[straube]

awm, on 2011-March-08, 17:24, said:

On to your 2♦ method, the issue is that your best fit will fairly often not be in opener's major suit. If you have a big minor suit fit, finding it will lead you to a better partial and quite possibly a nice game contract that you might otherwise miss bidding 1M-2♦-2M. I've seen people play this as "any invite" (for example I think it is part of Marshall Miles' unbalanced diamond system) but it doesn't seem to me that this works too well. I'd rather play 2♦ as natural and invitational-plus, with opener rebidding 2M on non-fitting minimums. This allows you to use it sometimes as a two-card limit raise (with side diamonds), but it also helps with: (1) diamond invite hands (2) hands where you have a real diamond fit you find it rather than having to play 2M on a 5-2 (3) two-suited minor invites, which can be hard to bid in other methods (4) certain kinds of fitting slam tries where knowledge of a nice side diamond suit in dummy may be key to a slam decision.

I've just looked at a few hands (maybe 50 or so) and I looked at what a liberal use of 2D might accomplish....coupled with 1S-2H being natural invitational and nf...

2D showing always only GI with...

A) 2 spades and 3+ diamonds
B) single-suited diamonds with usually less than 2 spades. With 2 spades one might miss a 6-2 fit unless starting with 1N
C) good 5/5 minor hands

I rather liked how it worked out.

1S-2D,
.....2H-minimum, 4H
..........2S-2-fit
.....2S-minimum
..........P-to play
..........2N-?
..........3C-5/5 minors
..........3D-6D
.....2N-maximum, no 7th spade or second 5-cd suit
..........3C-5/5
..........3D-6D
..........3H-2S/4H
..........3S-2S

I also didn't notice losing a lot of superior minor suit fits. I think I lost a 4/4 diamond fit when we had only a 5-2 spade fit.

Now when responder doesn't have 2 spades and doesn't have 5 hearts, it ought to pay more to bid a quiet 1N and see if partner can rebid a suit for which we are fitting. 1S-1N would still hold promising hands....just less of them.

I'm not convinced this is right. I'm dealing with a GCC restriction that 2D can't be used artificially except for another GF (and I have one of those). Also, we have a GCC restriction that 1S-2H be natural. But it seems like 1S-2H as invitational nf is useful.

[straube]

Looked at more hands and it confirms your point about missing superior fits. We're now thinking of 2D as a balanced hand or good diamonds or good 5m/5m. Do you like this any better?

1N-semiforcing, conceals GI 2-suited or clubs

2C-GF relay

2D-GI, nf, 3+ diamonds, balanced or good 6 diamonds or good 5m/5m
.....2H-4H, min
.....2S-5S, min
.....2N-ask
..........3C-good 5D/5C
..........3D-good 6D
..........3H-bal, 4 hearts
..........3S-bal, 5 hearts
..........3N-bal, 2 spades
.....3C-wants to show shortness, denies 4H
..........3D-relays for shortness
.....3D-4H and shortness
..........3H-relays for shortness
.....3H-5H
.....3S-5C
.....3N-5D

2H-6H GI

2S-constructive raise

2N-limit raise

3C-weak jump shift (meaning that 1N followed by 3C is GI)

3D-mixed raise

[glen]

straube, on 2011-March-08, 23:05, said:

2D-GI, nf, 3+ diamonds, balanced or good 6 diamonds or good 5m/5m
.....2H-4H, min

You really can't play 2♦ as nf, since what hands are going to pass this 3♦ showing bid? - hands with 4+♦s only, but then they risk finding opener with good 6+♦s or a shapely 5-5+. You get more value out of the bid by including other hand types, and not having opener show 4♥s, such as:

1♠-2♦: 3+ ♦s, GI+, various
-- 2♥ asks:
---- 2♠: GI 3♠s (I suggest that balanced without 3♠s start with a semi-forcing 1NT, but if you wish this can show the GI 2♠s balanced hand type even though getting to 2♠ instead of 2NT on GI hands produces just random swings, often based on the opponents distribution).
---- 2NT: 6+♦s and 4 of a rounded suit, GI. 3♣ asks, 3♦ showing ♣s
---- 3♣: 5-5+ minors, GI
---- 3♦: good 6+♦s, no second suit, GI
---- 3♥+: various GF and/or slam tries
-- 2♠: 6+♠s, no extras
-- 2NT: 6+♠s, establishes GF opposite all GI, asks hand description.

