Science and Uncertainty

In the 19th century many scientists sought to emulate Isaac Newton in the role of God’s messenger transmitting to mankind the immutable laws of the universe. Today’s view of Mankind is that we are observers whose incomplete knowledge is limited by the accuracy with which we can measure. As measurements become more refined, scientists have to be prepared to abandon cherished principles and move on to the next stage with the realization that there is no end in sight and uncertainty will always prevail.

Within the bridge world scientific bidding is synonymous with the extraction of precise information that reduces uncertainty. Thick books have been written on how to get the most information out of an auction that starts with a 1NT opening bid. Bidding contests promote the idea that bidding sequences can be devised to extract the essential features of a deal that determine desirability of reaching a particular contract. The basic idea is that the more information extracted, the better the chances of success. This is not entirely true.

In practice we know that the success of a contract depends upon the actions of the defenders – the less they know, the better for declarer’s chances. The aim of bidding is not to achieve the fullest disclosure, but to maximize the probability of scoring well. Normally, at IMPs especially, this translates into bidding a high scoring contract and making it with some help from the defenders. One wants to exchange information in order to judge accurately the efficacy of a high-scoring contract while keeping hidden information that may aid the defense. The trick is to achieve the best balance between the conflicting requirements. Recently in a match between Norway and Sweden a deal occurred that shows how the most sophisticated pairs seek to make use of uncertainty.

Dealer: East Vul: Both	Bertheau ♠ KT2 ♥ K53 ♦ T942 ♣ AQ
Groetheim ♠ Q ♥ AT ♦ AQ76 ♣ KJ6532		Tundal ♠ AJ63 ♥ 96 ♦ KJ53 ♣ 874
	Nystrom ♠ 9854 ♥ QJ8742 ♦ 8 ♣ T9

Here we see Glenn Groetheim, devisor of one of the most advanced relay bidding systems, jumping to 3NT with an inappropriate distribution, just like the Galloping Grannies at my local club. The difference is that an expert partnership has several options available that can be safely pursued without fear of an accident, 2♣ would have been a relay trigger, yet the expert chooses to blast away to the most likely making game hoping to have the timing to set up the winning tricks in clubs. One may surmise that he has chosen to maximize the Conceal-to-Reveal ratio in order to minimize the chance of an immediate killing lead. He may have anticipated a non-damaging spade lead, but Nystrom had the natural lead of the ♥7, and 3NT was easily down 3 for a loss of 14 IMPs.

Lest we think it was wrong to insist on reaching game on this combination, the Swedish EW pair at the other table bid and made the ‘impossible’ 5♣. Johan Upmark opened the bidding with a Strong Club and in a series of 5 relays Per-Ola Cullin provided the information that pointed to the game that can be defeated on a diamond lead, which would have been obvious if EW had revealed their 4-4 fit in that suit. The lead was the ♠7 which ran to the ♠Q. Upmark played a low club from hand, and the timing was right for discarding a heart loser on the ♠A. This shows one more time that full disclosure is not the way to go if one aims to maximize the gain. To be fair, a mundane result of +110 in a minor partial would have gained a useful 9 IMPs for the Swedes.

A week later in Monaco Bertheau and Nystrom got to bid a slam using their own relay methods against another famous Norwegian pair. The question here is what information does one include in the definition of one’s bids? To qualify as scientific, a bid should be defined exactly in numbers, not vaguely as an attitude, as follows at the other table.

Dealer: South Vul: Both	Ventin ♠ Q2 ♥ 8732 ♦ 76532 ♣ J4
Zimmermann ♠ AK6 ♥ A5 ♦ AT9 ♣ K8652		Multon ♠ 943 ♥ K964 ♦ KJ ♣ AQT7
	Wrang ♠ JT875 ♥ QJT ♦ Q84 ♣ 93

West has a strong 5-3-3-2 hand, the kind that gives me so much frustration when I am playing 2/1 game forcing at the local club. It pains me to open a nebulous 1♣ and jump to 2NT for all the world looking like a guy who hopes partner can raise to 3NT on 6 HCPs. My expectations of being raised to 4 clubs is practically non-existent, so I must settle for +690 and an average board. I have suggested starting with a forcing 1♦ response, but the above result from a top European pair shows that may get you one level higher to no good effect in the end. The problem is not so much a lack of bidding space, one can bid safely up to 5NT, as a poor definition of the bids, hence what information they convey. Even if one starts with game forcing response by East to a Big Club opening bid by West, there remains the question if what information is most relevant: responder’s distribution, of course, to establish a 9-card fit exists, but also the playing strength of responder’s holding with regard to a club slam. Controls are of prime importance.

