[newchemist]

Assume that you are playing symmetric relay or similiar where 1M - 1NT is an artificial gameforcing relay, and 1M is limited to (11)12-15pts

1. What are the possible schemes for 2-level responses, and what are the advantages/disadvantages of each? For example i think viking club uses 1M-2C as artificial, game-invite while other bids are presumably natural.
2. Since weak-ish or mildly invitational hands can no longer bid a waiting 1NT in response to the opening bid, wouldn't there be a large risk of partnership missing some <25HCP distributional games or failing to find good partscores on misfit hands? What can be done about this?

Of course one could shift the game-forcing relay to 2C and keep the 1NT response as a waiting bid, but the loss of a step really hurts the relay.

[hrothgar]

newchemist, on 2011-October-02, 02:27, said:

Assume that you are playing symmetric relay or similiar where 1M - 1NT is an artificial gameforcing relay, and 1M is limited to (11)12-15pts

The simple answer is don't do this.

It much more logical to play that step is an artificial game invitational relay

[Flameous]

http://toohighagain....t-gf-relay.html

My take on the subject. I think there is lot's of fine tuning to be done as I haven't really played the structure.

[newchemist]

Flameous, on 2011-October-02, 09:37, said:

http://toohighagain....t-gf-relay.html

My take on the subject. I think there is lot's of fine tuning to be done as I haven't really played the structure.

I like this set of continuations alot. I assume responses are similiar after 1H-2C, i.e. 2D shows most min hands while max hands make a natural rebid.
But what would the sequence 1H-2C-2D-2S show?

My suggestion is that to make the system exactly the same after 1H or 1S (so as to reduce memory load)
after 1H/S - 2C - 2D,

2H: Balanced/Semibalanced 8-9 HCP
2S: Balanced/Semibalanced 10-11 HCP

[Zelandakh]

I play 1H - 1S and 1S - 1NT as INV+ relays, so 1H - 1S - 1NT - 2C and 1S - 1NT - 2C - 2D are GF relays where Opener has denied 4 cards of the other major. The follow-ups are then pretty simple, 1st step relates to clubs, 2nd step (ie rebidding your suit) shows a 1-suiter and everything else shows diamonds. For example:-

1H - 1S; 1NT - 2C
==================
2D = 4+ clubs
2H = 6+ hearts, 1-suited
2S = 5+ diamonds
2N = 5 hearts, 4 diamonds
3C = 6 hearts, 4 diamonds, 0-1 spades
3D = 2641
3H = 3640
3S = 1741
etc

Obviously, you will use symmetric rather than this but that is secondary. If you feel your range is too large to handle without including a min-max in your relays then this approach is very efficient. If you are comfortable with the 4-5 point range then a purely GF relay is probably better for you.

There are some solutions to the issue you raise with invitational hands. Using the INV+ relay method you just relay once and then bid naturally. Other systems use an artificial bid for many of the invitational hands. As I recall TOSR does this and is worth looking at if you want to model something that is known to work.

[akhare]

newchemist, on 2011-October-02, 02:27, said:

Of course one could shift the game-forcing relay to 2C and keep the 1NT response as a waiting bid, but the loss of a step really hurts the relay.

I don't think the 2C is a big less since you don't need a step to show the reverser. Pard and I play the following symmetric structure and it works very well. You can follow this up with your favourite QP ask / DCB etc.

1M - 2C (artifical GF)
.....2D (various shapes):
...........2S: Short legged with exactly 4 clubs <follow short legged template>
...........2N: LL with 5+ clubs <follow LL template>
...........3C: 5332 shapes
...........3D+: LL with diamonds, high short <follow LL template>
......2H: 6M <Follow single suited template>
......2S: Short legged with exactly 4 diamonds <follow short legged template>
......2N: 5/5 or better with OM <follow LL template>
<Begin 2 suited short legged template>
......3C: 4 OM, high short
......3D: 4 OM, 5422 exactly
......3H: 4 OM, low short, 5431
......3S: 4OM, low short, 6421
......3N: 4OM, low short 6430
......4C: 4OM, low short 7411

<Begin 2 suited long legged template>
...........3D: LL with OM, high short
...........3H: LL with OM, even short
...............3N: 5611
...............4C: 6511
...........3S: LL with OM, 5521, low short
...........3N: LL with OM, 6520, low short

