[Kungsgeten]

I've been tinkering with symmetric relays after our Swedish club. Fredin-Lindkvist used to play a variant of this, but they had a problem that they couldn't find a 4-4 heart fit partscore when responder was balanced and opener had the weak NT. I think this structure solves this, but it may have other flaws. Perhaps the negative 1D is overloaded, but I'm not sure.

1♣--
1♦ = 0-7 any OR 8-10 NT, not 4♠ and no five card suit.
1♥ = 4+♠, 8+. May have longer minor.
1♠ = 4+♥, 8+. May have longer minor. Not 8--10 NT with 4♥.
1NT = 5m332 or unbal 5+♣, 8+. No major. (8--10 or 16+ if 5332)
2♣ = 5+♦ unbal, 8+. No major.
2♦ = Multi. 0--4 with 6+M or 18+ flat NT no major.
2♥ = 11--14 NT, no major
2♠ = Not sure. Currently play it at INV both minors, but don't think its needed now.
2NT = Natural INV, wants to declare
3X = Not sure, currently play it as preemptive.
3NT = 15--17 NT, flat hand no major

• After the transfer to spades, the weak NT can support spades or bid 1NT. The strong hand can relay with 1♠ or make a natural rebid.
  
• After the transfer to hearts, the weak NT bids 2♣ (2--3♥, minimum if 3), 2♦ (max with 3--4♥) or 2♥ (min with 4♥). The idea is that 1NT isn't that likely as a spot to play when responder is either unbalanced / 5♥332 or INV+. Instead 1NT by opener is a relay.
  
• After 1NT and 2♣, step 1 is the weak NT and step 2 is a relay.

With these I think we can relay at +1, which I actually prefer since that's what we do over our 1♦ opening.

Compared with our current structure responder's major + minor lengths will be ambiguous more often, and it will not be possible to find a heart partscore when responder is 4-4 majors 8--10 and opener has the weak NT with 4 hearts.

[straube]

How about....

1D-0-7
.....No idea but you're in a better position than 0-7 or strong meanings

1H-8+, all with 4+S excepting 4S/5+H
.....1S-GF relay
.....1N-12-14
.....2S-12-14, 4 spades

1S-8+, 6D or 5+D/4C or bal
.....1N-11-13
.....2C-GF relay

1N-8+, 5+H
.....2C-GF relay
.....2H-12-14

2C-8+, 6+C or 5+C/4+D
.....2D-GF relay
.....2N-12-14
.....3C-12-14, fit

2D-8+, 4+H/5+C
.....2H-GF relay
.....2N-12-14
.....3C-12-14, fit
.....3H-12-14, fit

2H-8+, 4S/5+H
.....2S-GF relay
.....2N-12-14
.....3H-12-14, fit
.....3S-12-14, fit

2S-8+, 3-suited short S
.....2N-12-14
.....3C-GF relay
.....etc-12-14, fit

The 2L responses get you a little high but it would be rare that you're both minimum, plus you have a good chance of locating your best fit.

[nullve]

Kungsgeten, on 2016-June-18, 12:17, said:

it will not be possible to find a heart partscore when responder is 4-4 majors 8--10 and opener has the weak NT with 4 hearts.

If it's ok to miss a 4-4 H fit in that case, then the structure I posted here, with the difference that bal. 11-13 hands with 4 H are not exluded from 1♣ and

* 1♣-1♦; 1♠ = 11-13, 4 H
* 1♣-1♦; 1N = 11-13, 3- H
* 1♣-1♥; 1N = 11-13, 3-S4-H,

might also work and besides be mostly +0 when Opener is strong and Responder has 8+.

[Kungsgeten]

nullve, on 2016-June-19, 13:36, said:

If it's ok to miss a 4-4 H fit in that case, then the structure I posted here, with the difference that bal. 11-13 hands with 4 H are not exluded from 1♣ and

* 1♣-1♦; 1♠ = 11-13, 4 H
* 1♣-1♦; 1N = 11-13, 3- H
* 1♣-1♥; 1N = 11-13, 3-S4-H,

might also work and besides be mostly +0 when Opener is strong and Responder has 8+.

