[raspeball]

I wanted feedback on the following idea.
Assumption you use 1nt-2nt as puppet stayman, normally only with hands with gamestrength and want to check if you should play 3nt or 4♥/♠.

My proposed idea has the following respones:

3♣= No 5 card major. Could have one or both 4 card major(s), or could be a hand without 4 or 5 card major.
3♦=No 4 card major. Could have a 5 card major, or could be a hand without 4 or 5 card major.

So a hand with no 4 or 5 card major could choose to bid 3♣ or 3♦ randomly.


Further responses could be as follows:
1nt-2nt
3♣

3♦=4 card ♥ and 4 card♠
3♥=4 card ♠
3♠=4 card ♥
3nt= No 4 card major (Was searching for 5 card major)

1nt-2nt
3[♦

3♥=3 card ♠
3♠=3 card ♥
3nt= Both majors, want to play 4 major if opener have a 5 card major.

The advantage (i think) is that there could be less information leakage if we end up in 3nt.


My questions are:

1: Assuming that you want to play 1nt-2nt as puppet, do you think this idea has merit, or has it some flaw that i have not seen?
2: Do you think the method have any problems with disclosure, since a hand without 4-5 major, could bid both 3 ♣ or 3♦?
3: Do you know somebody playing this already?

[nige1]

Interesting convention but artificial and complex.
Please would a Moderator move this to Non-Natural System Discussion.

[Vampyr]

 raspeball, on 2020-April-24, 16:02, said:

2: Do you think the method have any problems with disclosure, since a hand without 4-5 major, could bid both 3 ♣ or 3♦?

Random responses are not usually truly random. They can be if partner, say, counts up his spot cards in the black suits and bids one thing if they are odd and the other if they’re even. If you don’t do something like this, there will be a pattern, and even if you haven’t worked out what it is you will have to disclose it. So yes there will be a problem.

[nullve]

Interesting idea!

So Opener's rebid is

* always 3♣ with a 4c major;
* always 3♦ with a 5c major;
* 3♣ and 3♦ with probabilities p and 1-p, respectively, with no major.

Obviously, you want p to be much higher than 0 so that Opener cannot be counted on to have 4 OM after

1N-2N
3♣-3OM*
3N-P,

* 4 M and < 4 OM

but you also want it to be much lower than 1 so that Opener cannot be counted on to have 5 OM after

1N-2N
3♦-3OM*
3N-P.

* 3-4 M and < 3 OM

I don't know what the best value of p is, but maybe 1/2 is close enough. If so, notice how the opening leader might apply "restricted choice" when guessing whether Opener has the other major or no major!

[raspeball]

 Vampyr, on 2020-April-24, 22:46, said:

Random responses are not usually truly random. They can be if partner, say, counts up his spot cards in the black suits and bids one thing if they are odd and the other if they’re even. If you don’t do something like this, there will be a pattern, and even if you haven’t worked out what it is you will have to disclose it. So yes there will be a problem.

Thanks for your reply. Yes this is kind of what i was thinking of. If i choose to say that with no 4 or 5 card major i bid 3♣ if I have an odd number of black cards, and 3♦ with an even number of black card, would that be ok?

[raspeball]

 nullve, on 2020-April-25, 06:09, said:

Interesting idea!

So Opener's rebid is

* always 3♣ with a 4c major;
* always 3♦ with a 5c major;
* 3♣ and 3♦ with probabilities p and 1-p, respectively, with no major.

Obviously, you want p to be much higher than 0 so that Opener cannot be counted on to have 4 OM after

1N-2N
3♣-3OM*
3N-P,

* 4 M and < 4 OM

but you also want it to be much lower than 1 so that Opener cannot be counted on to have 5 OM after

1N-2N
3♦-3OM*
3N-P.

* 3-4 M and < 3 OM

I don't know what the best value of p is, but maybe 1/2 is close enough. If so, notice how the opening leader might apply "restricted choice" when guessing whether Opener has the other major or no major!

Good point. This is something that I did not think about.
Lets say that the hand with a 4 or 5 card major are distributed like this:
25% have a 5 card major, and 75% have one/two 4 card(s) major.

Then i think it make sense that the hands without a 4 or 5 card major should bid 3♦ on only 25% of those hands, and 3♣ on the remaining 75%.

