BBO Discussion Forums: Inside-out Stayman - BBO Discussion Forums

Jump to content

Page 1 of 1
  • You cannot start a new topic
  • You cannot reply to this topic

Inside-out Stayman

#1 User is offline   nullve 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 1,217
  • Joined: 2014-April-08
  • Gender:Male
  • Location:Norway
  • Interests:partscores

Posted 2019-April-22, 06:11

Has anyone tried something like

1N-2("Stayman"): ?:

2 = 4-5 M
...P = allowed
...2 = like Crawling Stayman
......P = 4-5 H (4S4H possible?)
......2 = 4-5 S (4S4H possible?)
...(...)
2 = 3-S3H (2S2H also possible?)
...P = 4+ H (but not 5+S4H), weak
...2 = 5+S4H, to play
...(...)
2 = 3S2H
...P = 4+ S, weak
...(...)

?

Right now (after thinking about it for maybe an hour) this seems better than Garbage Stayman + Crawling Stayman when Responder has both majors and less than GF values.

Agree?
0

#2 User is offline   spotlight7 

  • PipPipPipPip
  • Group: Full Members
  • Posts: 308
  • Joined: 2009-March-21

Posted 2019-April-22, 07:03

View Postnullve, on 2019-April-22, 06:11, said:

Has anyone tried something like

1N-2("Stayman"): ?:

2 = 4-5 M
...P = allowed
...2 = like Crawling Stayman
......P = 4-5 H (4S4H possible?)
......2 = 4-5 S (4S4H possible?)
...(...)
2 = 3-S3H (2S2H also possible?)
...P = 4+ H (but not 5+S4H), weak
...2 = 5+S4H, to play
...(...)
2 = 3S2H
...P = 4+ S, weak
...(...)

?

Right now (after thinking about it for maybe an hour) this seems better than Garbage Stayman + Crawling Stayman when Responder has both majors and less than GF values.

Agree?



I suspect partner will get ulcers playing in 4-2 fits.
0

#3 User is offline   nullve 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 1,217
  • Joined: 2014-April-08
  • Gender:Male
  • Location:Norway
  • Interests:partscores

Posted 2019-April-22, 07:33

View Postspotlight7, on 2019-April-22, 07:03, said:

I suspect partner will get ulcers playing in 4-2 fits.

You mean after

1N-2
2*-P

* now also with 2S2H

when Opener has 2S2H and Responder 4S4H?

If you play regular Stayman + Crawling Stayman and your 1N opening includes shapes with 2S2H, then this might happen after

1N-2
2-2

as well, depending on what Opener is supposed to do with these shapes.
0

#4 User is offline   awm 

  • PipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 7,985
  • Joined: 2005-February-09
  • Gender:Male
  • Location:Zurich, Switzerland

Posted 2019-April-22, 07:40

Iím not sure why you would think this is good. Some problems;

1. You canít use stayman on a weak (43)51 any more because partner will bid a three card major (could be your three card suit) instead of 2D when he has no 4M.
2. Itís not clear how you invite after 1nt-2c-2d; it seems like 2h is NF so you canít make puppet style bids? I guess maybe 2s is invite with four spades and 2nt is invite with four hearts? But this removes other possible uses of 2s in this auction (inv with five spades being one option).
3. You will often reveal more about openers shape on the way to 2nt/3nt than you really want to; for example after 1nt-2c-2s openers whole major suit holding is known.

I donít even really see any advantages here, except right siding some heart fits when responder has both majors (against which you also wrong side some when opener has 4h).
Adam W. Meyerson
a.k.a. Appeal Without Merit
0

#5 User is offline   steve2005 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,789
  • Joined: 2010-April-22
  • Gender:Male
  • Location:Hamilton, Canada
  • Interests:Bridge duh!

