Sunday, October 21, 2012

Extending lead-directing doubles to suit games

I have written many times about different usages of double, especially my favorite  because it is so versatile  the cooperative double.  And I spent some time discussing various types of penalty double, in two parts (there is a third part in draft it seems that never was published).  Penalty Doubles (part 1) covered, among other types, lead-directing doubles.  My overall conclusion is that you won't get rich playing penalty doubles simply to increase your score because you think they are going down, unless you are preserving equity.  So, as Theodore Lightner realized many years ago, there's more profit to be had from suggesting the killing lead than there is in increasing the penalty when they've overbid slightly.  Most people understand that a double of a slam asks for an unusual lead.  And good pairs have discussed what the double of a 3NT contract asks for.  But I haven't read much on how to extend these principles to suit games.

I've also argued previously that cooperative doubles can lead to the juiciest of penalties  because such doubles tend to be made when the high card assets of the defenders and the trump length assets are separated in the two hands over the hapless declarer, who may not even realize that he was overbidding, is outflanked.

So, is it ever right to double because you think they have made an error or are running into bad breaks? Yes  when one of the penalty double triggers has occurred and it's already been established that our side has the makings of pincer movement as described above.  How about this auction: you open 1, LHO overcalls 1 and your partner bids 1NT.  According to my previously published rules (DSIP Rule Summary), penalty doubles are now in effect.  Both sides seem to have plenty to say and, after a competitive auction, we end up bidding 4 with no great certainty of making.  After two passes, RHO bids 4 and we double.  Our hand might be ♠94 AKJT52 QJ8 ♣K6.  Partner's original 1NT call assures us that they don't have a nine-card fit  and we have at least 22 hcp.

But supposing the auction goes thus (we are vulnerable, they are not):  pass (1) 2 (3) pass (3) pass (4) X all pass.  Your hand, by the way, is ♠K3 9876 AKJ94 ♣T5.  Partner is sitting under the spade length, he has passed twice, we have shown a decent hand and suit but have not shown any extras.  Can it possibly be that partner expects to defeat this on power alone?  And he thinks he can get it two tricks?   No, that's not possible.  The double must be lead-directing.  But this is not a slam.  Even if he has a surprise or two (or a void) in clubs, this isn't one of those situations where we get a quick trick on a club lead and then cash an ace for down 1.  No, we have to generate four tricks for our side!

This situation is more akin to trying to defeat a notrump game. The effect of the 2 overcall has raised the level of the auction to a somewhat uncomfortable level and the 3 call, while obviously showing good cards, is looking for a diamond stopper which opener doesn't have.  These folks sound like they might even be on a seven card fit. Surely the only possible interpretation of the double is that partner has some useful trumps and that our advantage of having the lead, if used wisely, will defeat the game.  So, laying down a high diamond is obviously not the solution because if partner wanted us to do that, he would not double.  Could he have a void in hearts or clubs?  Somewhat unlikely given that he never raised diamonds.  And whatever cards he holds in dummy's suit(s), likely clubs, can surely wait their turn.

What would a double mean if declarer had ended in 3NT?  It would (by our agreements, at least) say, don't worry about the fact that you have a hole in your suit  I've got it filled.  In other words, double suggests possession of a high honor in opening leader's suit.  Without the double, leader might try to "find" partner's "ace" for a lead back through declarer.  The double says "I don't have the ace, but you can low because our suit is reasonably solid and we will set the contract."

So what about in this contract, could it be that partner wants a low diamond to the Q (or perhaps even a void)?  Yes, that's got to be it.  Now, declarer runs out of trumps trying to stop diamonds and ends up going down one.  A high diamond lead let's him cruise to ten tricks.

Yes, I was partner and made the double.  And, yes, I had Qx of diamonds and four spades to the J.  Opening leader was not on the same wavelength (and I'm sure my readers will have quite a bit of sympathy with him, as do I).  So we lost 5 imps (instead of gaining 10) but fortunately got two big swings on other boards, not only to win this match, but to pass our opponents in the rankings and take first place at a club swiss.

