As Time Goes By

My year is marked with two ritualistic viewings of black-and-white movies from my youth in which cynical hard cases are transformed overnight into idealistic good guys: Scrooge at Christmas and Casablanca on the fourth of July. It is one of those strokes of accidental genius for which Hollywood is famous that Dooley Wilson’s quintessential version of As Time Goes By got into the film and stayed there despite the fact that the actor could neither sing nor play the piano. I remained infused with an appropriately melancholic spirit of resignation after belatedly reading Eric Kokish’s comments on Board 5 of this year’s World Wide Bridge Contest. Once more we were faced by the dilemma of how to make the best out of a balanced hand that contained 17-19 HCP after being forced by the system to open a natural 1♣ . Truly, Sam was right when he sang that the fundamental things apply as time goes by. A year ago I had blogged about this problem suggesting a wide-ranging reverse of 2♦ that included the NT possibility, and recently in the USWBC we came across a pair, JoAnna Stansby and Migry Zur Campanile using a system in which 1♦ is used as a relay to 1♥ to cover a 1♣ opening bid that may be a balanced 17-19 HCP. So the problem has not gone away and it’s still the same old story. Let’s see what Kokish recommends in an uncontested natural auction.

North	South	North	South
♠97	♠AK64	Pass	1♣
♥ AJ92	♥ Q53	1♥	2NT
♦ 875	♦ K9	3NT	Pass
♣ K953	♣ AQ86

Koach, as he is known, comments ruefully that there is no ideal solution. He notes his approach may miss a 4-4 spade fit, but he prefers it to the more revealing auction that proceeds with total honesty: 1♣ – 1♥; 1♠- 2♣ ; 2NT – 3NT, or more surprisingly to his fans, 1♣ – 1♥; 1♠- 2♣ ; 2♦ * an artificial forcing bid seeking more information. He doesn’t comment on what might follow a jump to 2NT – would 3♦ be a transfer back to hearts? It seems as if the professor has resigned himself to the fact that in the real world perfection, like a rainbow, is both elusive and transitory. Bidding problems only coincidentally have ideal solutions – most of the time a compromise is required.

It is best to think of constructive bidding as an exercise in constrained optimization – optimizing the gain got by increasing the information content of the auction weighed against the loss engendered through guiding the opponents to their best defence. If we examine the above hands we see a division of sides of 6=7=5=8 with a total of 26 HCP. On that basis we may conclude the optimum contract is 3NT played by South, and that the sequence that gets NS to that destination is optimum if it increases the chances of a spade lead and decreases the chance of a diamond lead. The Kokish sequence does that. A more revealing auction in which South bids his 4-card spade suit will decrease the chances of a favorable spade lead, thus it is not optimal.

Of course, initially the players have only a partial picture, so they fill in the details as they proceed. As a passed hand North might decide to respond 1NT to 1♣ to limit his high-card content and show the balanced nature of his hand. This leaves a blank space, but it would be a reasonable approach if 1♣ were artificial and strong, and 8 HCP constituted a game forcing holding. Better still if under those circumstances North were to able bid an artificial 2♦ to show exactly that type of hand, leaving it to South to decide how to proceed. Rather than explore a major suit fit, South might just bid 3NT straight up, judging on the basis of probability that this might well maximize his gain at minimal cost.

This illustrates that it is the nature of the information as well as its gross amount that must be considered in the optimization process. It is not full disclosure, but useful information that we seek, information that guides us to the solution that has the greatest probability of success. There is risk involved by being selective and keeping partner in the dark, but full disclosure also involves risk, especially, as here, one is likely to end up in 3NT regardless. South, holding the strong hand, is in the better position to judge.

Next, we look at the results obtained when the deal was played at 4730 tables world-wide. From these we can see what would have been the result of various bidding sequences, for example, what would been the cost of stopping in a part score when holding a combined 26 HCP? Surprisingly, missing game might not be as costly as one expects.

Score	Frequency	Cumulative	Score	Frequency	Cumulative
630	23%	100%	140	3%	44%
620	6%	77%	130	4%	41%
600	8%	70%	120	5%	37%
180	6%	62%	-100	29%	32%
170	6%	57%	-200	4%	4%
150	7%	50%

From this table we gather that the most frequent contract was 3NT, reached about 60% of the time. (The score 600 is ambiguous as it could have resulted from a contract of 5♣ .) If declarer made 630, not difficult after a spade lead, he would score 87% of the matchpoints, whereas if he went down 1, he would score around 18%. The contract was defeated on more than 50% of the cases, so on aggregate it would appear not worthwhile to bid the game as the cards lay. At IMPs the aggregate when vulnerable is more favorable. However, as we have indicated above, making or going down is not merely a matter of a flipping a coin, it more a matter of loading the dice – making 630 means declarer did not get a diamond lead, and that happy situation depends greatly on the bidding sequence chosen. To paraphrase Shakespeare, if ‘tis to be done, ‘tis best done quickly – without mentioning spades.