Posted 2019-April-22, 08:49

This complicates a nice simple convention like Stayman that anyone can use with little or no discussion. Why to find 3-card majors,.
There is a simple convention called Stayman in doubt which handles this without screwing up Stayman.
Sarcasm is a state of mind
0

#6 User is offline   nullve 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 1,217
  • Joined: 2014-April-08
  • Gender:Male
  • Location:Norway
  • Interests:partscores

Posted 2019-April-22, 08:51

View Postawm, on 2019-April-22, 07:40, said:

I’m not sure why you would think this is good. Some problems;

1. You can’t use stayman on a weak (43)51 any more because partner will bid a three card major (could be your three card suit) instead of 2D when he has no 4M.
2. It’s not clear how you invite after 1nt-2c-2d; it seems like 2h is NF so you can’t make puppet style bids? I guess maybe 2s is invite with four spades and 2nt is invite with four hearts? But this removes other possible uses of 2s in this auction (inv with five spades being one option).
3. You will often reveal more about openers shape on the way to 2nt/3nt than you really want to; for example after 1nt-2c-2s openers whole major suit holding is known.

1. True. Didn't think of that.
2. Oops! Forgot that not everyone is thinking inside a "pass or bash" paradigm where the equivalent of tradtional invites like 1N-2; 2x-2N, 1N-[2M-1]; 2M-2N and 1N-2N doesn't exist.
3. Assume Puppet Stayman (2N or 3) and Jacoby transfers on "3N or 4M" hands.

View Postawm, on 2019-April-22, 07:40, said:

I don’t even really see any advantages here, except right siding some heart fits when responder has both majors (against which you also wrong side some when opener has 4h).

I think the advantage depends on exactly how regular Stayman + Crawling Stayman is played. But Inside-out Stayman can obviously handle weak hands with both majors just as well as

1N-2; 2-?:

2 = Crawling Stayman, but not with 5+S4H
2 = 5+S4H, weak

and better than

1N-2; 2-?:

2 = Crawling Stayman
2 = 5(+) S, INV

It can also handle INV hands with 5+M4+M better than the former structure and INV hands with 4+S5+H/5+S4+H probably better/worse (resp.) than the latter.

This post has been edited by nullve: 2019-April-24, 12:45

0

#7 User is offline   nullve 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 1,217
  • Joined: 2014-April-08
  • Gender:Male
  • Location:Norway
  • Interests:partscores

Posted 2019-April-22, 09:08

View Poststeve2005, on 2019-April-22, 08:49, said:

This complicates a nice simple convention like Stayman that anyone can use with little or no discussion. Why to find 3-card majors,.
There is a simple convention called Stayman in doubt which handles this without screwing up Stayman.

The point is not to find 3c majors when Responder is strong. I use different gadgets for that.

The point is to improve partscore and game bidding when Responder has less than GF values.
0

#8 User is offline   pescetom 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 1,572
  • Joined: 2014-February-18
  • Gender:Male
  • Location:Italy

Posted 2019-April-22, 14:58

View Postnullve, on 2019-April-22, 08:51, said:

I think the advantage depends on exactly how regular Stayman + Crawling Stayman is played.


Out of interest, how often do people really encounter Crawling Stayman situation gains?
I always liked the idea, never did the maths, but it happened very rarely in the 2 years we tried it.
0

#9 User is offline   steve2005 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,789
  • Joined: 2010-April-22
  • Gender:Male
  • Location:Hamilton, Canada
  • Interests:Bridge duh!

Posted 2019-April-24, 07:57

View Postpescetom, on 2019-April-22, 14:58, said:

Out of interest, how often do people really encounter Crawling Stayman situation gains?
I always liked the idea, never did the maths, but it happened very rarely in the 2 years we tried it.
Crawling Stayman only has a potential benefit with Weak NT and that is fleeting.

Here inability of being able to play in 2D using Garbage Stayman will use up any other benefits if they exist.
Sarcasm is a state of mind
0

#10 User is offline   pescetom 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 1,572
  • Joined: 2014-February-18
  • Gender:Male
  • Location:Italy

Posted 2019-April-24, 08:39

View Poststeve2005, on 2019-April-24, 07:57, said:

Crawling Stayman only has a potential benefit with Weak NT and that is fleeting.