Monday, October 15, 2012

Maine bridge

Occasionally, Kim and I are able to attend one of the Maine sectionals. For those of you who never play bridge far from a metropolitan area, you should give this kind of thing a try. The players are invariably friendly and over the years we have come to know many of them. The food is good and people are just, well, nice. And there's always a big welcome from Horace and Sonya who put on the tournaments. This weekend's tournament was in Bangor, about an hour's drive from Kim's mother's house.

This occasion was saddened by last week's death at 88 of Kim's aunt Helen, formerly a regular bridge player who sometimes played in the sectionals but mostly played in the non-sanctioned club game in Fairfield. Helen was more of a poker player than bridge player, though. She loved to go to Las Vegas whenever she got the chance, which was quite often. Invariably at these Maine tournaments, several people would ask after Aunt Helen when she wasn't there. She was a lot of fun to be around and we'll miss her a lot.

On Saturday, the cards were very much with us. It helps to be in control of the auction on a lot of hands. And, in particular, there were many opportunities for slam bidding in our direction and, because we are fairly regular partners and play quite a lot of gadgets, we usually do well in such circumstances. Of the 53 boards we played, we bid six slams and one was bid against us. We also missed a lay-down 6♠ on 23 hcp, but so did everyone else. It's true that one of our slams (6) didn't fare so well (down 2 for -100) but, as it turned out, 5 was always making the other way so we ended up with almost a top board. I will return to this board later.

Kim was also a demon, nay ruthless, defender, with the result that in the afternoon we went plus on 19 of 27 boards ending up with a personal best score of 74%.

Here's a slam from the evening session where our agreements were severely put to the test and, fortunately, were up to snuff.

Kim opened 1 as dealer (board one – none vulnerable) and RHO bid 1♠. My hand: ♠94 J5 AJT52 ♣KJ85. Although I dislike making a cue bid with two losers in the enemy suit, I chose to bid 2♠. LHO now bid 3♠. Kim bid 4♣ which, since she was going past 3 (our guaranteed contract in this auction) showed extras and, since it also bypassed 3NT (which I might have been able to call from my side if she had doubled), suggested at least a healthy interest in a slam. With nothing obvious to cue-bid, I had to mark time with a 4 call. Admittedly, I was minimum for my bidding so far, but my hand had improved significantly with the double fit and I had little wastage (the J). 4 was forcing of course, because Kim would not suggest playing in slam and then pass in a part-score bid. I considered 5♣ but I try to make it a rule never to cue-bid a king if I'm not going to be the declarer – it's much too easy for LHO to double for a club lead, holding AQ or something else good in clubs.

Kim now made the key call of the auction: 4NT. I admit that it took me quite a little while to interpret this, although we've discussed it many times in the past. Since 4 would have been asking for key-cards (we play "Kickback"), 4NT now does duty as showing a heart control. Knowing Kim as well as I do, she couldn't possibly be still thinking about slam if she didn't have both clubs and spades controlled. But she still didn't have quite enough to be sure twelve tricks would be there (indeed, slam can't make in our other nine-card fit, clubs). So, with my surfeit of working cards, I was able to confidently bid 6. Twelve tricks were easy on any lead, but we were the only pair of 13 to bid the slam. Kim's hand: ♠7 A92 KQ43 ♣AQ732. Note that our opponents could have bid a quasi-profitable sacrifice in 6♠ for a loss of 500 instead of 920. But in that field, they'd have scored the same zero.

Here's an amusing auction from the evening session: 4♣1 p 4p 4♠3 p 4NT4 p 55 p 6NT all pass. (1) "Namyats" showing a good hand with eight or more hearts (could be seven if solid) with typically an ace or a couple of kings on the side; (2) I think we probably have a slam; (3) I have the ace of spades; (4) how many key cards (for hearts) do you have altogether? (5) three. 6NT made exactly (on any lead) for a top shared with one other pair. Why do I think it amusing? Despite us having 11 hearts between us, neither of us ever put a heart bid on the table.

I think our only mix-up of the day was the slam that went down but which turned out to be an excellent sacrifice (mentioned above). With nobody vulnerable, I dealt myself ♠AT7 KJ873 ♣QJ63, and opened 1. LHO overcalled 2. Partner bid 3 and RHO bid 4. Since 3 took us past 3, this was clearly game forcing. Whether or not it was forcing all the way to five of a minor is not entirely clear. First, let's think about the calls that partner did not make: double which, if followed by a new suit, would be game-forcing; 2♠ which would be non-forcing (as would 3♣); 3 (simply competitive); 3♠ or 4♣ which would be fit-showing jumps showing a good fit with a good suit; 4 which would be a splinter raise of diamonds.

