Sunday, August 7, 2011

Event 3 - Match 1 - Board 1

For those of you who are just joining us, a word or two about the Gargoyle Chronicles. These articles discuss matches I play against the computer program Jack, the winner of seven out of the last ten World Computer Bridge Championhips. The deals are random, and I report on each one. Accordingly, the deals fairly represent the types of problems one actually faces at the table. Bridge books tend to over-represent the spectacular and gloss over bread-and-butter decisions. I believe this tendency leaves a gap in bridge literature. There is a good deal to learn from the most quotidian hands. How do you figure out what is going on? How do you try to conceal what’s going on to the opponents? How do you try to clarify the position to partner? The ability to answer these questions separates the expert from the intermediate player more than the ability to execute a backwash squeeze.

By sheer chance, the spectacular will appear in these articles from time to time. But, most of the time, I simply focus on the questions I ask myself and how I reach each decision. The emphasis is on process, not on results. I am more concerned with how I make each decision than whether it worked or not. As in real life, a decision I take great pains to make frequently turns out not to matter. Or it turns out be wrong, hopefully because of bad luck and not because I made a mistake.

Speaking of mistakes, if I do make one, I report it and try to figure out why I made it. I don't give myself any second chances. I will confess, however, that I take considerably more time in playing these deals than I could ever afford at the table. So I do make fewer mistakes here than I do in real life. Sadly, I still occasionally make them.

I am starting the third event of this series: a round-robin team event, scored at victory points. This is the first board of the first match. There will be nine matches in the event. In this match, my opponents, Sophie and Jacinta, play Polish club. My partner and I play Acol. If you wish to play the board yourself before reading my analysis, clicking on the hyperlink below will bring up a pdb file, which you can load into a variety of bridge-playing programs.

Board 1
Neither vulnerable

♠ A Q K 9 8 2 9 8 3 ♣ A K J 10

Partner opens one diamond. Normally, one responds one heart with this pattern. But with this hand, we may be in slam territory. If partner has, say, ace doubleton of hearts, it's easy to see a four-three club fit offering the best chance for slam. So I decide to bid two clubs, representing my clubs as a five-card suit.

Partner rebids two notrump, showing 15 to 17 high-card points. We are clearly in slam territory now. I bid three hearts, and partner bids three spades.

One never knows about Jack, but this bid should show doubt about the right strain, presumably because of lack of a spade stopper. The fact that partner's high cards are outside of spades is a definite plus. We might even have a grand slam:

♠ x x x A x A K Q x x ♣ Q x x

Not that I intend to make any attempt to get to seven. I can't possibly find out enough to bid a grand intelligently. But the fact that seven is odds on opposite the right minimum convinces me that my hand is worth driving to a small slam rather than simply inviting.

Unfortunately, we've wrong-sided notrump. If partner were barred and I had to pick the final contract right now, I would guess to bid six clubs. It's hard to construct a two notrump rebid with no honor in spades or clubs. And as long as partner has the club queen, six clubs can't be a terrible spot. But I'd just as soon avoid a four-two fit if possible to minimize the chance that an opponent has more trumps than I do. While it would be nice to protect my spade tenace, it might not be necessary. Sometimes, we will be able to take twelve tricks without losing the lead.

I don't relish having a sophisticated slam auction with Jack. (No offense, Jack.) But it can't hurt to give it a try, especially if I'm content to set a modest goal for myself: finding out whether partner has three clubs or not. Choosing six clubs whenever partner has three clubs and six notrump whenever he has two is probably better than just blasting six clubs, which is my alternative. I start by bidding four clubs. I hope partner thinks this is forcing.

Partner bids four hearts, presumably suggesting a four-three heart fit. Partner doesn't yet know I'm contemplating a slam, so he's bidding under the assumption that we are searching for the best game. Now that I know he doesn't have a doubleton heart, six clubs has become less attractive. But I decide to probe further by bidding four spades. I still don't think I've shown slam interest, by the way, although Jack may not agree. Four spades, to my mind, simply asks partner to choose between four notrump and five clubs. We haven't necessarily found our best strain yet, and choice of games always takes precedence over exploring for slam. Since how partner responds to a choice-of-games inquiry often helps you in a potential slam decision, this approach seldom causes much of a hardship.

LHO passes four spades, which is a good sign. She might have doubled with the spade king. And, if she choose not to, her failure to double might put her partner off the lead. I'm a little less worried about playing six notrump now.

