[antonylee]

Hi,
In one of my current partnerships (playing some classic 2/1 structure) we play 1m-2M as an intermediate jump shift, ~good 8 to bad 12 with 6+M. While we were discussing continuations, which we play as natural for now, I realized it may be a good opportunity to play some shape asking relays here, with 2(M+1) as the GF relay, perhaps promising 2+ in the suit for some safety. I have never played any kind of shape relays but if I understand well, we are two steps above e.g. symmetric relayers after e.g. 1♣-2♠-2N!-... Yes, this means that if we end up playing in that suit (which is likely), declarer's shape and honor location will have been revealed; on the other hand probably most other scientific schemes here will also involve 2N as some sort of feature ask so they will suffer from the same problem.
Any thoughts on whether this is playable?

[Kungsgeten]

Yes, you are two steps higher than symmetric relay. In symmetric relay though responder, holding a single suiter, has 5+ cards in the major and no 4+ side suit. The way I play it, this means 5332, 6322, 6331, 7+ cards and a singleton, 7+ cards and a void. With some extreme distributions you have to lie.

The thing you would have to consider, if you plan know responder's full distribution, is what kind of hands responder can hold. Can responder have more than 6 cards in the major? Can responder have a side suit, and if so can it be any side suit (other major, other minor, the minor opened)?

Let's say that responder can not hold a side suit, and not more than 6 cards in the major. This leaves only two distributions: 6322 and 6331. You could then keep things pretty simple:

1m-2H; 2S--
2N = Any 6322
..3C = relay
....3D = 2-6-3-2
....3H = 2-6-2-3
....3S = 3-6-2-2
3C = 3-6-3-1
3D = 3-6-1-3
3H = 1-6-3-3

1m-2S; 2N--
3C = Any 6322
..3D = Relay
....3H = 6-3-2-2
....3S = 6-2-2-3
....3N = 6-2-3-2
3D = 6-3-1-3
3H = 6-1-3-3
3S = 6-3-3-1

The above is just an example ofcourse. You could also zoom into showing controls or whatever at the last step. Over the 6322 response an alternative might be to use the relay + 1 as a direct control asking bid (or perhaps RKC or whatever), and pretend responder has 6333.

If responder could have a four card side suit things get a bit trickier. You then have to determine which suit is the side suit, and then which suit is short, and then if it is 6421 or 6430.

[Kungsgeten]

Another alternative would be to use 2NT as game forcing Stenbergs/Jacoby (where 2NT shows game forcing values and 2+ support in your case). Here's a popular variant in Sweden (some minor modifications to fit into your sequence):

1m-2M; 2NT--
3C = Minimum, no void
..3D = Slam interest, asking for shortness
....3H = singleton clubs
....3S = singleton diamonds
....3N = singleton other major
....4X = 6332, cue bid
..3H = Slam interest, opener has singleton in other minor
..3S = Slam interest, opener has singleton in other major
..3N = Suggestion to play
..4X = Cue bid, not interested in singleton (4M is to play)
3D = Max, no shortness (so any 6322)
..3H = Slam interest, singleton other minor
..3S = Slam interest, singleton other major
..3N = Suggestion to play
..4X = Cue bid (4M is to play)
3H = Max, singleton clubs. 4X are cue bids.
3S = Max, singleton diamonds. 4X are cue bids.
3N = Max, singleton other major. 4X are cue bids.
4C = Void clubs
4D = Void diamonds
4H = Void other major

[Zelandakh]

antonylee, on 2013-June-09, 23:48, said:

I have never played any kind of shape relays but if I understand well, we are two steps above e.g. symmetric relayers

It is actually worse than this since you might still have a 4 card side suit, whereas these can be spun off into the 1.5 suiters for symmetric. I think you would be better off using some sort of weak 2-style structure, perhaps 2NT = size/feature ask; 3♦ = shortage ask; 3♥ = good spade raise, or whatever. There are some other clever things you can do but it is probably not worth it; in the same way, adding relays only for a rare sequence without using them more generally is not really worth it.

[rbforster]

Shape and honor location relays also work better when your describing the dummy instead of declarer, giving away less useful info. Not sure if transfer jumps could fit into OPs scheme, or if they're meant to be partially preemptive.