[glen]

see above

[akhare]

straube, on 2011-March-08, 23:05, said:

Looked at more hands and it confirms your point about missing superior fits. We're now thinking of 2D as a balanced hand or good diamonds or good 5m/5m. Do you like this any better?
<snip>
2N-limit raise

3C-weak jump shift (meaning that 1N followed by 3C is GI)

3D-mixed raise

Glen's suggestion about 2♦ being various GI hand types with 3+ ♦ looks very good.

Also, instead of using 3♣ as WSJ:

2N: GF raise
3♣: Mixed raise / LR
......3♦: Ask, presumably wants to bid game opposite LR
..........3M: Mixed raise
..........Other: LR
......3♥ / 3♠: To play

[awm]

I'm sort of mystified at this point.

Straube seemed to indicate that he sees the point about finding minor suit fits, but then suggest a method which.... doesn't help on finding minor suit fits.

Glen continues with a method that seems to potentially lose not only minor suit fits when responder is balanced, but also to lose heart fits.

Then Akhare suggests wasting 2NT to show a game-forcing raise (only! not even limit-plus!) when we are already playing 2♣ as a GF relay.

My suggestion was something simple like:

2♦ = 4+♦ and forcing one round, at least invitational
..... 2♥ = natural, 4+♥
..... 2♠ = minimum, denies a four-card red suit
..... 2N = game values, balanced or a weak 6+ card spade suit
..... 3♣ = game values with a black two-suiter
..... 3♦ = 4+♦, minimum but perhaps there is a big fit
..... 3♥ = strong diamond raise (GF)
..... 3♠ = GF with a strong 6+ card spade suit

The 2♦ bidder will have one of:

(1) a balanced invite with diamonds, like 2344 or 2353 or 2443. This hand passes 2♠, bids 2♠ or 3♥ over 2♥, bids 2♠ (doubleton) over 2N, bids 3♥ or 3N over 3♣ (depending on heart stopper), passes 3♦, tries 3♠ over 3♥ (doubleton).

(2) a three-suited invite with diamonds and short spades. This hand bids 2NT over 2♠, 2NT or 4♥ over 2♥, 3NT over 2NT, otherwise basically as above.

(3) a diamond one-suited invite. This hand normally rebids 3♦ at next turn, or looks for diamond game/slam over a diamond raise (obviously a 6+♦/4♥ hand raises hearts if opener bids 2♥ but is otherwise pretty much the same)

(4) a club/diamond two-suited invite. this hand rebids 3♣ at next turn if possible, or otherwise cooperates with diamond raises

(5) a pure 3♠/5+♦ sort of GF raise. this hand normally bids 4♠ at next turn, with the idea that this helps opener with slam prospects more than trying to relay the hand.

[akhare]

awm, on 2011-March-09, 11:07, said:

Then Akhare suggests wasting 2NT to show a game-forcing raise (only! not even limit-plus!) when we are already playing 2♣ as a GF relay.

The point is taken, but the only reason to "waste" 2N when 2♣ GF is available is to handle preemption by fourth hand. The idea is that after 1M - 2N - (4♠ / 5m), opener might be better placed to make a decision knowing that we have at least a 9 card trump fit.

Granted, 2N doesn't have to be GF, but do you think that all hands with slam interest should go via 2♣ regardless of combined trump holding?

[glen]

awm, on 2011-March-09, 11:07, said:

... Glen continues with a method that seems to potentially lose not only minor suit fits when responder is balanced, but also to lose heart fits. ...

Actually my recommendation was to use 1NT semi-forcing with GI balanced, which finds the second suit fits

[straube]

awm, on 2011-March-09, 11:07, said:

Straube seemed to indicate that he sees the point about finding minor suit fits, but then suggest a method which.... doesn't help on finding minor suit fits.

Well, we did move quite a bit in that direction. We placed all the 2-suited (5/4) responding hands into 1N to increase the odds of finding those fits. We did lose diamond fits when opener had four diamonds which was insufficient to raise diamonds. We did find club fits when we bid 1N with Ax Ax Qxxx Qxxxx. How would you respond with this?

The structure solved the 13 opposite 12 dilemma more frequently.

I like your structure. I'd like for opener to be able to distinguish good 5S/4H from bad and good 5S/5H from bad. How would you go about that? I think 1S-2D, 3H is needed to show a strong 5/5. Maybe 1S-2D, 3S is a strong raise and 1S-2D, 3N is a jump rebid of spades. I don't know.