When I am dealt a hand with 8 controls (A=2, K=1) I am immediately alerted to the slam possibility it represents. Zimmermann had a chance when Multon jumped raised to 3NT showing decent values, if he could have bid 4♣ naturally and still managed to stop in 4NT if partner primarily held diamonds. This was IMP scoring, after all. The result shows that the 4-3-2-1 scale is not sufficient, as 12 tricks are easy even on a mere 31 HCP. (In fact, 7♣ makes without a finesse as South’s heart honors can be ruffed out.)

The aforementioned Swedish pair reached 6♣ in a straightforward manner, albeit in an auction that went on for 9 rounds. The first round was critical: Bertheau began with a Strong Club and Nystrom responded 1♥ to show 8 or more “zz points”, where A=3, K=2, and Q=1. This evaluation is well suited to distributional hands where controls are more important than slow stoppers in a short suit. No one can convince me that 3 jacks are the equivalent of a king. Bertheau knew after the first response that his partner held the ♣A, a red king, and possibly 3 queens or a second red king and 1 queen. It hardly seems efficient to have to take 7 more rounds to sort it out, but fun is fun. I’m sure the boys put in lots of time preparing for this, so let’s not begrudge them their moment of triumph.

Bidding according to the total number of HCPs is quantitative in nature, but it is not scientific, as it is not a sufficient description. The other factor at work is distribution. The value of isolated queens and jacks lies largely in their relationship with suits held by one’s partner. If a partnership withholds distributional information, then it removes the opportunity to evaluate accurately, and they must fall back on bidding by formula to what is most likely on average. Here is a perfect example from the Hainan Air Cup 2011.  Using Precision methods one pair bid to 6NT on 34 HCPs, the other to 7♦. Let’s put it in a hypothetical American 2/1 context.

The doubleton black queens motivate West to open 1NT (15-17 HCP). He has 5 controls which puts him comfortably at the top of the range. I prefer the range 14-16 HCP, as all too often 17 HCP puts one in the slam range. Responder has 6 controls, the equivalent of 20 HCPs. However, it is uncertain how much weight to put on the 3 jacks and the 2 tens. One can find out only by getting out of the NT mode and start bidding suits. East is not quite sure how to do that. If he does go through a procedure, how sure can he be that he will reach the correct contract? When East opts for a quantitative 4NT, West is happy to accept the invitation. He might consider it unlucky that the hands fit so well, because with 34 HCP in total, 6NT is most often the right place to stop.

West may decide to open 1♦ planning to reverse to 2♥ after the expected 1♠ response. When partner bids 2♣, West has to decide whether to enter the NT mode or to bid 2♥ to show a second suit, even though it is likely that there is no 4-4 heart fit. To bid 2♥ is to guide the defender towards a spade opening lead. When facing a close decision in a forcing auction I prefer to make the cheaper bid regardless and await developments. Here the decision is not even close as 13 of the 17 HCPs lie in the red suits. Here is how the bidding might proceed, in the manner of the Chinese pair involved. The key is the establishment of a diamond fit.

Of course, the bidding is not perfect, as East (Gao Fei) could correct to 7NT. This shows that it is not easy psychologically to switch back and forth between NT mode and suit contract mode. The disclosure of the diamond suit enabled East to give added value to ♦KJT, while the disclosure of the club suit enabled West to add value for ♣Qx. The presence of the ♠Q and ♥J was largely irrelevant.

Note that it doesn’t help greatly for West to open a Big Club. He has just 5 controls, not enough to be able comfortably to take charge of the auction. Usually 6 controls are required, 7 desired. It is East with his 6 controls who can offer the best evaluation, so cooperative bidding is indicated when the high card values are balanced and neither hand has shortage.