Cryptic Controls

In a previous blog we discussed an opener’s reaction to a 2NT Jacoby raise and suggested that 3♣ should be defined as an asking bid. The subsequent auction can be geared towards hiding as far as possible the holding of the eventual declarer. Potentially the greatest advantage over natural bidding comes when neither partner has enough strength to insist on slam even though together they hold 10 controls. If opener has 5 controls and responder has 5 controls, there are missing either an ace or 2 kings. Here is the asking bid structure when responder bids 3♦ to show 5 controls, either AAK or AKKK.

If opener holds 2 aces he knows his side holds all the aces, so it is a matter of the missing kings. If opener has 3 kings and just one ace, there is an ace missing, and he can find out which one by using the 3♠ asking bid. The CRS responses show 2 aces of the same colour (4♣), rank (4♦), or shape (4♥). Opener will know which ace is missing, but one defender will not, and that defender could be on opening lead, as in the following deal where declarer manages to make slam of 15 HCP opposite 12.

The bidding is crude, but opener knows the red kings are missing. The trumps will come home 50% of the time. Any further exploration may be counterproductive, decreasing the chances of making 12 tricks. A diamond lead would be inspired, but if one were led, the hearts must come home, after which declarer has a choice of how to play the spades to gain a diamond pitch. The computer lead was the ♣J, the ♥K was onside, but the spade finesse failed. No matter, the diamond switch came too late, and 12 tricks were taken.

On a double dummy basis one might argue that this is a bad slam that just happens to make, but this is not a good way to think about the contract, as the success of the contract depends on the opening lead. It is improbable that the killing lead will be found on every occasion, and the less information transmitted, the greater the probability that it will not be found. For example, on the above deal an opening lead from a diamond honour might be considered too dangerous.

When slam is bid, declarer doesn’t know which intermediate cards, if any, are held opposite. There are several possibilities that can help, for example, if responder holds the ♦Q, declarer is protected against a diamond lead. If responder holds the ♣Q, there is no club loser, and the slam is at worst on the trumps coming home. So, there are situations where declarer would like to ask about queens. This is more often the case when an ace is missing. Here is a case where the trump ace is a sure loser, so declarer needs the timing to avoid a second loser.

When responder shows the minor suit aces, opener knows he has a pitch for the third diamond, so it is largely a question of the strength of the trump suit. If 5♥ can be interpreted as asking for the ♥Q, responder will raise to slam.

Distributional Probabilities
Opener does not know partner’s distribution in detail, but he may assume a 4-3-3-3 shape with 4 trumps. This is the single most likely shape, and the most conservative with regard to hand evaluation as ruffs in the dummy will be hard to come by. In that situation side-suit queens are prime assets. If 4-4-3-2, the side suits are most probably aligned with the 2-card suit opposite opener’s longest side suit and the 3-card suit opposite declarer’s shortest. This is a happy situation as ruffs may be available in both hands. In addition opener may be able to arrange for an elimination and endplay.

As the bidding proceeds and controls are shown, the probabilities associated with the length of suits chance. It is likely that revealed controls come from the longer suits. For example, an AK is more likely to come from a 4-card suit than from a doubleton. This is the companion to the standard situation where a player who shows a long suit is more probable to hold honours in the suit than not.

To illustrate this point we look at a computer hand where opener held a 3=5=4=1 and responder showed the AK of clubs. It is reasonable to assume the controls shown came from a 4-card suit.

Before the 3♣ asking bid the single most likely distribution in responder’s hand is 3=4=3=3, but nearly as likely is 3=4=2=4, which would allow for diamond ruffs in the dummy. However, the bidding changes the expectations. From the bidding opener can place 3 cards in partner’s hand (♣AK, ♠ Q) and 3 cards in the defenders’ hands (♠ K, ♦Q, ♣Q), so the remaining cards may be placed in the current vacant places as follows.

The former 3=4=2=4 (long suit – short suit match) has been transformed into the most even split (2=2=2) and is now twice as likely as the former 3=4=3=3. Opener can be hopeful of being able to avoid the diamond finesse, losing just 1 spade trick, and he just might get a diamond lead.

It  is disappointing when dummy emerges with 2=4=4=3 shape, the odds being nearly 6:1 against that shape relative to 3=4=2=4. A passive trump is led and declarer must decide whether there is a better play than the 50% diamond finesse. Well, there is. After drawing trumps and discarding a spade on a top club, and ruffing a club to eliminate the suit, declarer plays off the top diamonds hoping to exit a diamond to a defender who has to lead away from the ♠ K. There is an additional chance: that the ♦Q is doubleton, which raises the chance of success well above 60%. Such was the case with the ♦Qx offside.  A spade lead followed by a diamond finesse would have resulted in defeat.

This deal illustrates that when trumps can be drawn easily, the timing is on declarer’s side. There may be more than one way to make 12 tricks and declarer must try to make use of the various possibilities. Besides which, sometimes he guesses right.

More Cryptic Controls
When responder shows 6 controls by bidding 3♠ /3♣, he holds either AKKK or AAKK.
To place the controls opener need only bid 3NT to ask if partner holds a suit that is topped by a king. 4NT is the response that says there are either none or 2 such suits. Opener will know which applies and is fully informed. With only one such suit, responder bids that suit. The answer tells all, while the defenders are kept largely in the dark. Here is an example.

Opener deduces that responder for his 6 controls must hold the ♠ A and the ♣A, therefore, 2 kings, one of which tops a suit, either the ♦K or the ♥K. When responder admits to the ♥K without the ace, opener knows the ♦K is missing. He may hope for a black queen, but the ♣Q is denied by the 5♦ response to 4NT, queen-asking, as queens are bid up-the-line. Opener signs off in 5♥ after showing slam interest, hoping responder can bid 5♠ to show a doubleton or 5NT to show the ♠ Q.  Responder should assume his partner knows where the controls lie, so there is no need to bid again when one has shown everything of interest there is to show.

In this manner slam is avoided, but if responder is in charge of the auction, as he would be in a standard Jacoby auction, he might get overly excited and use RKCB getting too high. This demonstrates the disadvantage of an embarrassment of riches in the trump suit when bidding to a slim slam – the ♠ J would be potentially more useful that the ♥J.

Bridge World Sub-Standard
It is oddly coincidental (as are so many happenings) that the Dec edition of the Bridge World’s Challenge the Champs feature presented these hands with the suggested Jacoby 2NT auction given below, where the standard bid of 3♣ shows shortage.

It is difficult for responder to evaluate upwards with 8 losers and wasted values in clubs, even after opener overbids 3♠ on a junky 3-card suit. Sadly, this is typical of standard bidding that one must lie in this manner just to keep the auction alive. Responder, accustomed to being poorly informed, is reluctant to proceed. Obviously the hands should be bid to 6♥, and it is possible to achieve this if the opening bidder takes control with a 3♣ asking bid on the off-chance he will hit pay dirt. The subsequent bidding works out fine.

On seeing the 4♦ bid, opener must refrain from shouting ‘Bingo!’