I think the problem here is the negative sequences when opener is strong: its basically a strong heart system! 1♣-1♦; 1♥-1♠. Now opener has 17+ and responder has 0-7, and we're at 1♠. It could perhaps work if lowering the range of the positive responses, but that has other problems.

[Kungsgeten]

straube, on 2016-June-19, 07:38, said:

How about....

1D-0-7
.....No idea but you're in a better position than 0-7 or strong meanings

1H-8+, all with 4+S excepting 4S/5+H
.....1S-GF relay
.....1N-12-14
.....2S-12-14, 4 spades

1S-8+, 6D or 5+D/4C or bal
.....1N-11-13
.....2C-GF relay

1N-8+, 5+H
.....2C-GF relay
.....2H-12-14

2C-8+, 6+C or 5+C/4+D
.....2D-GF relay
.....2N-12-14
.....3C-12-14, fit

2D-8+, 4+H/5+C
.....2H-GF relay
.....2N-12-14
.....3C-12-14, fit
.....3H-12-14, fit

2H-8+, 4S/5+H
.....2S-GF relay
.....2N-12-14
.....3H-12-14, fit
.....3S-12-14, fit

2S-8+, 3-suited short S
.....2N-12-14
.....3C-GF relay
.....etc-12-14, fit

The 2L responses get you a little high but it would be rare that you're both minimum, plus you have a good chance of locating your best fit.

What's the upside here? The 1♦ response is better defined, and the relays are at +0? It seems awkward when opener has the weak NT and 2♣+ responder can not force to game (no room for invites, and we're at the 3-level in a possible 5-2 or 4-3 fit). We also can not find a 4-4 hearts partscore when opener has the weak NT and responder is 8-10 with four hearts.

Fredin-Lindkvist played:

1C--- (11--13 NT or 17+)
1D = 0-7
1H = 8+ with 4+ spades, or 12+ balanced
1S = 8+ with 4+ hearts, unbalanced
1NT = 8--12 NT, not 4 spades
2C = Nat, F1
2D = Nat, no major, NF
2H = GF with 5+D and 4+C
2S = GF with 6+D
2NT = 8--11 1354 or 3154
3m = INV
3M = Pree
3NT = 14--18, no major
4m = SAT, 9--13
4M = Pree

[nullve]

Kungsgeten, on 2016-June-20, 00:45, said:

I think the problem here is the negative sequences when opener is strong: its basically a strong heart system! 1♣-1♦; 1♥-1♠. Now opener has 17+ and responder has 0-7, and we're at 1♠.

Assuming you still play Mexican 2♦, maybe

1♣-1♦; ?:

1♥ = "17-19", shapes consistent with (1N)-2♣+ (NT defence) / "20+", any
...1♠ = 0-7 (as before)
......1N = "20+", forcing²
......2♣+ = "17-19"¹, NT defence
...(...)
1♠ = 12-14, 4 H³
1N = 12-14, 2-3 H³
2♣+ = "17-19"¹, unbal. shapes not covered by NT defence⁴

?

¹ The idea is that this range (which may translate to 'rules of 26-28') will be so narrow that there will be no need for invites based on hcp alone.
² Like Midmac's 1N opening, only a tad stronger (rule of 29+?) if unbalanced.
³ Can play both 1♠ and 1N here as 12-14 bal. thanks to the Mexican 2♦ opening.
⁴ E.g. (31)(54) or 6m(331) if playing Asptro.

[Kungsgeten]

nullve: Yeah, something like that could work. Interesting

A downside is ofcourse that it is easier to extract penalties, compared to 12-14 NT semi-natural 1M rebids after 1C--1D.