Begin single suited template
............2N: No short, 6322 / 7222
...................3D: 6223
...................3H: 6232
...................3S: 6322
...................3N: 7222
............3C: High short
............3D: Mid short
............3H: 6331
............3S: 7321
............3N: 7330

[rbforster]

akhare, on 2011-October-02, 23:36, said:

<Begin 2 suited long legged template>
...........3D: LL with OM, high short
...........3H: LL with OM, even short
...............3N: 5611
...............4C: 6511
...........3S: LL with OM, 5521, low short
...........3N: LL with OM, 6520, low short

Don't you have 5530 in the above template somewhere, maybe at 3N and 6520 bumped up to 4C?

[Flameous]

akhare

Where are your 544 shapes?

I don't think playing inv+ relay really solves any of the real issues. You still need ways to bid shapely invites since simple min/max hardly does any good there.
Sure you get your balanced and semibalanced hands out but they were never the problem in the first place. I like to have min/max differentiation in my relays but
it seems inefficient to have it come first. Now I started thinking of having it after hand's nature is shown. Essentially this is just adding "minimum" as first
step to each of akhare's templates. If anyone has tested this, I'd like to hear how it worked out.

I think the practical best is to play 1S relay over 1H and 2C over 1S and open all major shapes with 1H. It ain't pretty but it's workable. And if you throw in
a reverse flannery opening, I think you have yourself pretty good set of relays.

[akhare]

Flameous, on 2011-October-03, 15:23, said:

akhare

Where are your 544 shapes?

As Rob and others have pointed out, there's some room for proof reading .

Regarding 5440, you either have the choice of treating it as 5431 or tacking it at the end (probably before 7411)..

[Zelandakh]

I disagree Flame. The advantage of having strength come at the beginning is that your invitational hands can now relay opposite a maximum and that you can sign off much earlier since you do not have to keep relaying just in case Opener is stronger, thus avoiding excess information-leakage, the big disadvantage of the method. Bidding shapely invites is simpler than in standard since you have already limited your hand and thus can bid your shape without partner thinking you want to force the bidding. You also gain the secondary advantage that your simple new suit bids become highly limited since they do not include any invitational hands.

In my own system I use both methods due to necessity. After 1S - 1N (relay); 2D (hearts) it is simpler to use 2H as a general relay so as to stay in symmetric levels. This is not so bad with both majors but the auctions are nonetheless occasionally not as smooth as those following a 2C rebid (min without hearts). Ideally strength-showing should always come first in these auctions imho since we are focused primarily on games rather than slams.

[newchemist]

I think the invi+ relay is a good idea. Another discussion for another thread

But here I am more interested in how the 2-level responses should work assuming 1NT is a 100% GF. Whether or not a 100% GF 1NT is a good idea is another problem altogether.

Logically speaking it seems that unlike standard 2/1, here you only have 1 gme-forcing response and everything else is free for weak/invi hands so you ought to have better definition on weak/invi hands if only a sensible response scheme can be found.

On an unrelated note, I was under the impression that a 2C relay is mathematicaly too high for full shape relays.
Also I think having 1M-1NT as a 100% game force does allow you more freedom with relay-breaks, run-on sequences etc etc so dont underestimate it!

[nigel_k]

If you want to keep it as close to symmetric as possible, after a 1S opening do the following:

2C-2D-2H: Either 5/5 majors or three suited (then after 2S, 2NT is 3 suited and 3C+ is 5/5 majors)
2C-2D-2S+: Spades and clubs (2S=5/5, 2NT+=5+/4 etc)
2D-2H-2S+: Spades and diamonds
2H-2S-2NT+: 4 hearts (and longer spades)
2S+: single suited

Similar after a 1H opening except that 5/5 majors is out. Obviously you can make changes so that the relayer ends up declarer more often (e.g. 2C=diamonds or other types, 2D=hearts, 2H=clubs). But my personal preference is to get back to 'standard' symmetric as early as possible because my memory is not that good.

The above can also be used if 1NT is invitational or better but in that case I would try to organise it so that opener doesn't ever bid 2♠ with 5332 because those are the hands where responder will want to invite with 2NT.