[Cyberyeti]

DavidKok, on 2022-December-05, 09:04, said:

My 2NT rebids are relatively simple:

1♦-1M; 2NT is 6(+)♦, 3M, strong (the BW death hand - it depends on partnership agreements whether you want this to be GF or just forcing to suit agreement at the 3-level. My preference is GF), though with an unbalanced diamond I think it's better to play Gazzilli so that this becomes a specific type 4-card major suit raise.
1♦-1NT; 2NT is stronger than 1♦-1NT; 3♦.

1♣-1♦* (Dutch Doubleton); 2NT does not exist (a 1NT rebid would have shown 17-19).
1♣-1M; 2NT is 17-19 balanced GF, since 1M promises 8HCP (although we may upgrade for this bid).
1♣-1NT; 2NT does not exist (17-19 balanced bids 3NT, hands with clubs bid 2/3♣ or their second suit) but I suppose it should be a soft 3♣ rebid.

It has become increasingly popular to play 1♣-1M; 2♦ as either 17-19 bal or a real reverse, with 1♣-1M; 2NT showing the BW death hand. I think this is an improvement.

But most importantly, the point is not to make some clever use of the 2NT rebid, it is to stay low on the 19 opposite 4-5 point hands. You can fill the 2NT with any artificial hand (GF long minor or GF artificial raise are popular) once it's freed up.

Edit: I forgot to include the major suit openings. 1M-1X (so NT or 1♥-1♠); 2NT shows 6M4m GF, part of standard Gazzilli.

We play 4 card majors so the structure is different, but 1x-1y-2N covers all 4 card GF raises of partner's suit except 4441s, BW death hand and huge 6-4s.

[DavidKok]

As long as you have a way to play your 19-opposite-4 in 2 of a suit it's all fine, like I said.

[mw64ahw]

Using a Transfer Walsh approach

1♣-1R-2NT for me shows a strong hand with
a) 6+♣ 3M
b) 6+♣ stronger than 1♣-1R-3♣
c) 5+♣ 4M GF not balanced
while
1♣-1R-2D is a strong hand without the other major or an intermediate hand with 3-card support, and
1♣-1R-2M is a strong hand with the other Major.

1♣-1R-1NT is used to show the semi-balanced (14-16) hand
1♣-1R-1NT

[Cyberyeti]

DavidKok, on 2022-December-05, 09:47, said:

As long as you have a way to play your 19-opposite-4 in 2 of a suit it's all fine, like I said.

We don't, we play it in 1N most of the time unless the 4 count has a long suit in which case we might play it in 2 of their suit

[pescetom]

mw64ahw, on 2022-December-05, 04:25, said:

I often wonder what the stats are on having the strong hand declare as often double dummy suggests it makes no difference and when it does, I've seen it work both ways.

I investigated this using double dummy analysis here last year, following similar speculation that right-siding suit contracts after an NT opening might not be important after all.
In a nutshell, when NT opener rather than responder plays in the suit of responder's 5-card major:
11-14 we gain 1+ trick(s) 3.1% and lose 2.7%
15-17 we gain 1+ trick(s) 5.9% and lose 2.6%
20-22 we gain 1+ trick(s) 7.5% and lose 2.6%

(unfortunately you won't currently be able to rerun the scripts, because shortly after that BBO took down the version of Dealer that supports double dummy calculation and has not yet restored it).

[mw64ahw]

pescetom, on 2022-December-05, 11:42, said:

I investigated this using double dummy analysis here last year, following similar speculation that right-siding suit contracts after an NT opening might not be important after all.
In a nutshell, when NT opener rather than responder plays in the suit of responder's 5-card major:
11-14 we gain 1+ trick(s) 3.1% and lose 2.7%
15-17 we gain 1+ trick(s) 5.9% and lose 2.6%
20-22 we gain 1+ trick(s) 7.5% and lose 2.6%

(unfortunately you won't currently be able to rerun the scripts, because shortly after that BBO took down the version of Dealer that supports double dummy calculation and has not yet restored it).

