[Siegmund]

I was debating whether or not to post this when the Gazzilli and Kaplan Inversion thread went up. Something I've been experimenting with, in the context of a system where the 1M openings include weak balanced hands, and opener is expected to respond 1S to 1H with most hands with 4 spades (no responding 2D on 4-2-4-3 hands to set up a force, etc.)

In such a system, opener has a lot of things he wants to do: get out at 1NT if he's weak and balanced and there is no major fit; find 5-3 spade fits if they exist (and, if he's weak and semibalaned, not be too afraid of 4-3 spade fits and 5-2 heart fits); still be able to show minimum, medium, or maximum hands in either major; and cope with the 3S6H version of the BW Death Hand.

Here is what I am trying for opener's rebids:
1NT = 11-14ish, natural, including all the 5332s (responder has ways to ask for 3 spades if he is INV+) and some of the 2-5-2-4s and 1-5-3-4s, but not 3-5-1-4s.
2C = 4+ diamonds, unlimited.
2D = 5+ clubs or 3-5-1-4, unlimited.


Over 2C, responder completes the transfer if he's willing to sit for 2D (and opener with a weak 3-5-4-1 corrects to 2S over 2D); can rebid 2H with 2 or 2S with 6, or a natural 2NT with inv values, and the usual array of inv or strong 3-level bids. Over 2D, responder goes farther out of his way to rebid 2H when he would rather hear a preference than go to 3C (for instance a weak 4-2-5-2).

Meanwhile, opener is able to put all of his 5-5s, weak and strong, into the transfer rebids, as well as the strong 6-4s (transfer then rebid hearts) . The immediate 3C and 3D rebids are freed up to take care artificially of hands that are hard to bid naturally. My tentative plan here:

1H-1S-3C: game-going hands with exactly 3 spades: 3-5-3-2/3-5-3-2, and 3-6-3-1/3-6-1-3.
1H-1S-3D: game-going hands with 6+ hearts and not 3 spades - just like a 1H-1S-3H bid but stronger, leaving room to explore for slam.
Now 1H-1S-3NT can promise 2-5-3-3. Opener's 2M and 3M rebids as in standard. Opener's splinters in support of spades as in standard.

The point of the transfers is sequence creation, not making responder declarer in minor-suit contracts, but when opener has the 11-count, you aren't wrongsiding the contract at all -- in fact you are making the hand in which 9 cards are already described dummy, and keeping the less well defined hand hidden. And it avoids completely wasting a round of the bidding on non-descriptive relays as Gazzilli's 2C by opener-then-2D-by-responder structure does.

There is much less need of such a structure over 1H-1NT since you can get away with 2S as a semi-artificial rebid to handle awkward strong hands, not needing so have all of 2H,2S,3H,3S natural and nonforcing.
Over 1S-1NTF, it's not 100% clear to me whether to include all 3 unbid suits in the transfer scheme, or just swap the minors.

Thoughts are welcome.

[Free]

Looking at Gazzilli, you have 2 major advantages:
- strong hands stay low, you can even use 1M-1NT-2♣-2♦-2OM artificial to combine several hand types).
- when responder is weak with misfit, you can play in responder's suit (which will be better in many cases).
The biggest disadvantages are some specific distributions, and if you play a (semi)forcing NT you might lose the ♣ suit. Another one is 1♥-1♠ auctions, which aren't always easy in my experience. So while I don't think there's much improvement possible after 1M-1NT auctions, 1♥-1♠ can do better. I'm glad you're looking for something better after this exact sequence.

The idea you present has some interesting features. You try to find a 4-3 or 5-3 ♠ fit which is definitely a plus. I just wonder why you don't just raise on a 3 card with suitable hands. The 3=5=1=4 hand for example, you know you want to play 2♠ if responder is weak, so why go through 2♦ hoping partner won't bid 3♣ on a 4=1=5=3 (admitted, it's a very specific situation )?