Edit: Another alternative instead of "NT-defence or not" might be transfer rebids by opener:

1C--1D;
1H = 17--19 / GF unbal or 20--24 NT
...1S = 0-7
......1NT = Clubs 17--19 or diamonds GF or 20--21 NT
......2C = Diamonds 17--19 or hearts GF or 22--24 NT
......2D = Hearts 17--19 or spades GF
......2H = Spades 17--19 or clubs GF
1S = 12--14 NT, 4H
1NT = 12--14 NT, 2--3H
2CDHS = "Acol". NF about (19)20--22
2NT = 25+ NT

[straube]

Kungsgeten, on 2016-June-20, 00:56, said:

What's the upside here? The 1♦ response is better defined, and the relays are at +0? It seems awkward when opener has the weak NT and 2♣+ responder can not force to game (no room for invites, and we're at the 3-level in a possible 5-2 or 4-3 fit). We also can not find a 4-4 hearts partscore when opener has the weak NT and responder is 8-10 with four hearts.

Yes, those are the biggest upsides. Btw, after 1C-1N, you have a spare 2D rebid for min/max or fit/no fit information. The 2C and higher responses are less than 10% of the responses. In the case of 2C you have loads of room for opener to show a min vs max weak NT, fitting or not. In the case of 2D, you have a spare 2S rebid available. So opposite the rare 2H or 2S responses, responder can only know whether opener is fitting or not with a weak NT. But his range is 12-14 which is pretty tight anyway.

True, this loses the 4-4 heart fit. So what really? Many folks open a weak NT and lose both major fits all the time. I play IMprecision and also lose this heart fit on occasion. If your responder has 4 hearts and can invite, you have Stayman available. If responder has 4H/5D he uses Stayman and can pass either the 2D or 2H response.

[nullve]

Trying to come up with something nice over 1♣-1♦; 1♥-1♠(0-7, any); 1N(20+, any):

1♣-1♦; 1♥-1♠; 1N-?:

[AFTER MUCH EDITING:]

2♣ = DN, 4- S / SP, either 4441 or bal. w/ red card majority / SP, "unbal. w/ 5+ C or 4(41)4"
...2♦ = catchall
......2♥ = DN, 4- S
.........2♠ = GF relay
............2N = 4- H
............3♣+ = 5+ H¹
.........2N = "23-25 bal."?
.........(...)
......2♠ = SP, either 4441 or bal. w/ red card majority
.........2N = relay
............3♣+ = Bal(red) [see Bal(c) below]
.........(...)
......2N+ = SP, "unbal. w/ 5+ C or 4(41)4", Unbal(♣) [see Unbal(x) below]
...2♥ = "Acol 2♥ or 20-22 bal."
......P = 5+ H, game unlikely²
......2♠ = catchall
.........2N = "20-22 bal."
.........3♣+ = 5+ H, nat., NF [was "Acol 2♥"]
......2N+ = 1-under transfers opposite 20-22 bal.?
...2♠ = "Acol 2♠"
...(...)
2♦ = DN, 5+ S / SP, "5+ D or 1444", unbal.
...2♥ = "Acol 2♥ or GF"
......2♠ = DN, 5+ S
.........2N = GF relay¹
.........3♣+ = 5+ H, nat., NF [was "Acol 2♥"]
......2N+ = SP, "unbal. w/ 5+ D or 1444", Unbal(♦) [see Unbal(x) below]
...2♠ = "20-22 bal."
...(...)
2♥ = SP, "5+ H, unbal."
...2♠ = relay
......2N+ = Unbal(♥) [see Unbal(x) below]
...(...)
2♠ = SP, bal. w/ black card majority
...2N = relay
......3♣+ = Bal(black) [see Bal(c) below]
...(...)
2N+ = SP, "5+ S, unbal.", Unbal(♠) [see Unbal(x) below]


Bal(c):

3♣ = 4M(c)3m(c) or 5M(c)(332)
...3♦ = relay
......3♥ = 4M(c)333 or 5M(c)(332)
.........3♠ = relay
............3N = 4M(c)333
............4♣+ = 5M(c)(332) [good news, so can bypass 3N]
.........(...)
......3♠ = 4M(c)3m(c)24
......3N = 4M(c)3m(c)42
...(...)
3♦ = 4M(c)4m(c)(32) or, if c=red, 4441
...3♥ = relay
......3♠ = 4M(c)4m(c)23
......3N = 4M(c)4m(c)32
......4♣(c=red) = 4441
...(...)
3♥ = 4m(c)333 or 5m(c)(332)
...3♠ = relay
......3N = 4m(c)333
......4♣+ = 5m(c)(332) [good news, so can bypass 3N]
...)...)
3♠ = 3M(c)4m(c)24
3N = 3M(c)4m(c)42

where 'm(black)', 'm(red)', 'M(red)' and 'M(black)' would stand for 'C', 'D', 'H' and 'S', respectively.


