Wednesday, June 23, 2010

Dan Dopps

I recently learned of the passing of a friend and former partner, Dan Dopps of Mansfield.  He was one of the four original contributors to the Modified Italian Canape System and played this regularly for many years. 

One of my fondest memories was our continuing debate about whether Losing Trick Count, or counting Winners, made more sense when dealing with unbalanced hands.  I ended up winning that debate, in a strange fashion.

Dan opened One Diamond, showing any number of possible hands (long clubs, long diamonds, or diamonds with a longer major).  I made some response, and he made a jump rebid showing, and now the debate...

1. Dan's position: 9 winners.
2. My position: 4 losers.

Seems like not much of a debate?  Dan thought that my approach was bizarre, and pessimistic.  I thought his approach was the exact same thing (I thought it was) such that using terminology of "losing tricks" made sense.

However, we soon found out why I was right.  With a CLEAR three tricks, I forced the small slam, which failed by one trick.  Dan thought that my hand was not good enough to force the slam, which seemed odd to me.  I mentioned that my three covers took care of three of his four losers, which suggested that the slam would make.  He retorted that I should count winners.  So, I added 9+3 and got 12.  He grumbled, I went for a smoke.  Next round.

Well, mid-auction on the last board of the next round, the director came over to ask Dan which of the 14 cards in his hand was not supposed to be there.  In the end, all 14 were in fact in his hand when he bid and played the hand earlier (a claim, and two people who could not count).

This sealed my victory in the debate.  Counting winners, he had my 3 tricks plus his 9, for 12 tricks.  But, counting losers, he had 14 cards, minus 9 winners, is 5 losers!  Thus the one-trick set.

So, if 14-card hands and 12-card hands occur much for you (especially in Rubber or party bridge), then I propose that LTC works much better in the long run than Dan's optomistic winners count.

But, I suppose that Dopps Count might help in some situations, where you need tricks and can claim, as long as the opponents do not grab theirs first.

I also remember two incidents that amused me.  We always discussed theory and missed the world around us.  In one event, we left with a round to play because the last hand in the next-to-last round was too fascinating for us to focus on the game, and we got confused.  We also accidentally entered a novice game for a similar reason, scoring up about a 260 on a 156 average (playing canape in the novice game).

Friday, June 11, 2010

The Strange World of Delayed Canape?

In practicing for Philadelphia with my good friend and partner, Ken Eichenbaum, we have discussed a very unusual phenomenon, the second-round canape.

Here's a few examples:

You are dealt 4-5-1-3 pattern and a fairly decent hand.  You open 1H and partner makes a forcing 1NT call.  Without sufficient strength to reverse, you opt for the clear 2C call.  Partner now trots out the "impossible 2S," showing values and club support.  So, you launch back 3S, which we now play as showing four spades and only three (or possibly two?) clubs, but a maximum.  A canape, in that your THIRD SUIT is longer than your SECOND SUIT.

How about another one?  You start with 1453 pattern with extras.  1D, but then partner bids 1S, which causes you to slow the auction down a bit.  So, you opt for a 2C rebid.  Partner courtesy-corrects to 2D.  You now bid 2H, agreed as showing 4H and possibly only 3 clubs.

Another.  You start with 3145 pattern, extras, but not quite sufficient for a 2D reverse after 1C-P-1H.  (You also spurn the idea of the tendency canape 1D opening.)  Your 1NT guarantees balanced (2-3 hearts), and 2C seems unappealing.  Not good enough to reverse because of the wrong stiff, but close.  No problem!  You have already discussed with partner that a 1S call is possible with this pattern.  When partner bids something below 2D, you complete your pattern with a 2D call, again showing longer in diamonds than spades.

I'd bet there are other situations where the 5431 "second and third suit canape" approach might make sense, but this is a new thing, at least for me, that I find fascinating to develop. 

Tuesday, June 8, 2010

Rexfordized Kokish, Part III

So, let's assume, now, that you have bought all of this so far.  What would happen after all of this if you were to reintroduce back into this approach the really big, balanced hands possibility?  Or, some of them?

This gets interesting.  Suppose Opener has the really big, balanced hand and can also bid 2H with THAT hand.

If Responder hears 2H and has a 5-card major, you will recall, he either bids 2NT (spades), 3C (hearts), or 3D (both).  If Opener is able to have the really big, balanced hand for the 2H relay, and if he has at least 3-3 in the majors, then any of these calls will be easy for him to handle -- he supports the major.

Suppose that Responder bids 3H or 3S showing a stiff or void in the indicated major, 2-3 in the other major, and 4+ in each minor, with values?  Again, if Opener has a hand with at least 3-3 in the majors, he should be well-placed to choose between 3NT and four of a minor, or 4NT.

Let's suppose Responder, instead, bids 2S waiting.  Opener now bids 2NT, even with the really big balanced hands.  If Responder has both four-card majors, he bids 3D.  If Opener does not himself have a 4-card major, which only occurs with 3-3 in the majors and a really big balanced hand, Opener bids 3NT, and all seems well, again.  If Opener has a four-card major (4-3), he bids his major and all is well.  With 4-4 in the majors, Opener could perhaps bid 4C or 4D or 4H of whatever makes sense.  Again, no problems.

Let's suppose Responder, instead, just bids 3C, asking Opener what he has.  Opener can now bid 3NT to show the really big balanced hand.  This creates one problem -- Responder could have one four-card major but not both, and Opener might have no 4-card major, one four-card major, or both four-card majors.

That's not great.  But, at least we are getting somewhere, and on many sequences we are doing great.  In this instance, sometimes we would play 3NT.  But, Responder has the option of a 4C call as Stayman when he can afford the next level to check on the major.  If Opener denies a major (bid 4D), Responder now can bid 4H or 4S as a minor flag, which would be nice, as well.