[pescetom]

 Vampyr, on 2020-April-24, 22:46, said:

Random responses are not usually truly random. They can be if partner, say, counts up his spot cards in the black suits and bids one thing if they are odd and the other if they’re even.

That would be legal, but not truly random to someone who knows anything at all about the deal (as partner does, even just looking at his own cards). It is an example of legal encryption, although difficult for partner to decrypt and pointless as it stands.

[Vampyr]

 raspeball, on 2020-April-25, 15:08, said:

Thanks for your reply. Yes this is kind of what i was thinking of. If i choose to say that with no 4 or 5 card major i bid 3♣ if I have an odd number of black cards, and 3♦ with an even number of black card, would that be ok?

Yes, but that is pretty revealing to declarer. Hand parity is an interesting idea though,

 pescetom, on 2020-April-25, 16:19, said:

That would be legal, but not truly random to someone who knows anything at all about the deal (as partner does, even just looking at his own cards). It is an example of legal encryption, although difficult for partner to decrypt and pointless as it stands.

You could make it more difficult by basing it on, say, your lowest red card. It is not pointless, as it is a way to (more or less) randomise two bids. The point is not whether partner can interpret it but whether declarer can.

[pescetom]

 Vampyr, on 2020-April-25, 18:59, said:

You could make it more difficult by basing it on, say, your lowest red card. It is not pointless, as it is a way to (more or less) randomise two bids. The point is not whether partner can interpret it but whether declarer can.

Sure, I just meant pointless as encryption in that partner has no interest in interpreting it. It's still a legal way to randomise two bids.

[awm]

When opener has a five-card major, it seems like you will always want to know what it is. If responder has 0-2 in each major there's no reason to bid puppet in the first place, so after 1nt-2nt-3♦ responder will always ask. After the common 1nt-2nt-3♦-3nt-pass, defenders know opener has no four or five card major, whereas using the more standard responses opener's four-card major holding will often be less obvious (i.e. 1nt-2nt-3♣-3nt-pass and opener could be 2-4 in each major, or 1nt-2nt-3♣-3♥-3nt and opener's heart holding is hidden). The 1nt-2nt-3♦-3M sequence is just asking for a lead directional double!

It seems like it's probably better to use the more standard rebids:

3♣ = no five-card major (may or may not have four-card major)
3♥/3♠ = that five-card major

[straube]

 awm, on 2020-April-26, 04:41, said:

When opener has a five-card major, it seems like you will always want to know what it is. If responder has 0-2 in each major there's no reason to bid puppet in the first place, so after 1nt-2nt-3♦ responder will always ask. After the common 1nt-2nt-3♦-3nt-pass, defenders know opener has no four or five card major, whereas using the more standard responses opener's four-card major holding will often be less obvious (i.e. 1nt-2nt-3♣-3nt-pass and opener could be 2-4 in each major, or 1nt-2nt-3♣-3♥-3nt and opener's heart holding is hidden). The 1nt-2nt-3♦-3M sequence is just asking for a lead directional double!

It seems like it's probably better to use the more standard rebids:

3♣ = no five-card major (may or may not have four-card major)
3♥/3♠ = that five-card major

I used to do this. Liked that it didn't leak much information. Why disclose whether opener has a 4-cd major if responder may not even have one? I also used 3D as any 4333 to be able to avoid mirrored 4M333 4M contracts. Using 3D for something so specific and relatively unimportant never seemed worthwhile but I never found a better use.

[raspeball]

 awm, on 2020-April-26, 04:41, said:

When opener has a five-card major, it seems like you will always want to know what it is. If responder has 0-2 in each major there's no reason to bid puppet in the first place, so after 1nt-2nt-3♦ responder will always ask. After the common 1nt-2nt-3♦-3nt-pass, defenders know opener has no four or five card major, whereas using the more standard responses opener's four-card major holding will often be less obvious (i.e. 1nt-2nt-3♣-3nt-pass and opener could be 2-4 in each major, or 1nt-2nt-3♣-3♥-3nt and opener's heart holding is hidden). The 1nt-2nt-3♦-3M sequence is just asking for a lead directional double!

It seems like it's probably better to use the more standard rebids:

3♣ = no five-card major (may or may not have four-card major)
3♥/3♠ = that five-card major

Thanks for your reply.