I'm not a fan of how you handle strong hands. Responder has to bid very high without knowing anything about opener's strength. In Gazzilli opener already showed he has <16HCP before responder shows his extras (1♥-1♠-2♣-2♦-2♥-...), or he knows partner has 16+HCP and we're already in a GF auction with probably more information (1♥-1♠-2♣-2♦-... shows opener's strong hand, his shape and his ♠ support). You need to go to 3-level to show/deny some ♠ support with strong hands and responder can't show a longer minor.

Also I don't think responder can't play in his minor when he holds 4♠-6m and a weak hand. Again this is quite rare and we expect a competitive auction, but it happens from time to time and Gazzilli is ready for this.

Quote

And it avoids completely wasting a round of the bidding on non-descriptive relays as Gazzilli's 2C by opener-then-2D-by-responder structure does

The so called non-descriptive relays are also "sequence creation"...
After the 2♦ positive you get one of the best sequences available: 2♠ makes the auction GF and shows a 3 card support, 2NT/3m/♥ also makes the auction GF, is natural, and denies 3♠, and 3♠ shows a 4 card support in a GF context -> you can just start cuebidding. There's no need to jump to 3NT with the 2=5=3=3, and responder can set any suit at 3-level.
After any other bid than 2♦ positive, the auction usually ends in 2M, but without fit it ends in a contract played by responder. The weak hand gets to play his suit, making sure the weak hand gets more playing strength while the opener still has his HCP strength to help.
Any other rebid by opener except 2♣ and 2NT show hands with less than 16HCP. You gain some advantages precision players have (1M-1NT-3m shows a good 5-5 for example).

While I don't want to crush the idea because it might have potential, I fail to see the advantages of this scheme over Gazzilli. Perhaps some example hands and comparisons may be useful.

[semeai]

Interesting setup. Let's go through the hand types and advantages/disadvantages:

On hands that would bid a NF 2♦, the systems are identical.

On hands that want to force 1 round with diamonds or with clubs, yours does better when responder is weak, especially with 5♠-1♥ and 4 of your minor. You'll usually do worse on hands where responder would bid 2♦ over the Gazzilli 2♣ bid, though, I believe.

On weak 3-5-1-4 hands for opener: You're at a slight advantage. I'm happy raising immediately to 2♠ on these, though, but if that isn't your style I understand.

Others: Gazzilli can already handle the bridge world death hand (though not with one bid). You lose the 3m jumps to show big distributional hands. These will be slightly worse going through 2m.

I'm going to guess that Gazzilli is preferable at imps, but yours has merit and may be desirable at matchpoints.

[Siegmund]

Glad to hear it described as better-than-insane, anyway

Yes there is potentially room to accommodate responder's weak 4-6es, and raising immediately with the weakest 3541/3514 hands is an option.

It's not so much that I believe its "better than Gazzilli," as me having a hard time believing the "clubs or not clubs" nature of the 2C rebid is optimal -- and me having a real dislike of the way 2C-then-2D sequences get used in XYZ / 2way NMF, so I am not predisposed to wanting another "2C is artificial and the 2D reply to it is also artificial" auction in my system to keep memory problems simpler. (Actually I had a need for 2C=diamonds and 2D=clubs in a few other auctions in the same system; otherwise I would have never thought to try it after 1H-1S.)

[benlessard]

We are playing transfers rebid but our 1S is a relay but doesnt show S.

If 1S showed S i would probably play something like...

1Nt = 4C or any Non GF hands with 3S or any 15-17 hands partner is expected to bid 2C with 3C and 2H
2C= 4D min or GF
2D= 6H min or GF
2H = min or GF S raise
2S medium S raise
2Nt and higher jumpshift with exactly 3S.

1Nt followed by a high level bid would show jumpshift with H/C without 3S.

With a 2533 you open a weak nt or a strong club if your system allow it otherwise you treat it as H+C. Sacrificing the ability to stop in 1Nt (rather than sacrificing 2C) is a very small price to pay for having "perfect jumpshifts"+ a way to show strenght.

The main problem i have with this setup is that after
1H-1S
2C-???

responder has to bid 3C to GF wich is too high for relaying perfect shapes.