Unbal(x):

Based on relay structure I play elsewhere:

2N = 4+ lo, not 5x5lo / 6+ x, 1-suited
...3♣ = relay
......3♦ = 6+ x, 1-suited
......3♥+ = U(x,lo) [see U(x,y) below]
...(...)
3♣ = 4+ mi, not 5x5mi
...3♦ = relay
......3♥+ = U(x,mi) [see U(x,y) below]
...(...)
3♦ = "5x5lower or (4441)", but not 4441
...3♥ = relay
.....3♠(x=♣) = 4414
.....3♠(x=/=♣) = 5x5C
.....3N(x=♣) = 4144
.....3N(x=♦) = 1444
.....3N(x=M) = 5M5D
.....4♣+(x=♠) = 5S5H
...(...)
3♥+ = 4+ hi, not 5x5hi, U(x,hi) [see U(x,y) below]


U(x,y):

Very simple version:

3♥ = 5x4y22 or 6+x4y
...3♠ = relay, usually 2+ x
......3N = 5x4y22
......4♣+ = 6+x4+y
...(...)
3♠ = 5x4y13 (5x4y04)
3N = 5x4y31 (5x4y40)
(...)

¹ Then 1♣-1♦; 1♥-1♠; 1N-2♣; 2♦-2♥(DN, 4-S S); 2♠(GF relay)-3♣+ can be very similar to 1♣-1♦; 1♥-1♠; 1N-2♦; 2♥-2♠(DN, 5+ S); 2N(GF relay)-3♣+.
² Gambling that partner has 20-22 bal.. I've seen Welland-Auken do this over 2♣(20-21 bal. or GF)-2♦(5+ H or waiting); 2♥(20-21 bal. or 5+ H) and I think it works!

This post has been edited by nullve: 2016-June-26, 23:26

[straube]

I've been looking at generated deals of weak NTs and the IMprecision-based response structure I suggested has been working very well. It should work even better for strong hands since it's set up for mostly +0 standard symmetric relays.

This doesn't solve the continuation difficulty of 1C-1D that has been created by lumping strong and weak NT together, but I think this is the best you're ever going to get for 1C-1H+ responses.

Maybe you can try it and let us know if it works for you as it did for me.

btw 1C-1S, 1N-2C is Stayman and 2D is just diamonds to play and 2H can be diamonds GF.
1C-1H, 1N and 2C is Stayman, 2D is just diamonds to play and 2H is a transfer.

[awm]

I'd go with something a little different to try and take advantage of the balanced nature of opener's weak hand. Perhaps:

1♦ = 0-7 any, or 8-11 balanced without 4M (all others 8+)
1♥ = 4+♠
1♠ = 4♥ balanced, or 4♥ and a longer minor, or 5+♥ and 4+♣
... 1NT = 11-13 no 4♥
... 2♣ = GF relay
... 2♥ = 11-13 with 4♥
1NT = 12+ balanced, or 8+ with 5+♣ (no 4M)
... 2♣ = 11-13 balanced
... 2♦ = GF relay (now 2♥ = 5+♣/4♦, 2♠ = balanced, 2N+ = 6+♣)
2♣ = 5+♥ (single suited or 4♦)
... 2♦ = GF relay
... 2♥ = 11-13 (maybe above here can be super-accept for heart with 11-13 and 4♥?)
2♦ = 6+♦ or 5+/5+ minors
... 2♥ = GF relay
... 2NT = 11-13 balanced, wanting to declare if we play 3NT (minimum responder bids 3♣ or 3♦ here)
2♥ = three-suited short in spades
... 2♠ = GF relay
... 2N = 11-13 want to play here (4333 or maybe other hands with strong spades)
... 3x = fit
2♠ = 5+♥ and 5+♦

This should get all the relays in at pretty much +0 steps. It also avoids some of the awkward sequences in straube's construction where you are forced at the two-level without a good stopping place. Here I think the only poorly placed sequence is 1♣-2♥ and even that will be okay most of the time (opps usually have 8+ spades and we usually have 8+ fit somewhere).