The reasoning for the proposed idea is that I hate the following auction: Responder search for one major, opener shows 5 card suit in the other, and we end up playing 3nt. Now opener have revealed way to much of his hand, and the defense gets much easier than it should be .
In my proposed idea the sequence:
1nt-2nt;
3♦-3♥(=3 or 4 ♠ and by implication 0-2♥)
3nt=Opener does not have 5 card ♠. Opener have now shown: No 4-5 card♠, not 4♥ (Opener might still have 5 card ♥).

But I see that my suggested alternative gives way to many lead directing options.

Inspired by the responses i have come up with an alternative to the alternative:
1nt-2nt;
?

3♣ responses. Denies 5 card major. May or may not have 4 card majors(Same as above)
3♦=5 card ♠, or no 4 or 5 card major.
3♥= 5 card ♥, or no 4 or 5 card major.


Now there is room for responder to show 3 card in the major that opener might have. There are different ways to do this, one is:

1nt-2nt;
3♦ ?

3♥=Asks opener to bid 3♠ with 5 spades
3♠=Asks opener to bid 4♠ with 5 ♠ (Responder wants to play the hand!).
3nt=Not interested in 5 card ♠.


1nt-2nt;
3♥ ?


3♠=Asks opener to bid 4 ♥ with 5 card hearts.
3nt=To play. Not interested in 5 card ♥.

True the proposed alternative also gives the opponent more lead directing opportunities, and occasionally gives away more information than needed about openers hand.
But it avoids the worst situation in my mind where opener shows a specific 5 card major and later ends up in 3nt.

[nullve]

The main problem with this second variant seems to be that whenever Responder is searching for a 5-3 major fit, Opener will not be able to deny 5 cards in the major without revealing that he has in fact no major.

(Compare

1N-2N
3♦(1)-3♥(2)
3N(3)-P

(1) 5c major or no major
(2) 3-4 S
(3) not 5 S (so either 5 H or no major)

with

1N-2N
3♦(1)-3♥(2)
3N(3)-P

(1) 5 S or no major
(2) 3-4 S
(3) not 5 S (and therefore no major).)

[raspeball]

 nullve, on 2020-April-28, 03:23, said:

The main problem with this second variant seems to be that whenever Responder is searching for a 5-3 major fit, Opener will not be able to deny 5 cards in the major without revealing that he has in fact no major.

(Compare

1N-2N
3♦(1)-3♥(2)
3N(3)-P

(1) 5c major or no major
(2) 3-4 S
(3) not 5 S (so either 5 H or no major)

with

1N-2N
3♦(1)-3♥(2)
3N(3)-P

(1) 5 S or no major
(2) 3-4 S
(3) not 5 S (and therefore no major).)

This is true. But it is less bad that the opponent knows that I don't have a 4 or 5 card major, than that they knows that I have 5 card major when I end up in 3nt.

[nullve]

Idea, using a fantasy step 3F between 2N and 3♣:

1N-2N = puppet to 3F

1N-2N; 3F -?:

3♣ = Puppet Stayman with 3-4 S and 3-4 H
3♦ = 4 H and 0-2 S
3♥ = 4 S and 0-2 H
3♠ = 3 H and 0-2 S
3N = 3 S and 0-2 H (NF)

Or one could, back in reality, use 2♠ as a puppet to 2N.

[nullve]

Another idea:

* Use the Jacoby transfer 2M-1 also as "M Stayman" on choice-of-game type hands with 4 M and 0-2 OM. (Problem: Must figure out how to sign off in 3N on these hands after 1N-2M-1; 2M(= 2-3 M or 4M333). Maybe use 3♠ over 1N-2M-1; 2M as a puppet to 3N?)
* Use the 2N response as a true puppet to 3♣, then over 1N-2N; 3♣,

P = weak hand with 6+ C
3♦ = GF, 4 cards in one major, 3 in the other
...3♥ = no 5c major
......3♠ = 4 H and 3 S
......3N = 4 S and 3 H
...3♠ = 5 H (=> e.g. 4♦ = wants partner to declare 4♥)
...3N = 5 S (=> e.g. 4♥ = wants partner to declare 4♠)
3♥ = GF, 3 S and 0-2 H
3♠ = GF, 3 H and 0-2 S
3N = NF, 3 S and 3 H

[awm]

How about:

1nt-2nt = 2 (or less)-3, 3-3, or 4-3 majors (either longer)
3♣ by opener forced and:
... 3♦ = 4-3 majors (opener bids three of a 4M or four of a 5M or 3nt with no interest)
... 3M = three cards in the other major, 0-2 here
... 3nt = 3-3 majors choice of games

With 4-4 or 2-4 majors just bid 2♣ stayman.