Here inability of being able to play in 2D using Garbage Stayman will use up any other benefits if they exist.

It certainly was rarely useful with strong NT. We had to give it up because it wouldn't fit into our current Puppet-like Stayman, but at least that gave us a relatively safe capacity to play in 2D.
0

#11 User is offline   nullve 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 1,217
  • Joined: 2014-April-08
  • Gender:Male
  • Location:Norway
  • Interests:partscores

Posted 2019-April-24, 14:22

I've corrected some of the gibberish in my previous post.

I could also have mentioned there that if Responder is weak with both majors, then

1N-2(Stayman); 2(no major)-?:

2 = Crawling Stayman, but not with 5+S4H
2 = 5+S4H, weak

is a lot like playing (Multi)Landy 2 in response to 1N, while

1N-2(Stayman); 2(no major)-?:

2 = Crawling Stayman (also with 5S4H)
2 = 5(+) S INV

is more like playing DONT 2 (or Cappelletti 2) in response to 1N.

Conversely, not playing (the equivalent of) Crawling Stayman might be a lot like having no way to show both majors when opps open 1N.
0

#12 User is offline   steve2005 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 2,789
  • Joined: 2010-April-22
  • Gender:Male
  • Location:Hamilton, Canada
  • Interests:Bridge duh!

Posted 2019-April-27, 18:00

What do you do with a weak hand, say zero points for sake of arguement and
4351 after 1N-2-2 or do you have another sequence for this distribution?
Sarcasm is a state of mind
0

#13 User is offline   nullve 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 1,217
  • Joined: 2014-April-08
  • Gender:Male
  • Location:Norway
  • Interests:partscores

Posted 2019-April-28, 04:21

View Poststeve2005, on 2019-April-27, 18:00, said:

What do you do with a weak hand, say zero points for sake of arguement and
4351 after 1N-2-2 or do you have another sequence for this distribution?

As awm pointed out, it's not safe to respond 2(Inside-out Stayman) on weak hands with this shape. True, over 1N-2; 2(3S2H) Responder can either just pass (and then 2 will probably be an OK Moysian) or get out in 3*, which will be on a 5-3 or better fit unless Opener is precisely 3226. But over 1N-2; 2(3-S3H) Responder has no choice but to get out in 3*, which could easily be on a 5-2 fit.

But passing isn't safe either, so suppose he chooses to respond 2 anyway. Then over 1N-2; 2 he can either pass, possibly missing a 4-4 S fit, or rebid 2(P/C**), reaching the same contracts as if playing Garbage Stayman, although 2 will now be played by Responder (after 1N-2; 2-2; P).

* say, via 3 as a puppet to 3
** but described as 'Crawling Stayman-like' in the OP
0

#14 User is offline   nullve 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 1,217
  • Joined: 2014-April-08
  • Gender:Male
  • Location:Norway
  • Interests:partscores

Posted 2019-May-04, 06:00

A variation:

1N-2; ?:

2 = 4-5 M or 33(52)
...P = allowed
...2 = 4+ H, weak
......P = 4-5H or 33(52)
......2 = 4-5S2-3H
...2 = 4S3-H5m, weak
......P = 4-5 S or 3S2m (or else 2N or P/C must be better)
......2N = 2533
......3 = P/C (but not 2533)
...(...)
2 = 2S3H or 33(43)
...P = 4+ H (but not 5+S4H), weak
...2 = 5+S4H, to play
...and e.g.
...3m-1 = PUP 3m
...(...)
2 = 3S2H
...P = 4+ S, weak
...and e.g.
...3m-1 = PUP 3m
...(...),

(conveniently ignoring the possibility that Opener is 22(54) or 22(63))

The idea is that if Responder is 34(51) or 4S3-H5m (yes, not just (43)51) and the bidding goes

1N-2
2-3m-1*
3m-P,

* PUP 3m

then a 5-3 or better m fit is guaranteed unless Opener is 23(62) with the wrong minor.