[straube]

I like Adam's structure and I see how it is better placed after a 2 level response. I'm also uncertain how Kungsgeten will continue after 1C-1D. If he needs 1C-1D, 1M-1N to be balanced 8-11, that's one less rebid for something else.

[Kungsgeten]

I also like awm's structure, but the 5D and 4C hand is missing at the moment.

After 1C-1D we play 1H as F1; natural weak hand or any unbal Acol strength hand. 1S is natural; weak or 17-19. Over these responder's 1NT rebid is natural 0-10 and forcing vs the strong hands.

[awm]

Yeah I had a feeling I was missing a hand type. My intent was to put:

8-12 5+♦/4♣ into 1♦
13+ (GF) with 5+♦/4♣ bids 2NT+

[nullve]

Possible improvement on awm's structure:

1♣-?:

1♦ = 0-7, any / 8-11 bal., 3-S3-H / 8-12, 3-S3-H4D5+C
(...)
1N = 8+, either 3-S3-H5+D4C or 1-suited with 6+ D / 12+ bal., 3-S3-H / ?
...2♣ = GF relay [the whole point]
...2♦ = 11-13 bal.
(...)
2♦ = 8+, either 5+D5+C or 1-suited with 6+ C
(...)
2N+ = 13+, 3-S3-H4D5+C

[Kungsgeten]

If the short spades three-suiter is included in 1NT, the 5+C, 4D hand can be excluded from the negative 1D response. Now opener can not get out at the 2-level, however.

1D = 0-7 / NT less than INV no major
1H = 4+S
1S = 4H bal / 4H longer minor / 5+H and 4+C
1N = GF NT, no major / 5+D, 4C / 6+D / Three-suiter short spades
2C = 5+H (single suited or 4D)
2D = 6+C or 5-5 minors
2H = 5+C, 4D
2S = 5+H and 5+D
2N = INV, no major
3X = Weak

[nullve]

Suppose that in 1st and 2nd seats there are ways to open all 6+ M3-OM and 7+m3-S3-H hands that would be worth a GF response to a strong 1♣ of type T by partner.*

If

P-1♣ = weak NT w/ no 5c M OR type T strong 1♣,

then it looks to me that after a positive (= GF opposite the type T strong 1♣) response the following structure is both pretty much +0 and allows the partnership to find all 4-4 major partials:

P-1♣; ?:

