[straube]

What if we aimed lower, like only offering to stop relays after discovering a major suit fit?

After 1C-1H strong, GF H

1S-asks normal relays
1N-I want to blast to (probably) 4H
.....4H-p/c
.....etc-relay at +1


Let's say that responder insisted on patterning. If his pattern was shapely (by partnership agreement), his base would be normal. So for example, responder has Qx Axxxxxxx void Qxxxx and GF base is set at 5 QPs and this is his base after showing pattern. But if responder had say 3532 or even 3514, his shape may not warrant relays unless he had 3 QPs more over base. So after revealing his pattern, opener knows base is 8. You could potentially be ahead here.

Or you could get even fancier and say that responder can't pattern out unless his QP base is the minimum for a positive response (e.g. 5 in this example) plus the length of his two longest suits -7 (to account for 4333) for a total of (for sake of argument) 8.

So responder has Qx Axxxxxxx void Qxxxx which is 5 QPs + 7 + 4 - 7 = 9. In this example, responder has 1 more QP than is required for permission to pattern. Responder does so, opener then computes that 4 QPs were added for distribution and adjusts responder's hand to 5 QPs for PCB purposes.

OTOH, you have a hand like xxx Axxx KQxx AK and you get 11 + 8 - 7 = 12. Opener does the math, corrects mentally back to 11 QPs and launches PCB.

This is very much a straw man meant to show how one could approach the accommodation of both weak shapely hands an strong balanced hands. To me it's just theoretical. I wouldn't want to spend my time fine-tuning such a structure or working everything out at the table. I think we're pursuing a goal that is very hard for not all that much.

IMprecision addresses a lot of David's concerns. It doesn't really let you blast until you figure out that a semipositive is misfitting or balanced, but oh well. Let the opponents know.

[foobar]

straube, on 2019-November-22, 00:08, said:

IMprecision addresses a lot of David's concerns. It doesn't really let you blast until you figure out that a semipositive is misfitting or balanced, but oh well. Let the opponents know.

Well, assuming opener has a GF hand and slam interest only opposite a really shapely hand, can't one play something along these lines if this is a prime consideration?

1♣ - 1N (5+ ♥; 2-6 QPs) - 4♥ (12+ QPs; but hopeless opposite lower end of QP range and/or hands without shape):
....Pass: Nothing more to say
....4♠+: RKC responses or whatever else makes sense

Playing something like it will probably require using 2♣ = 5+♠ or something, but it's a thought.

[ulven]

My 2 cents. The hands that benefits from a weak relay and where the risk of information leakage are the greatest are the minimum balanced ones. 15-16, 16-17 etc depending on the rest of your opening structure. IMO, those are also the hands that most frequently puts you at a disadvantage when the opponents compete over 1C.

If you decide to use 1NT as 15-17, like for example Muller-de Wijs (except favorable in their case) does, this problem isn't so much of a problem anymore.

[pilun]

David,

I don't visit this forum much these days.
Here are a few thoughts.

When you learn relay, you use it all the time to make sure it sinks in. When bidding boards, we often grind it out for practice, safe enough against robot opponents. At the table, we relay less and less each year, breaking the chain more often. Previously, we would proudly relay to slams and cute games that most would miss, ignoring the ones that they bid and we didn't - where texture is the issue.

We are comfortable bidding opposite a limited opener.
Big winners are hands that go

1♥ - 1♠ R - 2♣ (bal) - 3NT,

where 1♥ = 4+ ♥s, 11-15, denies 4 ♠s.

The concern is the 5-3 major fits. Responder with 3 ♥s often relays to check whether opener has five.
The big risk is seeing opener bidding 2NT (worst case) to reveal 2-4-3-4 to the opponents. Yuk!

Just as bad if responder has 5 ♠s. You could get

1♥ - 1♠ - 2♣ - 2♦ R - 2♠ (♥&♦) - 2NT - 3♣ (2-4-4-3) - 3NT.

At least you play the hand this time but the cost is great. The opening leader should work out that the most likely reason for bidding this way was to check for a 5-3 ♠ fit.

What to do? We tend to forgo the search for the 5-3 most of the time and blast 3NT over 2♣, hoping to gain in the cardplay.

We used to relay on big balanced hands with 4-card support but these days prefer 2NT limit+ raise instead. Basically, slam requires a shapely partner who can show a shortage, much the better way to go.

Strong club auctions feature more relays since - as you point out - the asker needs to cater for a responder with extra strength and/or shape.
Again we have started to break the chain more often. We raise a major with a balanced minimum, use 2NT as a splinter raise, while suit breaks are natural minimums, based on a misfit. These have helped by getting a strong responder to cool his heels.

We also often break into natural after one shape relay, trying to match the natural bidders who have been able to show texture and stoppers.
Nothing worse than finding partner with some 5431 then taking a long time-out to try to guess where the stuff is.

We are happy to break at the drop of a hat but - as you know - I am averse to adding complexity, so we are content with most of these breaks being natural. Must say the (your) 4♦ "mild slam try" has worked well. We use it as "I am about to make a natural slam try, needing extra strength and good trumps. Please bid bad potential trump suits upwards ..."

When we play strong pass, we stick with the old 1♣ = 6-10, 1♦ = 0-6, others 11+ GF. Realise this is technically inferior but the "symmetry" with strong club auctions is a plus. Haven't noticed losses yet.

Nick