The main downside compared with the OP structure seems to be that if the bidding goes

1N-2
2-2
P,

Opener is 33(52) and Responder is weak with 5M4OM, then the wrong major has been reached half of the time.

It also gets more complicated over 1N-2; 2 when Responder has something (54)(31) and enough to force to game opposite 4-5 M but wants to stay out of game opposite 33(52). But here's a half-baked idea for a mechanism that allows him to do both things:

1N-2; 2-?:

P/2/2: as above
2N = PUP 3
3M-2 = PUP 3M-1 unless Opener is 33(52)
...3M-1 = GF, not 33(52)
......(...)
......4 = 5+M4+OM, asks Opener to bid his major
......(...)
...3M = 33(52)
......P = 5+M4+OM, lost interest in game
......(...)
(...)
0

#15 User is online   straube 

  • PipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 3,828
  • Joined: 2009-January-18
  • Gender:Male
  • Location:Vancouver WA USA

Posted 2019-May-04, 09:14

Your focus seems to be on escaping 1N with certain weak distributional hands.

I know you use Puppet Stayman but that's a rather blunt instrument; there is a lot less room to react to opener's information.

How do you handle things that are routinely accomplished by Stayman for most folks like....

1) GI with one or both majors? Arrive sometimes in 3M?
2) GI with 5S?
3) GF with a major and a longer minor?

Maybe you use immediate splinters for certain hands (like 4144 for example) but how do you handle...

1) 4513 (can you show shortness and find all major suit fits?)
2) 2416 slam going (splinter in support of hearts?)
0

#16 User is offline   nullve 

  • PipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 1,217
  • Joined: 2014-April-08
  • Gender:Male
  • Location:Norway
  • Interests:partscores

Posted 2019-May-05, 10:20

View Poststraube, on 2019-May-04, 09:14, said:

Your focus seems to be on escaping 1N with certain weak distributional hands.

I know you use Puppet Stayman but that's a rather blunt instrument; there is a lot less room to react to opener's information.

How do you handle things that are routinely accomplished by Stayman for most folks like....

1) GI with one or both majors? Arrive sometimes in 3M?
2) GI with 5S?
3) GF with a major and a longer minor?

Maybe you use immediate splinters for certain hands (like 4144 for example) but how do you handle...

1) 4513 (can you show shortness and find all major suit fits?)
2) 2416 slam going (splinter in support of hearts?)

I usually want to distinguish between the following 5 Responder ranges in my 1N structures,

WK: The weakest range. Responder has no game interest.
INV: The only invitational range. Responder wants to be in game if and only if a sufficient trump fit exists.
GF1: The weakest GF range. Slam is at best a remote possibility and the focus is solely on finding the right game.
GF2: The intermediate GF range. Slam is possible only if Responder is unbalanced, but then he is not strong enough violate the balanced hand principle.
GF3: The strongest GF range. Slam is likely and Responder is strong enough to relay, possibly violating the balanced hand principle,

But a lot of things have to be different than in the regular Stayman + Crawling Stayman-based structure(s) I currently play, and this is what I've come up with so far:

1N-?:

(...)
2 = WK, various shapes* / INV, 5+M4+OM / GF1, either 4S4H1-m or 5+M4+OM / GF2, either 44(41) (44(50)), (54)22 or "unBAL w/ D"
2 = WK, 3-S5+H / INV, 5H5+m / GF1, either 3-S4+H1-m or 1-S4+H / GF2, "unBAL w/ H" / GF3
2 = WK, 5+S3-H / INV, 5S5+m / GF1, either 4+S3-H1-m or 5+S1-H / GF2, unBAL w/ S"
2 = GF2, "unBAL w/ C"
2N = "nullve/Kungsgeten Puppet" (described here), here supposed to take care also of all GF2 (5422) shapes except (54)22
3: currently WK with 5+D5+C, but now I can probably respond 2 with this hand type as well, so...
3 = GF1, 1-M3-OM
3 = GF1, 3-S3-H1-C
3 = GF1, 3-S3-H1-D
3N = to play
(...)