Partner bids four notrump. If we were on the same wavelength, this would be natural, and I would raise it to six. I know from past experience, however, that Jack plays most four notrump bids as Blackwood. He probably thinks my four spade bid was a slam try, showing the spade ace and confirming hearts as trump. I decide to humor him by bidding five clubs, showing zero or three keycards. If partner bids six hearts, I'll correct to six notrump.

Partner surprises me by bidding six clubs. Well. This was easier than I thought it would to be. Jack must be offering me a choice between six clubs and six hearts. If partner thinks six clubs might be the right spot, surely it is. I pass. LHO leads the five of spades.


NORTH
Jack
♠ K 10 4
A Q 5
A Q 10 7 2
♣ 6 2






SOUTH
Phillip
♠ A Q
K 9 8 2
9 8 3
♣ A K J 10



West North East South
Sophie Jack Jacinta Phillip
1 Pass 2 ♣
Pass 2 NT Pass 3
Pass 3 ♠1 Pass 4 ♣
Pass 4 Pass 4 ♠
Pass 4 NT2 Pass 5 ♣3
Pass 6 ♣ (All pass)
14th suit gameforcing
2Ace asking for hearts
30 or 3 aces

I guess I didn't achieve my modest goal. I don't understand three spades or six clubs. This is a pretty silly contract, although I don't see how I could sensibly have avoided it. As I said, if I had decided just to blast a slam instead of trying to have a cooperative auction, six clubs is what I would have chosen. I did try to steer us back into notrump. Other than just deciding that partner's three spade bid bore no relation to his hand, I don't see what more I could have done.

I play the spade ten (might as well find out where the jack is), East plays the deuce, and I win with the ace (the card I'm known to hold). I play the deuce of hearts--four--queen--three, then a club from dummy--four--jack--three.

 If West has four hearts, it would be a good play to duck the club queen. If I return to dummy with the heart ace and repeat the club finesse, she can give her partner a ruff. But it would be dangerous to duck with queen doubleton or third. For all she knows, repeating the finesse isn't an option, since she doesn't know I have the club ten. And she would be unlikely to duck with queen fourth, since she doesn't expect her partner to have a third trump to ruff with. I think I'm safe in assuming the club queen is onside, which means I have eleven tricks.

I have lots of prospects for a twelfth. I can play a diamond to the queen, finding the king onside, or to the ten, finding the jack onside. If that fails, I need three-three hearts or the other diamond honor dropping doubleton or a red-suit squeeze. To preserve the squeeze chances, however, I must play diamonds now. If I play a heart to the ace, repeat the club finesse, then take a losing diamond finesse, a heart return will break up the squeeze.

The opponents' convention card says they do not signal on declarer's leads. I know from past experience with Jack that this does not mean they card randomly (as it should) but that they always play up the line. Against such opponents, the normal rules for falsecarding don't apply. Instead, I must also play up the line. Each opponent will know that her partner doesn't have a card lower than the one she plays, so I will be marked with any low cards I don't play. Accordingly, I play the three of diamonds. West plays the five. Should I finesse the queen or the ten? The king and jack are equally likely to be onside, but the queen has the advantage of being more likely to win the trick. This could be important for two reasons: (1) West might have a singleton heart. (2) Clubs might be five-two. If East, for example, has

♠ x x x x x x J x ♣ Q x x x x

I can still make this if I play the diamond queen.

I play the queen, and East follows with the four. I play another club--seven--ten--five, then cash the club ace--eight--diamond deuce--club nine. I've made six. If I can bring home the hearts or catch someone in a red-suit squeeze, I'll make an overtrick. Neither comes to pass, however. I wind up making only six.


NORTH
Jack
♠ K 10 4
A Q 5
A Q 10 7 2
♣ 6 2


WEST
Sophie
♠ J 9 7 6 5
6 4
K J 5
♣ 8 5 3


EAST
Jacinta
♠ 8 3 2
J 10 7 3
6 4
♣ Q 9 7 4


SOUTH
Phillip
♠ A Q
K 9 8 2
9 8 3
♣ A K J 10


Not surprisingly, our opponents are in six notrump making seven (no reason not to play a diamond to the ten in notrump), so we lose three imps.

Table 1: +920
Table 2: -1020

Result on Board 1: -3 imps
Total: -3 imps

1 comment: