My last article applied SST analysis to some Jacoby 2NT auctions. I pointed out how 6-3-2-2 and 5-4-2-2 shapes can be significantly stronger than 5-3-3-2 and explained why our responding methods for opener should differentiate them. This article will look at J2N auctions when opener has a singleton. Consider this auction. Do you know what to do next?

If you do, can you tell me? I know that:

• The SST is 4, hence 28-29 working HCP will be needed to make slam.
• I know my 13 HCP are working, so I need partner to hold around 15 working HCP to make a slam.

However, I can not tell how many working HCP partner actually has. Did 4♦ promise extras? Or could he cooperate with a minimum in case I had extras? Since I don't know, I can't be sure whether I should signoff or bid on.

This is a common problem in auctions where both hands are unlimited. Opener's 3♠ call was unlimited. If 3♠ had shown a 3-HCP range like 11-13, responder would never have bid 4♣ at all; he would know that slam is unlikely with his minimum opposite 11-13 HCP. Unfortunately, in the unlimited auction where opener might still have 19 HCP, it's a bad idea to sign off before we know how much opener has.

Solving this problem is not difficult using the 3NT slam try convention. After the 3♠ splinter, responder will make one of three choices:

• Bid 4 of opener's major as a signoff with a terrible hand -- usually a hand that is no longer worth a 2NT response due to the wasted values in partner's shortness
• Cue bid with enough extra to think slam is possible when opener holds a working minimum.
• Bid 3NT to show an in-between hand.

Using this convention, responder will bid 3NT on the above hand to send the message, "My cards are useful, but I am not strong enough to try for slam if you are in the minimum range." If responder had held an extra ace, he would bid 4♣, but now his call would say, "I have significant extras--please cooperate if you have a working minimum." in addition to "I have a club control." After 3NT, opener can cuebid with extras (14+) or signoff with a minimum.

The 3NT slam try I presented above is called non-serious or frivolous 3NT. You can read my earlier article to learn how it is used in other auctions. In J2N auctions, non-serious 3NT should apply when the auction has reached 3 of partner's major. If the auction is less than 3 of his major, I have a different suggestion.

The Size Ask

Again we don't know how many values opener has, but at least we have the 3-level to sort things out. We could just start cuebidding but I think there is a better approach. I suggest that when opener's singleton is lower-ranking than his suit, we add one conventional followup to help us: After opener shows a singleton, responder's cheapest call asks for opener's range. Opener bids in steps:

• Step 1 = approx 11-13 HCP
• Step 2 = approx 14-16 HCP
• Step 3 = 17+ HCP

For example:

• 3♥ = 11-13 HCP
• 3♠ = 14-16 HCP
• 3NT = 17+ HCP

Continuations after a size ask

After the size ask, responder is the boss since he now has all the information to decide whether the partnership is in the slam zone. His position is similar to opener's position after a splinter raise. He knows opener's strength and his shape. He can calculate the SST accurately. When slam is possible he will cuebid or bid RKC. And when slam is unlikely, he can simply sign off.

What does it mean if responder does not make a size ask?

Responder can decline to ask for opener's size under two conditions. First, if he already knows enough to drive to slam. For example, responder could just bid RKC without first asking for size. The more common reason is when responder holds a singleton. If responder is too strong for a direct splinter (15-17 HCP), he can bid J2N planning to show his shortness at the 4-level with his next call.

When responder shows a singleton, he passes control to opener, who becomes captain. 

SST when responder has a singleton 

SST analysis provided valuable insights about how opener should bid when he holds two doubletons. What about when opener holds a singleton? Is there a difference in SST between different patterns that include a singleton?

♠-♥-♦-♣

4-4-3-2 (responder)

5-4-1-3 (opener)

SST is 3. Slam should require approximately 26 working HCP.

♠-♥-♦-♣

4-4-3-2 (responder)

6-3-1-3 (opener)

SST is still 3. Slam should still require around 26 working HCP.

♠-♥-♦-♣

4-4-3-2 (responder)

6-2-1-4 (opener)

SST is still 3, but this time we have an additional working doubleton. Does slam still need 26 working HCP?

That is a shock. Slam is nearly cold on only 21 working HCP! Why? Once again, the extra working doubleton reduced the loser count. In effect, the SST is 2, not 3. In addition, we had the 10th trump effect, reducing the number of HCP we need to run the trump suit. Together these two things reduced the number of working HCP needed to make slam from 26 to 21. What happens if we change the shape a little:

Opposite opener's 5-3-1-4 pattern, we need an additional ♥Q and ♠Q to make slam cold, which would bring the working HCP total to 25, about what we would expect for an SST of 3.

Opposite opener's 6-3-1-3 pattern, we need the ♥Q but not the ♠Q. Combined HCP needed is 23, 2 less than normal because of the 10th trump.

This example teaches a valuable lesson: hands with 10 cards in 2 suits should not be treated the same as 5-3-1-4 and 6-3-1-3 patterns (9 cards in 2 suits). The 10-card hands are potentially much more powerful because they have two short suits (a singleton and a doubleton). What can we do to separate these patterns without making our system more complex and artificial? Fortunately, there is a simple fix. We already have a bid that shows a doubleton and a singleton in the unbid suits: the jump to 4 of a new suit. 

Our SST analysis of 6-4-2-1 patterns suggests that 6-4-2-1 and 7-3-2-1 will be much closer in trick taking strength to 5-5-2-1 than to 5-4-3-1 and 6-3-3-1. Why not have opener jump in his side suit on all hands that have 10+ cards in 2 suits? For example:

Note that this is a better slam than when opener is 5-5 because the extra trump length makes it safer to take ruffs in dummy. 

Grouping 6-4-2-1, 7-3-2-1 and 5-5-2-1 together by jumping to the 4-level on all such patterns is not ideal. The 6- and 7-card trump suits are stronger because the QJ of trumps are often unnecessary when there is a 10+ card fit. However, it is a clear improvement over grouping 6-4-2-1 with the much weaker 6-3-1-3 and 5-4-3-1 shapes.

Conclusion

Accurate slam bidding requires one or both players to limit their hands. After J2N, when opener shows a singleton, he is unlimited. Adding a size ask allows responder to get all the information he needs to know when the partnership is in the slam zone. If the auction is too high for a size ask, then you can fall back on non-serious 3NT to assess when the partnership does and does not have the values for slam.

7-3-2-1 and 6-4-2-1 patterns are often significantly stronger than 6-3-3-1 and 5-4-3-1 patterns. They are better shown with a jump to the 4-level in the side suit to express their high playing strength.

In these articles, i have tried to provide some tips to improve J2N auctions for those players unwilling or unable to adopt a complex relay structure. These half measures are not perfect. In a serious partnership, I recommend that you look at something like Larry Cohen's relay-based J2N structure.  And when evaluating an artificial bidding structure, don't forget to think about SST analysis.  It is the key to recognizing which shapes should and should not be grouped together.