* including 4+S4+H, (43)(51), 4S2-H5m and 4-S4-H6+m

1N-2; ?:

2+: as in the variation above where 2 = 4-5 M or 33(52) (= Almost inside-out Stayman?)

(I'm still not sure what Opener should do with 22(54) if that those shapes are allowed. One option is

2N = 2245
3 = 2254,

which is e.g. what Bocchi-Duboin played over 1N-2; 2-2(Crawling Stayman), but clearly that won't always work either.)

1N-2; 2-?:

P = WK, either 3-S3-H6+D or 5+D5+C
2/: as above
2N = WK, 3-s3-H6+C / INV, 5+M4+OM / GF1, either 4S4H1-m or 5+M4+OM / GF2, either 44(41) (44(50)) or (54)22
...3 = forced
......P = WK, 3-S3-H6+C
......3+: not sure, but maybe
......3 = INV, 5+M4+OM / GF2, (54)22
.........3 = 33(52)
............P = INV, 4+S5+H
............3 = INV, 5+S4+H
............(...)
.........(...)
.........4M = to play
......3 = GF1, 4S4H1-C / GF2, 4441 (4450)
.........3 = range ask
............3N = GF1
............4 = GF2
........(...)
......3 = GF1, 4S4H1-D
......3N = GF2, 4414 (4405)
......4 = GF1, 5+M4+OM
.........4 = 4S4H ("bid your shorter major, partner!")
.........4M = 4-5 M, to play
......(...)
3+ = GF2, "unBAL w/ D"

1M-2; 2M-?:

P/2: as above
2N = WK, 3-S3-H6+C / GF1, either 4S4H1-m or 5+M4+OM / GF2, either 44(41) (44(50)) or (54)22
...3 = forced
......P = WK, 3-S3-H6+C
......3+: not sure, but maybe
......3 = GF1, 4+M5+OM / GF2, 4M5OM22
.........3 = anti-SPL C
.........3 = anti-SPL D
.........(...)
......3 = GF1, 4S4H1-C / GF2, 4441 (4450)
.........3 = range ask
............3N = GF1
............4 = GF2
.........(...)
......3 = GF1, 4S4H1-D
......3N = GF2, 4414 (4405)
......(...)
......4M = to play
3+ = GF2, "unBAL w/ D"

* just like 3 over 1N-2; 2-2N; 3, which is nice, although it really goes against my principles to invite here

1N-[2M-1]; ?:

2M = 2-3 M or 4M333
2M+1 = 4-5 M, not 4M333

1N-[2M-1]; 2M-?:

2(M=) = GF3 (relay)
2N = INV, 5M5+m / GF1, either 4+M3-OM1-m or 4+M1-OM / GF2, 4M144
...3 = 2 M
......P = INV, 5M5+C
......3 = INV, 5M5+D
......3 = GF1, 4+M1-OM / GF2, 4M144
.........3 = range ask
............3N = GF1, 4-5 M
............4 = GF2
............(...)
.........(...)
......3 = GF1, 4+M3-OM1-C
......3N = GF1, 4-5M3-OM1-D
......(...)
...3 = 3 M or 4M333
......3 = GF1, 4M1-OM / GF2, 4M144
.........3 = range ask
............3N = GF1
............4 = GF2
............(...)
.........(...)
......3 = GF1, 4M3-OM1-C
......3N = GF1, 4M3-OM1-D
......(...)
......4M = 5+ M, to play
3+ = GF2, "unBAL w/ M" except that 3, a n actual puppet to 3), can also be WK with 4-M5+D

Here's an idea for the unBAL 3+ structure on GF2 hands:

1N-2; 2x-3+ = Can(). See Can(x) below.
1N-[2M-1]; 2M-3+ = Can(M). See Can(x) below.
1N-2; 3+ = Can(). See Can(x) below.

