Sunday, May 29, 2011

More on (moron?) "Impossible Two Spades"

A discussion on BBF underlined yet again a concept that I believe to make sense.

The origin to the idea was a combination of a discussion with Ken Eichenbaum in Philadelphia concerning a specific "exception" where one might suppress a 4-card spade suit after a 1H opening, coupled with a Bridge Bulletin hand with the same principle.  The idea is that with a light hand, a practical response to a 1H opening with four spades and six clubs is occasionally 1NT, planning to bid clubs later.  (For the skeptics, the hand in question in the expert forum was handled frequently with that approach).  This suggested that a "impossible two spades" is no so impossible:

1H-P-1NT-P-
2D-P-2S = weak with four spades, six clubs?

This could be tweaked somewhat to include 3136 hands.  This caters well to the hand where Opener (not playing Flannery) has 4531 shape and finds himself "forced" to rebid 2D, disliking this, of course.

The first question, then, is whether Responder or Opener should have the fourth spade (to bid "impossible 2S" or to pass the same).  It seems that the hands where Responder bypasses a 4-card major are less frequent than the 3-card options for Responder, such that Opener should be 4-card to pass.

This, then, leads to the BBF problem.  Opener has a trashy hand with 4630 and lousy (9-high) hearts.With the "not impossible two spades" approach, 2D comes to mind, as it seems to increase our chances of landing well (you could end up in 2S for a fine result, and 2D therefore is less frequently passed when that is bad).  One concern was propelling the auction when Opener has 5-5 reds, but it still seems that you end up OK when and if Responder hits Opener with the "not impossible two spades" response.

Tuesday, May 17, 2011

To Cue a Side Queen? Or, "Almost Last Train?"

I may have erred in my bidding ecently, but I think I was right.  If so, an interesting sequence came up.

Critical to the problem is that Responder initially passes, advertising less than an opening hand.  After two passes, third seat opens One Heart holding:

AKxx-AKxxxx-Kx-x

The opponents pass throughout.  Responder bids One Spade. Opener now leaps to Four Clubs. 

Initially, I thought that perhas my 4C call was perhaps poor.  This hand is extremely powerful.  But, two factors convince me (feel free to comment if you disagree) that 4C is right.  First, the splinter is opposite a passed hand, which by definition is limited; hence, the splinter should probably be a tad richer than a similar splinter opposite an unpassed hand.  Second, the space available for more exploration is maximized after a cheapest possible splinter, which allows Responder space to dangle carrots with les cause.  (After 4C, partner can bid 4D, bid 4H, bid 4D and accept-or-reject 4H; after 4D, only 4H is available to invite slam back.)

In any event, back to Responder, who held:

Qxxx-J10-Qxxx-Axx.

The other problem to consider with the splinter is that the holding of the AK-AK-K in trumps, hearts, and diamonds means that partner is less likely to have traditional cues available.  In my cuebidding style Responder can cue 4H with the heart Queen (if that call is a traditional cue), but he will only be able to cue diamonds with the Ace.

Or will he?

This sequence made me think of two alternative possibilities.

First, one might decide that a cue of 4D simply shows one of the top three honors in diamonds.  The idea would be that a splinter bid inferentially is the same (or similar) to bidding the suit (or suits) that are not short.  For example, if you assume a pure splinter in rsponse to a major opening (e.g., 1S-P-4C) as showing a pure 4441 pattern (it does not, but bear with me), then this is sort of like bidding spades, diamonds, and hearts at the same time, at least in theory.  Thus, cuebidding "partner's shown suit" would show (per my agreements) one of the top THREE honors, not one of the top TWO.  Extending this out too much would perhaps be insanity, but the general principles could be that a space-deprived cue of an inferentially-shown suit might be the Queen.  In this situation, using that idea might justifya 4D call by Responder (planning to reject a 4H last train try).  Opener would accept any sniff.

Alternatively, I wonder whether a different principle might make sense here.  When only two calls are available (the three-under splinter being the classic example), should cues be cues?  Or, should cues be Last Train and Almost Last Train?  LTTC and ALTTC?

ALTTC allows a counter LTTC.

It seems like a "critical card" concept might work well in this situation.  LTTC might deny the critical card but show extras otherwise.  ALTTC might show that critical card but not necessarily defining "extras" versus "EXTRAS," with a counter LTTC (CLTTC?) asking for general strength.

In the actual auction (1H-1S-4C), the "critical card" might be diamond control or it might be a missing heart card.  I'm not sure which would be more important in the long run.  I think the missing honor (assuming a ore frequent semi-solid hearts than solid) in a splinter-maker's long suit (if one had been shown) is probably a good candidate for the default "critical card."  Assuming this:

1H-1S-4C-4H = LTTC = extras but no heart honor
1H-1S-4C-4D = ALTTC = at least mild interest, plus a heart honor
1H-1S-4C-4D-4H = do you have mild or strong interest?

What about a non-descript sequence, like 1S-P-4C?  Now, there is no focal side suit.  In that event, the default might be internl trumps:

1S-4C-4H = LTTC = extras but poor trumps
1S-4C-4D = ALTTC = good trumps, extras to EXTRAS
1S-4C-4D-4H = extras, or EXTRAS???