Auction	I	Auction	II
North	South	North	South
Pass	1♣	Pass	1♣
1♥	1♠	1♥	1♠
2♣	2♥	2♣	2♥
Pass		3♣	Pass

Let’s consider the effect of a broken sequence where in an uncontested auction the partnership does not come to grips with the fact that they should normally reach 3NT on a combined 26 HCP. Suppose the bidding proceeded as in the following 2 sequences

In the first sequence an overly subtle South was disappointed when North passed as he felt his 2♥ was an encouraging move when, if minimum, he could have passed 2♣ in a known fit. North made 170 and was rewarded with a score of 53%. In the second sequence North perceived a weakness in diamonds, so chose the safest part score, making 150 for a matchpoint score of 48%. Despite the weak efforts both contracts scored close to average.

Some NS pairs might congratulate themselves on avoiding a 3NT contract that ‘should be defeated’. Their thinking is directed towards minimizing the loss if they happen to make the wrong decision by overbidding. Certainly, neither 4♥ on a 4-3 fit nor 5♣ on an 8-card fit appeals to many players, so it becomes largely a matter of 3NT or a part score.

The Contested Auction

The art of competitive bidding lies in withholding information from the opponents while welcoming information coming from their direction. Very often the opposition’s contributions, if accurately evaluated, given some credence if not afforded full credibility, help greatly in determining the optimum contract. Light actions are especially useful. Here is the full deal with a hypothetical construction of EW actions.

Dlr: North		North
NS Vul		♠97
		♥ AJ92
		♦ 875
		♣ K953
	West		East
	♠ QJT85		♠32
	♥ 876		♥ KT4
	♦ T63		♦ AQJ42
	♣ T4		♣ J72
		South
		♠ AK64
		♥ Q53
		♦ K9
		♣ AQ86

West	North	East	South
—	Pass	1♦	Dbl
1♠	2♥	Pass	2NT
Pass	3♣	Pass	3♥
Pass	4♥	All Pass

East is happy for a change to be able to open the bidding in a minor suit in which he holds full values. South doubles planning to bid NT later – with better diamonds he might bid 1NT immediately, but with good spades he can afford to await developments. West makes a silly bid on the belief that he who holds the spades rules the world. South shows extra strength by bidding 2NT and North has full values under the circumstances. When South belatedly shows tolerance for hearts, North is able to raise himself to game on the decent quality of his trump suit and his lack of values in the suits advertised by the opposition. With secondary values in diamonds he would be more cautious. He is prepared to believe East has an opening bid in which case West is exposed as a fraud.

In competition a player should believe his partner and suspect the opposition. It is very helpful if partner has a bid that promises values in a balanced hand. The takeout double serves nicely in that capacity. In the uncontested auction South had a problem separating the tasks of showing a strong balanced hand and showing a hand that could play in spades. A double is more flexible than a bid of 1♠in the uncontested auction. West’s silly bid simplifies the NS auction by eliminating the spade suit from consideration, but if North is put off by this nonsense NS may stop short of game, nevertheless, if NS play even in a partial heart contract they will score above average thanks to this dubious action.

If East’s opening 1♦ bid is of the nebulous variety and South ends up declaring in 3NT, West may lead a spade in the mistaken belief that it is safe. With little hope of developing tricks in the spade suit and no entry to cash them if they were to be happily supported, the most likely result is that East will eventually find himself endplayed after an asute declarer has obtained a full count of the hand and exits with a diamond. That may account for some of the 600 scores. As Kokish points out, the ♦ T is a valuable entry to the West hand which could enable East to avoid an endplay in the heart suit, so leading the ♦ T at the start, as some would, could prove disastrous. The ♦3 is best.

One further point: if West boldly jumps to 2♠ , can NS possibly extract a penalty? An examination of the results indicates few pairs were able to extract penalties, but when they did, they scored very well. Even +300 scores above 62%. All that is needed is for North to double 2♠competitively, showing a balanced hand, and for South to leave it in. Easy to see after, but at the table most players bid on in the hope of making game, so West goes unpunished and can continue to act with impunity. In this ‘let’s pretend it didn’t happen’ mode, NS are merely attempting to return to the normal contract of 3NT. South can see the danger in 3NT as the diamond situation is fragile, so should be inclined to pass for penalty in hopes of a good score no matter how many undertricks are realized.

That argument may not convince some who may ask, ‘what is North to bid if he has an unbalanced hand with hearts and clubs?’ The answer is 2NT, for takeout. This is logical, for with a balanced hand and the values to bid a natural 2NT, it is better to double and give partner the option of converting to penalty. For distributional hands with game ambitions, North has a choice of 2 cuebids, otherwise, with no defence, he just bids his best suit. That’s a fine collection of bidding tools based on a balanced double.

We conclude this one deal encompasses many of the features of tournament bridge. Fate was kind to some bold ones who took their chances as well as to some undistinguished others who were awarded an average plus for merely muddling through. It rewarded skill in declarers who devised a clever endplay when the opportunity presented itself. A few lucky ones get on the night plane to Lisbon, while most are left behind in Casablanca. Prominent among the innocent victims were the Wests whose partners wouldn’t open in a minor suit on less than 12 HCPs, and the Easts whose partners had a tin ear and led the ♠ Q, regardless. Yes, it’s still the same old story – so shuffle, deal, and play ‘em again, Sam.