So, one might reintroduce some of the really strong balanced hands back into "Rexfordized Kokish," limited to maybe 24-25 HCP, with 3-4 in each major.  With a 5-card major, 2-2 in the majors, or 4-2 in the majors, Opener would rebid 2NT directly (2C-P-2D-P-2NT), as with 26+ hands.

Of course, one might decide to include some other balanced hands that do not wualify into this structure.  For instance, Opener might treat a hand with 4-2 in the majors as a major-MINOR canape with, say, 2425, 2452, 4225, or 4252.  He might even treat 4243, 4234, 2434, or 2443 as a "major-MINOR canape" if that seems right and manageable.  The auctions where this would create a problem would be when Responder shows a 5-card suit in the short major (bidding 2NT or 3C directly after Opener's 2H call), which seems to be not all that bad.

Monday, June 7, 2010

Rexfordized Kokish, Part II

So, back up.  Opener starts 2C and Responder bids 2D.  Opener now bids 2H.  If Responder has a 5-card major, he can show it immediately, in case Opener has 4-3 or 3-4 in the majors and a long minor, or a 5+ hearts and unbalanced hand, with three spades.

We save space, however, to give room.

With 5-5 or better in the majors, a major fit exists, one way or another.  So, Responder bids 3D.  Whatever Opener's options, he has a major of at least 4-card length.  Opener simply sets trumps, and cuebidding follows.  Maybe Responder bids his stiff, though.

With five cards in hearts, and 0-4 spades, Responder bids 3C.  When Opener has the hearts-and-unbalanced hands, we have a fit.  With the heart-minor canape, we have a fit.  With the spade-minor canape and a 3-card heart fragment, we have a fit.  In all three circumstances, and to be consistent with what follows, Opener bids 3D to artificially agree hearts, and cuebidding follows, or Opener could blast 4C or 4D to show maybe the heart-minor canape and a "picture bid."  With no heart fit, Opener has spades and a longer minor with 0-2 hearts; in this event, he flags his long minor (3H = clubs, 3S = diamonds).

Why the flags?  Consistency.  For, if Responder has five spades (and 0-4 hearts), he bids 2NT.  If Opener has the heart-minor canape with 0-2 hearts, he flags the minor (3H for clubs, 3S for diamonds).  With the unbalanced hands with 5+ hearts (and 0-2 spades), Opener bids his longer minor (3C or 3D), but 3C might be bid with 2623 (Responder bids 3D to express interest in clubs, Opener bidding 3S to confirm real clubs).  With spade support, Opener bids 3NT or something at the 4-level (maybe canape completions).

Responder calso has two other main rejects of the relay -- 3H or 3S to show a stiff in the other major, two or three cards in the bid major, and 4+ in each minor, with promising values.  The reason why this shows shortness in the other major is to allow Opener to focus hearts after the 3H call:

2C-2D
2H!-3H(short spades, 2-3 hearts)
3S(I have five hearts)...
(or 3NT = hearts agreed)

So, in summary so far:

Opener bids 2C
Responder bids 2D
Opener bids 2H (hearts and unbalanced, OR four-card major with 5+ minor)
Responder bids:

1. 2S waiting
2. 2NT with 5 spades, 0-4 hearts
3. 3C with 5 hearts, 0-4 spades
4. 3D with 5-5 majors
5. 3H with 0-1 spades, 2-3 hearts, 4+ in each minor, values
6. 3S with 0-1 hearts, 2-3 spades, 4+ in each minor, values

If Responder bids 2S waiting, Opener bids:

a. 2NT = 4-card major with longher minor
b. 3C or 3D = natural with 5+ hearts
c. 3H = six hearts
d. 3S = 6H/4S
e. 3NT = 5H/4S

If Opener bids 2NT, Responder bids:

I. 3C asking
II. 3D with 4-4 majors

If Responder bids 3C asking, opener bids:

i. 3D with diamonds
ii. 3H with hearts and longer clubs
iii. 3S with spades and longer clubs

A Different Spin on "Kokish"

Or, call this "New Frontiers Light."  Suppose we decided to change the Kokish Transfer to handle problem patterns rather than a problem range for balanced hands.  Could this be effective?  I think so.

The basic idea is this.  When Opener starts 2C, hears 2D (waiting or GF waiting), and then bids 2H, this would instead show either the classic Kokish half (rebids above 2NT show unbalanced hands with hearts) or a new meaning for 2NT -- minor canapes.  Thus:

2C-2D
2H-2S
3-bid = unbalanced with 5+ hearts

2C-2D
2H-2S
2NT = 4-card major, plus a 5+ minor

There is more to this that I have worked out, which I will share later.  But, the basic idea is that Opener's 2NT shows a hand with four of either major and 5+ in either minor, problem hands that have caused difficulty for years. 

The most frequent unwind is for Responder to just bid 3C to find out what Opener has.  If Opener has diamonds (with either four-card major), Opener bids 3D.  Responder can then bid 3H to check on a 4-4 heart fit or 3S to check on a 4-4 spade fit.  If Opener, instead, has a hand with clubs and a 4-card major, he just bids the major.  So:

2C-2D
2H-2S
2NT-3C
???

3D = diamonds, plus a major
3H = clubs plus hearts
3S = clubs plus spades

So far, so good.  The second most common unwind for Responder is a 3D reply, showing 4-4 in the majors.  When that happens, Opener just sets trumps.

2C-2D
2H-2S
2NT-3D
???

3H = hearts agreed, I have a long minor
3S = spades agreed, I have a long minor

There is more to this, including what Responder does with a 5-card major, what Opener's direct 3C and 3D bids mean (no 4-card major), some special treatments, and another hand type that can be handled through this "Rexford-ed Kokish."  That will come next.