Btw, what do you think of our 1S-2H as GI six hearts?

[straube]

awm, on 2011-March-09, 11:07, said:

The 2♦ bidder will have one of:

(1) a balanced invite with diamonds, like 2344 or 2353 or 2443. This hand passes 2♠, bids 2♠ or 3♥ over 2♥, bids 2♠ (doubleton) over 2N, bids 3♥ or 3N over 3♣ (depending on heart stopper), passes 3♦, tries 3♠ over 3♥ (doubleton).

(2) a three-suited invite with diamonds and short spades. This hand bids 2NT over 2♠, 2NT or 4♥ over 2♥, 3NT over 2NT, otherwise basically as above.

(3) a diamond one-suited invite. This hand normally rebids 3♦ at next turn, or looks for diamond game/slam over a diamond raise (obviously a 6+♦/4♥ hand raises hearts if opener bids 2♥ but is otherwise pretty much the same)

(4) a club/diamond two-suited invite. this hand rebids 3♣ at next turn if possible, or otherwise cooperates with diamond raises

(5) a pure 3♠/5+♦ sort of GF raise. this hand normally bids 4♠ at next turn, with the idea that this helps opener with slam prospects more than trying to relay the hand.

I really do appreciate your help, but I have concerns about this.

1) is not so much of a problem. At worst we'll miss a club fit now and then when we have 2344. My last structure misses that all the time. Probably the worst thing is when partner rebids 2H and his strength is ambiguous and I have to raise. That is a loss though.

2) I want to respond 1N on these hands very badly. I don't have any direction on this. I'm usually happy if partner passes 1N and I get to play this. I'm happy bidding 1N and raising clubs if partner has them.

3) diamond suit invites are problematic for both structures. It's nice to show diamonds, but it's also nice for opener to show a sixth spade when we have two spades. If we respond 1N, we'll sometimes learn this. If we respond 2D, we often won't. 1S-2D, 2S-3D and now we're in a 6-0 diamond fit instead of a 6-2 spade fit. We solved this for clubs by responding 1N first and then agreeing that a rebid of 3C promised the GI sort (partner will raise spades when opener rebids them); we miss spade fits on the weak side, but we don't miss our game and 1S-3C is an optional bid.

I'm almost thinking of 2D as...

1) promising 4 diamonds, could have 4 or 5 hearts
2) denying 4 clubs unless 5/5 (planning to show them later)
3) denying 6D/2S. These hands would start with 1N and then decide later what to do.
4) promising 2S unless 5D/5D or 6D

My concern is that I'm increasingly making use of 2D restrictive, but the bid is quite preemptive to our auctions. What do you think?

[awm]

There is some issue about what you're comparing with. To address the concerns:

(1) The main advantage on these hands over bidding 1NT forcing is that you can now play in 2♠ when opener is minimum. You seem to think this is advantageous, considering that your entire "two-card limit raise" idea was based on this fact. However, unlike the "two-card limit raise" 2♦ bid, you still have the opportunity to find a 4-4 or 5-4 diamond fit. This method seems strictly better than making a "two card limit raise" on these particular hands.

(2) You could bid 1NT forcing on these if you want; however my sequence has several advantages over that. Partner will often know about the singleton spade (for example 1♠-2♦-2♠-2NT shows a singleton spade whereas 1♠-1NT-2♣-2NT could easily be a doubleton spade. This may help when opener has a minimum 6-4 (for example). Raising clubs isn't really an option for these hands if you play a forcing/semi-forcing notrump (i.e. 1♠-1NT-2♣ could be three anyway) and responder could be 1453 in any case. Taking these hands out of 1NT creates some interesting inferences about the 1NT...2NT sequence that can help you find club fits on occasion too.

(3) Obviously if you have 2♠-6♦ there will be a decision about whether to pass after 1♠-2♦-2♠, or continue with 3♦. However, consider what would happen playing the other methods available. If you make a "two-card limit raise" with these hands then you could miss a ten or eleven card diamond fit which is obviously horrible! If you bid 1NT, then when partner rebids 2♣ or 2♥ you will not have the option of playing in 2♠ at all (because correcting would be a weaker hand) and will have to move to the possibly 6-0 diamond fit. Even if you "get lucky" and bid 1♠-1NT-2♠, you will have to play at the three-level if partner's not bidding game, and partner won't know to up-evaluate diamond length and down-evaluate diamond shortage (since you never showed your diamonds). Heck, even 1♠-1NT-2♦ is non-trivial, since bidding 3♦ now undersells the hand and you have rather a lack of forcing descriptive bids.