As with most cue-bids, there are several possible meanings for 3:
  • Partner, please bid 3NT if you have the hearts well stopped;
  • I have a diamond fit and am willing to play 4 opposite a flat minimum;
  • I have a very strong hand and right now, I'm not quite sure where we should end up so please tell me more about your hand.
After the 4 bid on my right, I reasoned that we were now forced to play at least at the five-level. Therefore, I felt that an immediate bid of 5♣ would show the weakest possible hand, offering clubs as a second place to play and fulfilling the third interpretation of the 3 call. That is therefore what I bid. 

Kim assumed that this meant I had extras (I could have passed) and so bid 6. No double was forthcoming (we had confidently bid this with no suggestion of sacrificing). The K was led and dummy turned out to be ♠QJ3 A6542 ♣KT74. After winning the first trick, LHO switched to a spade. But diamonds were 3-0 offside and I still had to lose a club for -100. As mentioned above, the opponents could make 5 (six if a club isn't led) and our blunder actually earned us 7 matchpoints out of 8. 

So, what's the proper treatment? I'm not sure. But I'm coming around to Kim's way of thinking for the following reasons:
  • she might have wanted to play 4♠ all along but didn't want to jump straight to 4♠ because that would tend to shut out the possibility of a spade slam – still, I'm not sure about this because double followed by a spade bid would have had a similar meaning albeit perhaps with fewer spades;
  • she might have wanted to bid 4NT which, I think, would be a slam try with a heart stopper all along (5NT would be pick-a-slam).
I therefore invite your comments on this subject of bids by the partner of a cue-bidder. 

However, on a day when the bridge gods are with you and you can do no wrong, even a mixup like this one can turn out to be rosy!

Wednesday, October 10, 2012

Sequences (part one)

If music be the food of love, then sequences are the life-blood of bridge. Every bridge writer remarks that defense is harder than offense, but in my opinion none have adequately described why. Yes, the declarer can see all of his "assets". But it's specifically the sequences that make the crucial difference. A defender can sometimes immediately see that one of his significant cards cannot be in sequence with his partner (the dummy holds the surrounding cards) but he can rarely be sure that partner has a fitting honor until later in the hand when there has been time for signalling. By then, it may well be too late.

Sequences are most important in the first couple of leads of a suit. After that, assuming leader is trying to develop long cards, the opponents are likely to start showing out of the suit or at least having to play their sequence-breakers and what cards are left are more likely, now, to be sequential.

The next best card combination after a sequence is a quasi-sequence, known as a tenace. That's to say we almost have a sequence but there is a card missing. A tenace is positional in normal play, in other words its effectiveness depends on the location of the missing card. However, even if the card is wrong, it may be possible to win both cards of the tenace on an endplay.

Sequences can be split between the two partnership hands and still be effective. In fact, such an arrangement can provide transportation between the hands. Tenaces can be split and retain some effectiveness but it is generally reduced. For example, AQJ opposite xx may provide three tricks if the king is right. But Axx opposite QJ will produce only two tricks even if the king is onside. Longer split tenaces fare better: A5432 opposite QJT9 will always produce five tricks if the king is onside, assuming one outside entry to the long card. Sometimes, a tenace can even kill a sequence sitting under it, for example AKT sitting over QJx.

Of course any sequence or tenace has to have a reasonable rank to be useful. Holding the 65432 in a suit (a straight flush) may look pretty but it has no value whatsoever as a sequence. When a suit has no significant sequences at any given point of the play, it may be said to be "frozen": whichever side leads the suit gives up a trick. We'll examine these more closely in part two.

On defense, the opening lead is particularly dangerous for the simple reason that you don't know which cards in your hand are part of a partnership sequence and you don't usually know where partner has a tenace sitting over the dummy. This is why having your own sequence(s) is so very important, lest you "blow up" a suit that was frozen. Leading from a safe sequence (AK, KQ, QJT, JT98) can never lose a trick in the suit directly but it can sometimes lose a tempo if there was something more urgent to do. How many times have we seen the Abbot lead from a sequence like KQJT only to have Brother Xavier remark after the hand "not the best of leads, Abbot."

