQT72 AK7322 ♣A5, a nice hand by any measure.  You open 1♣ and partner bids 2 (or 2♣ if playing transfer positives) showing diamonds and enough points for game.  Naturally, you bid 2♠ ("alpha") showing five spades and asking more about partner's hand.  Partner bids 2NT denying Qxx/xxxx or better in spades and fewer than four controls (A=2, K=1) in total.  Actually, there are five responses in all (in steps: first, second... fifth):
  • bad support, 3– ctrls;
  • bad support, 4+ ctrls;
  • good support, 3– ctrls;
  • good support, 4+ ctrls;
  • very good support (Qxxx or better), 4+ ctrls.
Now, we trot out 3 ("second alpha").  This time partner has only three responses, since we already know whether he has four or more controls from the previous response:
  • bad support;
  • good support;
  • very good support.
Let's say that this time partner bids 3NT (good support for hearts).  It probably doesn't matter whether we are playing second alpha or natural at this point because we are going to stop in 4 regardless.

But suppose partner's first response had been 3♣ (bad support for spades but 4+ controls).  Now, when we bid 3, we get the second alpha response of 3NT, leaving us the entire four level for more bidding.  If we were playing natural bids at this point, partner would be obliged to bid 4 to show the support.  Let's give partner the following hand:  ♠K3 Q65AKJ92 ♣J94.  After 3NT, 4♣ is beta (hopefully!) and will elicit the response of 4 (exactly four controls).  Now, we follow up with 4♠ ("espilon" or control-asking bid – presumably, there's no law against invoking epsilon in a suit previously alpha'd?).  Partner bids the third step (5) showing second round control (either the K or a singleton).  We're running out of room a little (we'd like to know if partner has ♠K(x) or ♠x but bidding 5♠ will commit us to 6 anyway).  But, just for fun, we'll bid 5♠ and partner responds 6♣ (K).  We can now bid 6, 6NT (played by partner) or even 7/7NT if we're really stretching.  If both major suits behave, we will have all thirteen tricks in hearts or notrump.

To conclude:  after a minimum/no-support game force response to alpha (1st step), we might as well go to natural bidding because slam is generally (not always) out of reach and getting to the right game contract should take preference.  But, after a four-control response (2nd step), our next new suit should be "second" alpha because slam is still very much in the picture.