This situation comes up so few, I prefer 2N-3N natural. No opportunity to dbl for the lead or gain info about 2N opener hand distribution.

This. 2N-3N definitely needs to be natural. If you have to bid 2N-3S to get to 3N, you are giving the opps way too much, and there are other ways to handle the 54 majors hand. Usually 2N-3S is used to show some sort of minor suit slam try; often 'clubs or diamonds or both minors' which is difficult to sort out in the space available.

If you are willing to give up Gerber, you can remove the diamond hands from 3S:
--> 2N-3S = slammish with clubs or both minors; opener usually bids 3N but can super-accept clubs by bidding 4C (or 4D RKC for clubs). If opener bids 3N, R bids 4C (or 4D RKC) with just clubs, or 4M shortness with both minors

--> 2N-4C = slammish with diamonds; opener can bid 4D, 4H (RKC) or 4NT ('terrible hand for diamonds').

So do you compensate with a particularly complicated puppet stayman structure (I suspect not, since this gives more opportunity to X for the lead)? Or do you just treat 5=4=2=2 hands as 4=4=(32) hands or 5=3=(32) hands as responder?

This is fine, but you didn't suggest another way to handle the 5=4 majors hand.

1. Smolen (ie, don't play Puppet Stayman over 2N; a lot of top players seem to be going in this direction).

2. Muppet Stayman: 2N-3C-3H = no 4 or 5-card major. 2N-3C-3N = 5H. With 4S5H, responder can transfer then bid 3S; with 5S4H, responder can stayman then bid 3S (over, say 3H showing no 4 or 5M).

3. There are also some other options that I don't know well enough to describe.

My current preference is Smolen and regular Stayman over 2N.

I think there are 2N-3c variations of puppet stayman that allow you to also find a 5-3 spade fit if responder has 5s+4h. I never played them though, we just bid them as if it was a 4-4 or 5-3.

Also note that the 3s transfer to NT to later show clubs or diamonds got booted for 3s minor suit ask. I created a nifty system, which included things like 6-keycard rkc

obv never occurred.

puppet stayman over 2NT, like gerber, is a convention i could easily live without and never really miss it. i understand the theoretical gains from it, but i think in practice the actual gains are almost completely negligible. smolen works fine for the 5-4 hands, so you only lose on the hands where a lot of things go wrong and you have a vulnerable shortness and partner can't cover them and we're taking 10 tricks in a suit contract instead of 9 and all these variables aren't offset by the times when you have 9 tricks in NT and not 10 tricks in 4M when you manage to find a 5-3 fit.

with that said, i like a 3C response to 1NT to be a form of puppet stayman to conceal opener's major suit holdings.

many people play 6-keycard RKC after 2NT 3S 3NT 4M (showing shortness in that major and 5-5 minors)

with 5-4 minors and 4-4 minors you can bid 4S over 2NT or go through 3S then bid 4NT, both of which are quantitative, and that works fine compared to keycard. i think it's slightly superior to mss.

I agree that over 1N you need a way to find out if opener has a 5M; however I don't like 3C for that - the opps get a chance to double (even a failure to double gives opening leader some information).

I prefer 1N-2N asking for a 5M (with 1N-2N-3C = no 5M, giving the opps no unneccssary information about opener's 4Ms). Obviously, this requires other adjustments to the structure after 1N.

Of course there is some risk associated with it. In my partnerships, we bid 3C in response to 1NT with many hands that everyone else normally bids 2C with, so it takes a very specific situation where having 3C as stayman is costly (ie something like 3352 where you would just bid 3NT without puppet but 3C gets doubled and you have no 5-3 major fit and no way to recover from the information that a club is going to be led).

There's a lot of luck in bridge.

2NT – 3NT should theoretically be something artificial, but I promise partner will pass you in it. I dropped that after partner went 0-for-3, once as responder and twice as opener. And while that partner was particularly prone to such things, it's such a normal, predictable mistake that you need to allow for it in system design.

Delighted to see "Don't play Puppet Stayman" becoming a more popular answer to the 5-4 majors question

I just remembered the story behind the last timethis happened, the one that pushed us back to a more normal system.

Playing in an A/X pairs at the Long Beach NABC, I think it was, I opened 2NT (which we were playing standard 20 – 21; we hadn't converted yet to our bizarre canapé system). Partner responded 3NT, and the following conversation ensued:

Me: Alert!

RHO: Please explain.

Me: Artificial; a puppet to 4C. He can have a variety of hands, including a club one-suiter and various two-suiters not including clubs. I'll also tell you based on experience that it's extremely likely that he forgot and just wants to play 3NT.

RHO: How does he get to 3NT?

Me: BY first bidding 3S, which is a marionette to 3NT. With certain hands I can break the puppet to 4C, but a 3NT rebid after his 3S is absolutely required.

RHO: Does he forget that one too?

Me: Every time.


Partner stayed stone-faced throughout, to his credit, but as he laid his dummy down in 4NT or 5NT he explained that he had in fact forgotten for the third time running; whereupon we agreed that we needed to play something else.

So, what is the response when you have 5-6 points without a 4 card major and the opener bids 2NT?

Isn't 2N-3N to play?

he's suggesting using 2NT 3S as a puppet to 3NT either to play or presumably to show some other very specific hand type.

Technically a marionette, not a puppet, meaning it can't be declined under any circumstances; a puppet can be bypassed with an extreme hand. But yes.

I actually don't remember exactly what system we were using, some sort of weirdness I dreamed up, but 3S was the way to get to 3NT.

Note, of course, that using 1NT – 3NT as something artificial has the same practical problem: someone will screw it up.

... except I just realized I may have this backwards— now I can't remember. Lol at me. It was all discussed in Bridge World, but I don't remember when.

I'm getting old.

At my level, I tend to ignore all the exotic bidding theory talk, and treat every play problem with great interest.

Couple of bidding problems for you

Problem 1
T3
T732
AQ72
K83

Opp are vulnerable, you are not. IMPs

2c-pass-2d-3c
pass-??

2c=GF (any) or weak with both majors (say, 8-10)
2d=relay

Problem 2
JT74
5
AJ4
QT965

You are vulnerable, opp are not. IMPs.
1h-dbl-??

Right, so in (1) you give up puppet. It's a tradeoff.

In (2), you now allow them to double an artificial 3H for the lead, which doesn't seem much different than doubling 3S in the auction 2N - 3S - (x) - 3N.

I don't think the concerns about 2N-3S --> 3N giving too much to the opps are particularly grave is my only point.

1. Pass? 4C seems like a weird enterprise.

2. Pass. Not ideal but nothing better available.

Edit: Re Gabe's post obv

You are terribly wrong here ducy?

1. I guess I have to bid 4C. Seems like partner could be loaded.

2. Torn between 1S and pass, but going to try 1S.

1. I agree that giving up Puppet is a tradeoff, but when players like Meckwell, Hampson, and others don't use Puppet over 2N (but do have a 5M ask over 1N) I tend to respect their judgement.

2. The person doubling 3H rates to be on lead against 3NT, whereas if 2N-3S-3N is the only way to get to 3N, you are giving the opps information every time - even if they don't double (if opening leader has a weak hand and wants to try to hit his pard, he knows that pard isn't very interested in a spade lead). This is a price I'm willing to pay to make some sort of slam try, but not just to get to 3NT. Also, if you play 2N-3N is artificial, you have to budget for a few disasters before the partnership drills it into the memory banks. It takes a long time for the benefits of the convention to outweigh the disasters of its introduction.