1♦ = negative
1♥ = positive, either 4 S, S+H 2- or 3-suiter or minor 1-suiter w/ "high shortage" (13(63), 22(63) or 23(62))
...1♠ = 4 H or strong
......1N = 4S3-H
.........P = weak NT
.........2♣ = GF relay
............2♦ = BAL
............2♥ = S+C reverser
............2♠ = 3-suited, short H
............2N+ = S+D reverser
.........(...)
......2♣ = S+H 2-suiter
.........2♦ = GF relay
.........2♥ = weak NT (and 4 H)
.........(...)
......2♦ = 4423 or S+H+C 3-suiter
.........2♥ = weak NT (and 4 H)
.........2♠ = GF relay
............2N = 4423
............3♣ = 4414
............3♦ = 4405
............3♥ = 4504
............3♠ = 5404
.........(...)
......2♥ = 4432 or S+H+D 3-suiter
.........P = weak NT (and 4 H)
.........2♠ = GF relay
............2N = 4432
............3♣ = 4441
............3♦ = 4450
............3♥ = 4540
............3♠ = 5440
.........(...)
......2♠ = either 1336, 2236 or 2326 (C 1-suiter, "high shortage")
.........2N = GF relay
............3♣ = 1336
............3♦ = 2236
............3♥ = 2326
.........3♣ = weak NT
.........(...)
......2N = either 1363, 2263 or 2362 (D 1-suiter, "high shortage")
.........3♣ = GF relay
............3♦ = 1363
............3♥ = 2263
............3♠ = 2362
.........3♦ = weak NT
.........(...)
...1N = weak NT, < 4 S
...(...)
1♠ = positive, either D+C 2-suiter or BAL w/ no major
...1N = weak NT
...2♣ = GF relay
......2♦ = BAL
......2♥ = D+C reverser
......2♠ = 5+D5+C
......2N+ = D+C anti-reverser
...(...)
1N = positive, 4H3-S
...P = weak NT, < 4 H
...2♣ = GF relay
......2♦ = BAL
......2♥ = H+C reverser
......2♠ = 3-suited, short S
......2N+ = H+D reverser
...(...)
...2♥ = weak NT, 4 H
...(...)
2♣ = positive, either M+D non-reverser, 0544 or 5H(332)
...2♦ = GF relay
......2♥ = S+D non-reverser
.........2♠ = GF relay
......2♠ = 0544 or 5H(332)
.........2N = GF relay
............3♣ = 2533
............3♦ = 3523
............3♥ = 3532
............3♠ = 0544
......2N+ = H+D non-reverser
...2♥ = weak NT (P/C)
...(...)
2♦ = positive, either S+C non-reverser, 5044 or 5S(332)
...2♥ = GF relay
......2♠ = 5044 or 5S(332)
.........2N = GF relay
............3♣ = 5233
............3♦ = 5323
............3♥ = 5332
............3♠ = 5044
......2N+ = S+C non-reverser
...2♠ = weak NT
...(...)
2♥ = positive, H+C non-reverser
...P = weak NT
...2♠ = GF relay
......2N = S (= high) shortage
......3♣ = even shortage
.........3♦
............3♥ = 2524
............3♠ = 1516
......3♦ = 3514
......3♥ = 2515
......3♠ = 3505
......3N = 2506
...(...)
2♠ = positive, either 3136 or 3236 (C 1-suiter, "middle shortage")
...2N = GF relay
......3♣ = 3136
......3♦ = 3236
...3♣ = weak NT
...(...)
2N = positive, either 3163 or 3262 (D 1-suiter, "middle shortage")
...3♣ = GF relay
......3♦ = 3361
......3♥ = 3262
...3♦ = weak NT
...(...)
3♣ = positive, 3316 (C 1-suiter, "low shortage")
...P = weak NT
...(...)
3♦ = positive, 3361 (D 1-suiter, "low shortage")
...P = weak NT
...(...)

My own interest in Swedish Club-like P-1♣ structures is mainly in a context where

P(1st/2nd seat): up to 13 if BAL but does not include unBAL "opening hands" (e.g. hands that meet the rule of 19)

and some very weak hands therefore need to be opened in 3rd/4th seat just to protect the partscore. One attractive possibility is then to let

P-1♣ = "mini-NT" with 8 walrus** - 10 OR "strong club",

which also allows

P-1♣; 1♦-?:

P = the "mini-NT"
1♥+ = the "strong club" (only)

.

* For example, if T is standard Precision (so that 1♣ is ~ 16+ if unBAL but ~ 17+ if BAL), then we assume there are ways to open all 6+M3-OM and 7+m3-S3-H hands with 8+ hcp.
** Minimum 8+ Milton Work points is required in order for the system to be non-HUM

[ulven]

Kungsgeten, on 2016-June-18, 12:17, said:

Fredin-Lindkvist used to play a variant of this, but they had a problem that they couldn't find a 4-4 heart fit partscore when responder was balanced and opener had the weak NT.

That structure was invented in 1990 in Magnus and my partnership. Lindkvist-Fredin started playing 1997, and this was pulled from the shelves.

In the Junior Europeans 1990,I played it with Johan Ebenius, and one success of this design "flaw" came on Viewgraph vs Denmark. We bid 1C-1NT (8-11,bal,denies 4-card spades) down one and teammates reached making vulnerable 4S after being able to overcall 1S in 4th position.