[PrecisionL]

We are going to a two-tier structure for the 4441 positive responses to 1♣  as Bary Rigal suggested in Precision in the 90s (Advanced Section) and I wonder if anyone has any experience with something like:

1♣ - 3♦ = 4=1=4=4 and 8-11 hcp
1♣ - 3♥ = 1=4=4=4 and 8-11 hcp
1♣ - 3♠ = 4=1=4=4 and 12+ hcp with 4+ Controls
1♣ - 3NT = 1=4=4=4 and 12+ hcp with 4+ Controls

Minor suit singletons Tier I & II are compacted into the 1NT response.

Larry

Edit 3/23/09:

We play transfer positive responses but 5332 and 4441 are treated differently to keep the responses 'pure.'

The URL below is now 2 years old, we need to update once the above issue is resolved.

[awm]

I'd recommend reversing these. Advantages:

3NT would show a limited hand, and therefore be passable.

You have more room to set the suit and ask cuebids/bid keycard over the cheaper options, especially when a major is to be trump. The space seems more useful opposite the stronger (more slammish) hand types.

One could sort things out over 8-11 via relays fairly easily, since you are limited both in values and controls, so it is feasible to use "4♣ = slam try relay/control ask" over these after a 3♠/3NT reply. Doing the same for 12+ is much more awkward, as there is a wider range of controls and especially because there may be enough extras for responder to want to force a cuebid.

[shevek]

Forget about devoting 4 bids to the pure 3-suiters, which are 4% of hand patterns.
Maybe even bid 1♥/1♠ with these and unravel, or throw them all in the same box. Whatever you ultimately choose has to be better.

[david_c]

shevek, on Mar 23 2009, 07:23 AM, said:

Forget about devoting 4 bids to the pure 3-suiters, which are 4% of hand patterns.

Surely you want your 3-level bids to be very specific. Partner is unlimited in strength and shape, so you don't want to go around pre-empting him unless you have something very descriptive to say.

[shevek]

david_c, on Mar 23 2009, 03:58 AM, said:

shevek, on Mar 23 2009, 07:23 AM, said:

Forget about devoting 4 bids to the pure 3-suiters, which are 4% of hand patterns.

Surely you want your 3-level bids to be very specific. Partner is unlimited in strength and shape, so you don't want to go around pre-empting him unless you have something very descriptive to say.

Agreed that symmetric relay would be an answer but if you can't find a copy of that, how about 3x with KQJxxxx & out or similar, some one loser suit or 2 of top three, etc? That would be useful.

[ansoe]

Can I suggest you consider adopting transfer positive responses. If you do then a response of one spade shows positive values with no 5 card or longer suit. Opener rebids one no trump which asks responder for more information. With all 4441 hands responder will now bid two clubs over which opener relays with two diamonds and responder bids the suit below the singleton. If opener now bids the short suit it can be used to ask the range of the responders hand, or number of controls.

This treatment has the added advantage that NT contracts are usually played by opener and openers rebid of 1NT is unlimited.

[TylerE]

I'm a fan of the Berkowitz/Cohen structure here:

3♣ = 4441, 8-11 or stronger but quacky (e.g. not 4 controls), black stiff, 3♦ asks, then 3♥ = ♣
3♦ = 4441, 8-11 or stronger but quacky (e.g. not 4 controls), red stiff, 3♥ asks, then 3♠ = ♦
3♥ = 1=4=4=4, 12+/4 controls+
3♠ = A solid 6+ suit, no outside controls.
3N = 4=4=4=1 12+/4+
4C = 4=4=1=4 12+/4+
4D = 4=1=4=4 12+/4+

[keylime]

Larry and I have been wrestling with this for a while now, because of the nature of our system. We already use transfer positives in our structure, and rather like the overall structure that we have now. This handtype tho, is the one that can cause some troubles.

[ASkolnick]

We also use the Berkowitz structure for the unusual positives.

[jmc]

We use the Berkowitz-Cohen structure but allow the 1c-3m 4441's to contain 4 controls IF its 2 bare aces. This really helps p after 1c-(3NT-4H) because partner will always have fillers with the controls. We use cue bidding the stiff to ask controls. Answers are 0-1, 2, etc for the cheaper hands and 4, 5, 6 etc for the bigger hands.

I'm sure symetric relay would be better, but with 5 of us playing the system, not everyone is ready to take the plunge.

jmc

[rbforster]

Another option is to use the "impossible negative" approach and respond 1♦ (forcing, generally negative) on minimum 4441's. If partner rebids 1NT, your strong NT systems can easily show this shape. If partner rebids a suit, you can splinter in support or jump to 3N. At least the way I play it, negative hands can pretty much never have the strength to bid 3N opposite a minimum response by the strong club hand, so these are unused sequences anyway.

For potentially slam invitational 4441 hands, I think you'd have to be more careful about whether or not you could handle the different 1♣-1♦ sequences. Either you'd likely need more detailed methods (so there's always a forcing bid responder can make), or perhaps you'd make a direct response to 1♣ (like those in OP) with the stronger hand types.

[ASkolnick]

We also use the "impossible negative" with the "impossible" positive, but only for really strong hands (19+).

Minimum 4-4-4-1's, since we play a canape response style to 1D gets answered with 1 of a major.

[dake50]

What goes on after a negative that so interferes with Unusual Positives? Seems you would plan neg rebids and later developing. Look for UnuP to fill unused bids. Ie what do you insist is needed neg rebids then let rest/else be UnuP. I saw no problem except with 4+controls UnuP so took those into 3C =4441, 4+controls, 11+hcp. Thus UnuP was 8-12hcp, less than 4+controls if 11-12. Often opener's rebid put UnuP into 'big' splinter else jump NT.

[mycroft]

I've learned to like the Berkowitz/Cohen version as well. These are bids that usually take only one or two rounds to resolve, as they should after taking away three rounds of bidding. And what else are you going to use them for (in a non-relay style)?

I haven't run into the "two bare aces" thing, because there is the HCP requirement to go with it; I guess you could find that your suit with QJTx is trump against AKxxx, and Jxxx matches Axx; but alternatively, you could have a standard 2NT opener go 1C-3C; 3NT "not enough" and look really silly.

That 3S vs 3NT one is dangerous - "forget" dangerous, both ways - but the positives for doing so (minimizing wrongsiding the NT (either the "gambling" way or the "playing the relay" way)) is worth taking the time to not get it wrong.

[mycroft]

And as to why not the traditional unusual negative, 1C-p-1D-3H. 2S is probably enough, even. Responder has to do something drastic to show the UnuN, which will not be always the right thing; he's guessing in the dark.

[PrecisionL]

Thanks for all your replies.

We have put the 44(41) and the 44(50) with a minor singleton / void into the 1♣ - 1NT scheme and the (41)44 with a major singleton into the 1♣ - 2♣ scheme.

This leaves 1♣ - 3♣/♦/♥/♠ available for new uses.

Does anyone have frequent uses for the 3-level responses?