[straube]

Just for clarity, the three-suited hand short in spades is either 1-4-4-4 or 0-4-(54), not 1-4-5-3, right? Opener on occasion will want to rebid 3C over 2N to play.

I see the attractiveness of your proposal and I'm not bothered so much by guessing with the 2S/6D hand.

I would like to be able to sort out the heart hands though. Can anything be done? Having 1S-2D, 2H as nf was nice but I could give that up. There's still the issue of sorting out the good 5/5s from the bad 5/5s.

[straube]

1S-2D

.....P-10-11 with 4 diamonds, yes we'll lose occasional 5m, I don't like not announcing a fit but if it makes the rest of the structure more playable...
.....2H-10-11, 4H
.....2S-10-11, all else
..........P-12-13, 2 spades
..........2N-12-13, 1444
...............3C-fit
..........3C-5/5 minors
..........3D-6 diamonds, may have two spades
.....2N-ask, 12+
..........3C-5/5 minors, technically nf but rare to pass
...............3D-preference, technically nf
..........3D-6 minor, GI
..........3H-4H, 2S
..........3S-2S
..........3N-1444
.....3C-5 clubs, forcing
.....3D-fit, 12-15
.....3H-5 hearts
.....3S-5143
.....3N-5341

[straube]

1S-2D

.....P-10-11
.....2H-10-11, 4H
.....2S-10-11, all else
..........P-12-13, 2 spades
..........2N-12-13, 1444
...............3C-fit
..........3C-5/5 minors
..........3D-6 diamonds, may have two spades
.....2N-GF ask, 12+, may conceal a diamond fit
..........3C-5/5 minors, GF
..........3D-6 minor, GF
..........3H-4H, 2S
..........3S-2S
..........3N-1444
.....3C-5 clubs, forcing
.....3D-fit, 12
.....3H-5 hearts, forcing
.....3S-diamond fit, heart shortness
.....3N-club fit, heart shortness

[awm]

I don't think you can sort out the heart hands the way you describe.

The main issue is the 5-5 majors hands. These hands have the properties that:

(1) They can be quite light in values, since usually this is one of the most effective patterns for a light 1M opening. This makes it unappealing to force game on these hands opposite an invite without a known fit as you seem to be doing in your current structure. It's easy to imagine winding up in 3NT holding a 5512 9-count opposite a 2263 11-count, and there's no reason to think a 20 hcp misfitting game will have any hope to make.

(2) They are powerful when there is a nine-card fit. The alternative approach of bidding 2♥ on any minimum loses out here; if you bid 2♥ on your 5512 9-count, you could easily make game opposite partner's 2443 12-count (for example) based on the big heart fit. Of course, you could say "I will just never pass 2♥ with four-card support" but now you are getting closer to the situation where 2♥ may as well be forcing anyway.

I don't really see a big issue with being forced to the three-level on a 4-4 fit and "almost-game" values; this will very rarely cost you (and the whole field has the same "issue").

[straube]

awm, on 2011-March-11, 12:43, said:

I don't think you can sort out the heart hands the way you describe.

The main issue is the 5-5 majors hands. These hands have the properties that:

(1) They can be quite light in values, since usually this is one of the most effective patterns for a light 1M opening. This makes it unappealing to force game on these hands opposite an invite without a known fit as you seem to be doing in your current structure. It's easy to imagine winding up in 3NT holding a 5512 9-count opposite a 2263 11-count, and there's no reason to think a 20 hcp misfitting game will have any hope to make.

(2) They are powerful when there is a nine-card fit. The alternative approach of bidding 2♥ on any minimum loses out here; if you bid 2♥ on your 5512 9-count, you could easily make game opposite partner's 2443 12-count (for example) based on the big heart fit. Of course, you could say "I will just never pass 2♥ with four-card support" but now you are getting closer to the situation where 2♥ may as well be forcing anyway.

I don't really see a big issue with being forced to the three-level on a 4-4 fit and "almost-game" values; this will very rarely cost you (and the whole field has the same "issue").

Makes sense to me. I think I'm willing to risk losing 5D (passing 2D) but I don't want to lose heart games. Do you like this yet?