Can(x) ('Can' for 'canapé'):

This is essentially a more sophisticated and canapé version of the relay structure I described here

If lo, mi and hi are resp. the lowest, middle and highest suits outside x, then

3 = 4+x5+lo, but not 5x5lo OR 1-suited
...3 = relay
......3 = 1-suited
......3+ = U(x,lo). See U(x,y) below.
...(...)
3 = 4+x5+mi, not 5x5mi
...3 = relay
......3+ = U(x,mi). See U(x,y) below.
3 = "5-5" (although I'm not sure how to best split shapes corresponding to the six combinations of 5a5b (a and b different suits) evenly among the four "5-5" bids (one for each x)*)
3+ = U(x,hi). See U(x,y) below.

* But if

ab(x) = the "adjacent" suit "below" x (S,C,D,H if x is C,D,H,S, respectively)
na(x) = the suit not "adjacent" to x (H,S,C,D if x is C,D,H,S, respectively)
Or(x) = the other suit of the same rank as x,

to borrow some definitions from the same post, then maybe

3 = 5x5ab(x) or 5x5na(x)2+Or(x)
...3 = relay
......3N = 5x5ab(x), high shortage
.........4 = relay
............4 = 5x5ab(x)03*
............4 = 5x5ab(x)12*
......4 = 5x5na(x)2+Or(x)
......4 = 5x5ab(x)30*
......4 = 5x5ab(x)21*
...(...)

* Resolving (5530) shapes more cheaply than (5521) shapes makes sense in a key card-oriented relay structure if the fragment is a potential trump suit but not the doubleton.

?

U(x,y):

3 = 4x5y13 (4x5y04)
3N = 4x5y31 (4x5y40)
4 = 4+x6+y0hi (but not 4x6y0hi) or 5x6y11
...4
......4 = 5x6y11
......4 = 5x6y02
......4N = 4x7x02
......etc.
4 = 4x6y(21), final shape resolution after spiral scan for key cards
4 = 4x7y11
4 = 5x6y20
4N = 4x7y20
etc.

(The "even shortage" shapes (5x6y11 and 4x7y11) are split between the "high shortage branch" (4) and "the low shortage branch" (4+).)

(Why canapé? Because responses to Opener's 3 relay commits us to the 4-level, which is ok if Responder promises a 6c major (as in Can), but not necessarily ok if he only promises a 6c minor (as in a non-canapé version of Can).)

If you wonder where the (6430) hands have gone, the answer is that I want this hand type to be shown below 4 as long as there are three possible trump suits, so it has to be shown independently of Can. Then maybe the above structure can be adjusted so that

1N-2; 2x-2N; 3-3; 3-4/ = GF2, 4630/6430 (only 2 possible trump suits)
1N-2; 2x-2N; 3-4/ = GF2, 4603/6403 (only 2 possible trump suits)
1N-2; 2-2N; 3-3; 3N-4/ = GF2, 3460/3604
1N-2; 2-2N; 3-4/ = GF2, 3406/3604
1N-2; 2-2N; 3-3; 3N-4/ = GF2, 4360/6340
1N-2; 2-2N; 3-4/ = GF2, 4306/6304
1N-3; 3-3; 3N-4/ = GF2, 3046/3064
1N-3; 3-4/ = GF2, 0346/0364

?

WARNING: I will almost certainly continue editing this post!

11 May 2019: Here are 100 EW hands (with some auctions after a "14-16" NT) dealt randomly on the condition that West has either 15 walrus points and strictly balanced shape or 14 walrus points and 5m(422):

Spoiler


12 May 2019: 100 more hands (with some auctions) randomly dealt with the above constraint:

Spoiler

This post has been edited by nullve: 2019-May-13, 14:08

0

Share this topic:


Page 1 of 1
  • You cannot start a new topic
  • You cannot reply to this topic

1 User(s) are reading this topic
0 members, 1 guests, 0 anonymous users