Interesting, it seems that this is based on the likelihood of having a tenace(s) in the stronger hand and being led to, but maybe marginal enough that I don't notice.

[mikeh]

pescetom, on 2022-December-05, 11:42, said:

I investigated this using double dummy analysis here last year, following similar speculation that right-siding suit contracts after an NT opening might not be important after all.
In a nutshell, when NT opener rather than responder plays in the suit of responder's 5-card major:
11-14 we gain 1+ trick(s) 3.1% and lose 2.7%
15-17 we gain 1+ trick(s) 5.9% and lose 2.6%
20-22 we gain 1+ trick(s) 7.5% and lose 2.6%

(unfortunately you won't currently be able to rerun the scripts, because shortly after that BBO took down the version of Dealer that supports double dummy calculation and has not yet restored it).

I think double dummy analysis is not particularly useful in this sort of simulation.

Even if it were, in real life one’s lead depends in part on how strong their auction was and on the form of scoring. At imps, versus a game, I may choose an aggressive lead, not worryin* about overtricks. Against a slam, it’s rare that an aggressive lead stands out…rare but not impossible. DD analysis usually ignores the nuances of the auction.

[smerriman]

mw64ahw, on 2022-December-05, 16:54, said:

Interesting, it seems that this is based on the likelihood of having a tenace(s) in the stronger hand and being led to, but maybe marginal enough that I don't notice.

I mentioned it in that thread, but I'll mention it again here; the double dummy numbers are basically showing how often you have *four* tenaces you need to protect. Any time there's a single safe lead (which is virtually almost always), it assumes the defense will find it. The true value of right-siding has to take into account the probability the defense won't find a safe lead, which those numbers don't show at all.

[mw64ahw]

smerriman, on 2022-December-05, 20:46, said:

I mentioned it in that thread, but I'll mention it again here; the double dummy numbers are basically showing how often you have *four* tenaces you need to protect. Any time there's a single safe lead (which is virtually almost always), it assumes the defense will find it. The true value of right-siding has to take into account the probability the defense won't find a safe lead, which those numbers don't show at all.

Agreed, but this only applies where detrimental wrong-siding occurs so making the benefit a touch more marginal

[smerriman]

mw64ahw, on 2022-December-05, 21:19, said:

Agreed, but this only applies where detrimental wrong-siding occurs so making the benefit a touch more marginal

No, it's the other way around - the advantage of right-siding should be greater than the double dummy numbers show. When the above numbers show an advantage, it's a guaranteed advantage - all 4 suits fail when leading into the strong hand, so you always win. Well, not guaranteed; the winning lead on the other side may not be found. But there should be more hands where you'll get a benefit that weren't in the numbers, because the leader usually has a harder decision to make when leading into the strong hand, making more room for mistakes.

[mw64ahw]

smerriman, on 2022-December-05, 21:27, said:

No, it's the other way around - the advantage of right-siding should be greater than the double dummy numbers show. When the above numbers show an advantage, it's a guaranteed advantage - all 4 suits fail when leading into the strong hand, so you always win. Well, not guaranteed; the winning lead on the other side may not be found. But there should be more hands where you'll get a benefit that weren't in the numbers, because the leader usually has a harder decision to make when leading into the strong hand, making more room for mistakes.

Yes that's what I was trying to say; that the benefit of right-siding is marginal and that in the wrong-sided case the marginality percentage increases, but is still marginal. Still sounds a bit convoluted!

[pescetom]

mw64ahw, on 2022-December-06, 00:53, said:

Yes that's what I was trying to say; that the benefit of right-siding is marginal and that in the wrong-sided case the marginality percentage increases, but is still marginal. Still sounds a bit convoluted!

I still think you missed his point. Any benefits that show up even double dummy are just the tip of the iceberg and so the real world benefits are less marginal than they appear in such analysis.