For our first example we will go to another deal involving the Abbot and Brother Xavier (from Celestial Cardplay by David Bird, 2009). This time, the Abbot is, through the Good Lord's mysterious handiwork, partnering the disrespectful novice, Brother Cameron. In this hand we will see how the natural advantage of the defenders in having the opening lead is, as so often happens, nullified by the wrong choice of suit. And why does Bro. Xavier go wrong this time when he leads the obvious Q? Because the key sequence in his partnership's hands is hidden in his partner's hand.

Perhaps if Ethan had had a chance to double an artificial club bid, the story might have ended differently. But then the two club honors might have been split. There's something to be said for an attacking lead against a slam when you know dummy is coming down with a good suit but that was not clear on this auction. And to lead from an unsupported honor when you have such an "obvious" sequence in another suit, might require some good arguments in the post-mortem.

No, the reason that the slam came home, apart from a relatively lucky lie of the cards, was declarer's intrinsic advantage and the defender's corresponding disadvantage: the ability or inability to see your sequences. It's true that even a diamond lead sets the slam but for a different reason: the trump situation provides no entry to the diamonds after they have been set up (East has to be sufficiently alert to continue diamonds at trick 2).

To be continued.

Wednesday, October 3, 2012

Extending the Michaels cue-bid

Let's say the auction starts with 1 by LHO and your partner bids 3. What?? You would understand 2 as a Michaels cue-bid. Well, logically partner's bid is "Super-Michaels." But what exactly does it show? According to one online reference, this bid shows a strong two-suited hand. But is that really necessary? Many people play mini-maxi Michaels which says that if you make an unforced bid after the Michaels cuebid, you are showing a maxi. Maybe a two-suiter in the 15-18 point range. Presumably with more than that you might double first and then try to show both suits. So, to some extent the strong 5-5 hands are covered.

But what about 6-5 (or more) hands? I once had the good fortune to play with a young Australian at an NABC whose name unfortunately I have forgotten. He was one of the best players I've ever partnered. Most hands ended with a claim at about trick six regardless of whether we were on offense or defense. In a Swiss, we had played six of the seven boards and were working on the last when the director came over and said we were playing the wrong opponents. We had to play the entire round again. No problem. We finished in plenty of time!

But I digress. He taught me the importance of the sixth card in a 6-5 or better hand. Never use Michaels with 6-5(+), he said. Partner will never be able to judge what to do because of that sixth card. I've always taken that to heart. But how do you handle such hands? It isn't easy.

One possibility that leaps to mind is a jump cue-bid. Suppose you have the following hand: ♠AKT85 J QJT983 ♣J and your right-hand opponent opens 1 (all are vulnerable). You've got the magic 6-5 "come alive" hand. But what should you bid? At my table, Bob McCaw held this hand and overcalled 2. The auction continued 2 pass 4. Undaunted, Bob bid 4♠. This was the final contract (it makes 5). Our side could try a sacrifice in 5HX for -500 but their side could always bid on. For that, Bob earned 5.5 out of 7 matchpoints. While it's impossible to argue with success, Bob's bidding was certainly aggressive by any standard. Not everyone will have the stomach for that kind of sequence vulnerable.

So, how about defining the jump to 3 as this sort of hand? Especially where it may be necessary to "reverse" your suits to show their lengths properly as Bob did. As with Michaels, a notrump (3NT) call by partner would ask "which is your minor?" Or, if you feel that you just have to be able to bid 3NT naturally, you could probably make a pretty good stab in the dark to guess which is partner's unknown six-card suit. In this auction, the 3 call would force the bidding to either 3♠ or 4. That's one level lower than with what actually happened at our table. Presumably, if you had a really good hand, say ♠AKJT5 8 AQJT83 ♣9 (a four-loser hand), you would be justified in raising whichever suit partner chose.

What would you be giving up if you adopted this convention? Have you ever even discussed with your partner what this bid might mean? Admittedly, it wouldn't be coming up frequently but it would certainly get your attention if it did come up.