Jacoby 2NT – Concealment and Disclosure

Beginners love 4NT Blackwood because it asks a simple question and gives a simple answer. What a relief that is. Of course, the answers don’t guarantee the right contract will be reached. One may ask for aces, bid a slam that should go down, and make it because of the uncertainty with regard to which aces are held. Declarer isn’t aware of any deficiency, but he bids on the probability that the missing ace and king(s) are in different suits. The odds on declarer’s side are improved, when the opening leader doesn’t know, either. There are methods for locating specific aces, but the ordinary player doesn’t like to assume the extra memory load, besides which uncertainty is often advantageous.

There have been developed forms of Blackwood where the answers depend on what has gone before. RKCB is an example where the responses are geared to an agreed trump suit. Exclusion Blackwood operates below 4NT while identifying a suit of no concern. In a response structure to a Jacoby 2NT, it is possible to use all three modifications: ace identification, adaptive answering, and asking bid displacement with the aim always to retain a degree of uncertainty in the minds of the defenders, while giving enough information to the declarer to arrive at a reasonable slam others may miss. The keys are the losing trick count and the number of controls held, not the number of HCPs. The former are the more relevant measures when a 9-card fit is known to exist.

The Jacoby 2NT is a bid favoured by many, including the late Marshall Miles, because it immediately announces a good hand with 4-card support for opener’s major and no shortage. In the classic version, opener is required to bid a short suit if he has one, but that does not limit his hand. There sometimes arises the problem of whether it is better to bid a second suit descriptively rather than the short suit. Furthermore, if opener is allowed to open light with an 8-loser hand he has to let partner know how bad his opening bid was in order to avoid getting too high. With an 8-loser hand opposite an 8-loser response it may be wiser to stop in 3 of the major or 3NT. We can do better.

A Simple Scheme
Here is a scheme based on losing trick count for the rebids after 2NT.

 3♣                  6 or fewer losers without further disclosure
new suit          a help-suit slam try with a 5-loser hand
3 of the Major minimum 8-loser hand
4 of the Major 6-card suit, no slam interest, 7-losers
jump suit         singleton in a 7-loser hand
3NT                 a flat 7-loser hand

The above scheme provides opener with the following six options: 1) sign off in game with no slam interest; 2) show a minimum without insisting on game; 2) indicate that a lack of controls makes 3NT more attractive than 4 of the major; 4) show a normal opening bid based on shortage in a minor; 5) start the exploration of slam without disclosure of specific assets;  6) issue a descriptive slam invitation. One of the major advantages is that opener may open light and avoid being punished by the responder.
Monaco vs Italy
Here is an example where one of the great pairs came a cropper because opener was unable to limit his hand after 2NT. By Board 20 in the 2013 Bermuda Bowl Final, Italy led Monaco by 24 IMPs when Helgemo and Helness, who tend to open light, bid to slam going off 2 when Versace and Lauria stopped safely in game. The Norwegians use 2NT as a forcing raise, but opener doesn’t seem to be able to limit his hand immediately, so his partner was encouraged to overbid to slam on an 8-loser hand.

Helgemo was at the 4-level before he revealed his short suit. Helness expected a better hand opposite, so he pushed to slam with no wastage in diamonds. The ♥Q was onside so the heart finesse worked, the ♣K fell singleton under the ♣A, but there was too much work to do with the trump split proving to be 4-1. Down 2 was the result. Under the above scheme Helgemo would have rebid 4♦ immediately, an aggressive Helness might have overbid a desperate 4♥ on 8 losers before respecting partner’s signoff.

Italy stopped safely in 4♠ when Versace opened 1♠ and rebid 2♠/2♣ to limit his hand and curb Lauria’s enthusiasm.  He never revealed his shortage as it was unnecessary to do so in a situation where game was the limit of the hand.

A Grand Slam
In the slam zone information is golden, but it is not about HCPs. The key bits of information have to do with losers and controls, as in the following computer example.

The losing trick count indicates the hand should be played in a Grand Slam. One needs to know that the controls are in place. Once opener suggests 5 tricks in hearts, responder can take the hint with his control-rich hand.

3♣ as an Asking Bid
Devotees of the losing trick count needn’t be given more examples like the one above to convince them of how good the system works when everything falls into place. What we do need are methods to place the necessary controls. We shall demonstrate how concealment and disclosure go together in the response structure to a 3♣ asking bid. To make this work we require certain restrictions on the definition of the 2NT response: 4-card support, 12+HCP, at least 4 controls (not KKKK), not more than 6. (With more than 6 controls, responder takes charge.) The initial responses to 3♣ are in accordance with the number of controls held.

When 2 aces are shown, the responses are in order of colour (4♣), rank (4♦), and shape (4♥), so it is expected that opener can identify exactly which aces are held. As 12 HCP are required initially, responder holding only 2 aces can be expected to hold 2 queens or an equivalent JTx combination, so slam may be a live possibility still.

After a response of 3♦ opener may wish to investigate slam by locating the controls in responder’s hand without revealing his own. In the case where opener has 5 controls, he knows there are missing either 2 kings or an ace. He has two possible asking bids, the most suitable choice depending on what he can see in his own hand. If he has 1 king (and a minimum of 5 controls), he must have at least 2 aces in order to ask, so he asks for kings by bidding 3♥, the responses to which are 3NT (AAK),  or 2 kings: 4♣ (colour), 4♦ (rank) and 4♥ (shape). If he has 2 kings, he knows responder has 2 aces and a king. He may ask which 2 aces are held by bidding 3♠, the responses to which are 3NT (max), 4♣ (colour), 4♦ (rank) and 4♥ (shape). 

Miles’ Examples
These responses cater for the situation in which opener with 4 controls is seeking a slam missing an ace and a king, at best a 50% chance if opener has no shortage. If he has an undisclosed short suit, slam is still possible if there is no wastage in that suit. Note that relays do not guarantee all aces are held, so responder is reduced to answering the questions reliably. This might have angered Marshall Miles who liked giving responder freedom of choice. Here are 2 examples he proposed in  Bridge from the Top – Book I to demonstrate responder’s flexibility after opener rebids 3♠ to show a singleton spade. Here is our treatment of these examples.

4♠ asks which king is held. If the spades and diamonds are interchanged in responder’s hand, the response is 4NT and one has to stop short of slam.

The Question of Shape
Once the controls are set in place, there may still remain the question of shape. This is the reverse of natural methods that deal first with the details of shape and later with controls. Certainly responder could help in that regard, but it is rather too late in the game for descriptive bids as opener does not guarantee that all aces are held. Opener can investigate above game when seeking the necessary third round controls by bidding 4NT.

After the 3♥ response showing AKK, opener knows which kings are held, so he relays to 3♠ to ask which ace. Finding the ♥A leaves the ♦A at large, but ♦KQ are adequate safety against a diamond lead. Slam might fail if responder holds ♥Axx. It behooves opener to ask for third round control of hearts, his main worry. Opener doesn’t care whether it’s the ♥Q or a doubleton. With classic methods opener has to rebid 3♦/2NT to show his singleton, and responder is at a loss what to do next as the ♦K appears to be a wasted value within a minimum response on 12 HCPs.