1S-2D

.....P-10-11
.....2H-four+ hearts, forcing (could be 5/5 minimum)
.....2S-10-11, all else
..........P-12-13, 2 spades
..........2N-12-13, 1444
...............3C-fit
..........3C-5/5 minors
..........3D-6 diamonds, may have two spades
.....2N-GF ask, 12+, may conceal a diamond fit
..........3C-5/5 minors, GF
..........3D-6 minor, GF
..........3H-2542
..........3S-2S
..........3N-1444
.....3C-5 clubs, forcing
.....3D-fit, 12
.....3H-5 hearts, GF
.....3S-diamond fit, heart shortness
.....3N-club fit, heart shortness

[Zelandakh]

Have you considered "playing the game" and making all of your 2-level responses natural but also part of a relay system? Perhaps 1S - 2C = GF with 3+ clubs; 1S - 2D = GF with 3+ diamonds and 0-2 clubs; 1S - 2H = GF with 5+ hearts, 0-2 clubs, 0-2 diamonds. Your 1S - 2D relays are now 1 step higher than normal - perhaps relay breaks to take advantage of the known lack of club length can pick up the slack. After 1S - 2H you could have Opener become Captain since more is known about Responder's hand, or just play normal 2/1 methods with Responder's hand better defined than usual. This is probably the simplest solution and requires little work to make the adjustments.

Another alternative would be to switch to a 1NT relay but include a single hand type below invitational strength, perhaps weak with long clubs. Then 1S - 1NT - 2C = "most minimums" can be passed and you can begin proper relays on the next round. This method would have an issue with invitational hands with long clubs so it might also be necessary to give up a sequence like 1S - 3C for this purpose. This is probably more work than you are willing to do for a few tournaments but is at least a way of preserving the single relay route while staying legal.

Note also that the other suggestion of using 1S - 2D as a general invite in combination with 2C GFR is a reasonably common idea; I certainly saw it a couple of times when looking at the relay systems out there before working on one of my own. Finally, the 2D response in Double-Barelled Stayman is a relay system. What else could you call it? It does seem inconsistent to allow this GF relay method and not something simialr after 1S.

[straube]

Zelandakh, on 2011-March-12, 12:13, said:

Have you considered "playing the game" and making all of your 2-level responses natural but also part of a relay system? Perhaps 1S - 2C = GF with 3+ clubs; 1S - 2D = GF with 3+ diamonds and 0-2 clubs; 1S - 2H = GF with 5+ hearts, 0-2 clubs, 0-2 diamonds. Your 1S - 2D relays are now 1 step higher than normal - perhaps relay breaks to take advantage of the known lack of club length can pick up the slack. After 1S - 2H you could have Opener become Captain since more is known about Responder's hand, or just play normal 2/1 methods with Responder's hand better defined than usual. This is probably the simplest solution and requires little work to make the adjustments.

Another alternative would be to switch to a 1NT relay but include a single hand type below invitational strength, perhaps weak with long clubs. Then 1S - 1NT - 2C = "most minimums" can be passed and you can begin proper relays on the next round. This method would have an issue with invitational hands with long clubs so it might also be necessary to give up a sequence like 1S - 3C for this purpose. This is probably more work than you are willing to do for a few tournaments but is at least a way of preserving the single relay route while staying legal.

Note also that the other suggestion of using 1S - 2D as a general invite in combination with 2C GFR is a reasonably common idea; I certainly saw it a couple of times when looking at the relay systems out there before working on one of my own. Finally, the 2D response in Double-Barelled Stayman is a relay system. What else could you call it? It does seem inconsistent to allow this GF relay method and not something simialr after 1S.

Our 1S-2C is +1 on average compared to standard symmetric. For 1S-2D to be a relay, we would be +2. It's unmanageable. We tried to collapse a few shapes to be +1 most of the time, but it was a memory load and suffered on the collapsed shapes. 1S-2H as a reverse relay sounds promising but then the limited hand is captain and doesn't know whether to look for slam.

We think that 1S-3C should be to play. Let the GI hands with a club suit bid 1N semiforcing and then bid 3C unless a 6-2 spade fit or 4-4 heart fit is discovered.

I'd consider 1S-2D as a general invite but it isn't GCC legal. For Midchart it might also make sense, but using it to show hearts is very attractive in concert with 1S-2H as a constructive raise.