[nige1]

 awm, on 2020-April-28, 15:19, said:

How about 1NT - 2NT = PUP to 3♣ with 3 M and 4 OM; or with 0-2 M and 3 OM.
Then after 1NT - 2NT - 3♣ -

• 3♦ = ART 3 M and 4 OM. Then opener bids 3M with 4M, 4M with 5M, 3NT with no interest.
• 3M = ART 0-2 M 3 OM.
• 3NT = P/C 33 MM.

With 44 MM or 0-2 M and 4 OM, just bid 2♣ Stayman.

Simple, cunning, and conceals opener's hand. Although
1N - 2N - 3♣ - 3M: might invite a double for the lead

This post has been edited by nige1: 2020-May-02, 12:48

[raspeball]

 awm, on 2020-April-28, 15:19, said:

How about:

1nt-2nt = 2 (or less)-3, 3-3, or 4-3 majors (either longer)
3♣ by opener forced and:
... 3♦ = 4-3 majors (opener bids three of a 4M or four of a 5M or 3nt with no interest)
... 3M = three cards in the other major, 0-2 here
... 3nt = 3-3 majors choice of games

With 4-4 or 2-4 majors just bid 2♣ stayman.

That works.
But after the sequence
1nt-2nt
3♣-3♦
3♥(=4♥)-3nt(4 ♠ and 3♥)
the defenders know to much.

Also after the normal stayman sequence we provide the opponents with more information than we want to (With the 4-4 major hand with responder I think we have to accept it,but with the 4-2 hands it is avoidable.).


I actually thought about a similar scheme, where openers 3♣ was forced.
The scheme would then be:
1nt-2nt
3♣-?
3♦=4-4 in majors.
3♥=4 ♠, not interested in ♥.
3♠=4 ♥, not interested in ♠.
3nt=3-3 in majors.

The downside here is that you cannot show the hands with 3 cards in one major, and 0-2 cards in the other (But some of those hand might be bidding 1nt-3x to show a three-suited hand anyway, depending on your agreements).

[nullve]

 raspeball, on 2020-April-28, 23:25, said:

I actually thought about a similar scheme, where openers 3♣ was forced.
The scheme would then be:
1nt-2nt
3♣-?
3♦=4-4 in majors.
3♥=4 ♠, not interested in ♥.
3♠=4 ♥, not interested in ♠.
3nt=3-3 in majors.

No Puppet variant can beat regular Stayman when Responder is 4-4 in the majors. In fact, it can only do worse. Here's why:

Suppose Responder is 4-4 in the majors. Then,

* if Opener has a major suit M, the best possible auction would be one where the M fit gets established without Responder revealing that he also has 4 OM

(Compare e.g. your action

1N-2N(1)
3♣(2)-3♦(3)
3♥(4)-4♥
P

(1) puppet to 3♣
(2) forced
(3) 4 S and 4 H
(4) 4-5 H

with the regular Stayman auction (which also happens to be the LIPS (= Low Information Puppet Stayman) auction)

1N-2♣
2♥-4♥
P.)

* if Opener has no major, the best possible auction would be one where he shows that (which is equivalent to answering 'No' to the two questions 'Do you have 4 or 5 hearts?' and 'Do you have 4 or 5 spades?') and Responder doesn't show a specific major

(Compare your auction

1N-2N(1)
3♣(2)-3♦(3)
3N(4)-P

(1) puppet to 3♣
(2) forced
(3) 4 S and 4 H
(4) no major

or the LIPS auction

1N-2♣(1)
2♦(2)-3N(3)
P

(1) like regular Stayman, but not used on choice-of-game type hands with only one 4c major (which go through 1N-3♣, or LIPS proper)
(2) no major
(3) 4 S and 4 H

with the regular Stayman auction

1N-2♣
2♦-3N(1)
P

(1) 4 S or 4H

.)

This post has been edited by nullve: 2020-April-29, 05:04