[the hog]

You do now.

han, on 2012-October-31, 05:23, said:

I don't know which assumptions you tried but I found that 4H is down 33% and 3NT only 16%. I simply gave opener exactly 5-4 in the majors and 18-19 HCP.

Back to the discussion, I don't know anybody who plays that 3H could be either 4 or 5 hearts and it seems quite impossible.

You do now.

[bluecalm]

Quote

I don't know which assumptions you tried but I found that 4H is down 33% and 3NT only 16%. I simply gave opener exactly 5-4 in the majors and 18-19 HCP.

Maybe I made a mistake somewhere, will rerun later and update this post as those layouts are slow for dd. I tried 17-21 from I remember and 5-6spades.

Quote

I don't know anybody who plays that 3H could be either 4 or 5 hearts and it seems quite impossible.

It sucks but it's standard in America from what I gather.

I run dd simul again:
a)17-21 with 5-6spades-4hearts:
3NT 62%
4H 56%

b)17-21, 5-6spades-5hearts:
4H - 95%
3N - 72%

I could improve a bit maybe by removing 5-5 "pure" hands but I doubt it affects overall score by much.

[han]

the hog, on 2012-October-31, 06:06, said:

You do now.

I am sure that mickyb does not play this convention with his regular partner.

bluecalm, on 2012-October-31, 06:19, said:

It sucks but it's standard in America from what I gather.

It's standard among Dutch beginners as well, but I was assuming that we are not discussing beginners bridge here. I realize that most Americans don't play Gazzilli (although I am not sure why) nor 2NT as GF, but I thought that this 1S - 1NT - 3C gadget to distinguish 5-4 from 5-5 was popular among American experts. I certainly learned it from Americans.

I stand by my comment that you can't play 1S - 1NT - 3H as either 6-4, 5-4 or 5-5.

[han]

Bluecalm, doing it separately for 5-4 hands is good because then you know partner won't correct to (for example) 4S. He might with some 6-4's, and certainly with many 6-5's.

Also, 5-4 17-counts wouldn't usually rebid 3H and 6-4 21-counts wouldn't open 1S.

[gnasher]

I have to confess that with Catch22 I play the equivalent sequence as any of 5-4, 5-5 and 6-4. When we first started playing together, we agreed something simple with a note that at some point we'd replace it with Gazilli, and we never got around to changing it. I can't remember any hand where this has cost us - the sequence is fairly rare, hands where responder's action is unclear are rarer, and often there's not much to choose between the different games.

[MickyB]

han, on 2012-October-31, 08:20, said:

I am sure that mickyb does not play this convention with his regular partner.

Wrong, sorry!

I'm not a fan of Gazz. I played 1M:1N, 2N as ART GF for a while but that was in the context of a 15-17 NT, now I'm playing 14-16 and non-forcing NT it's beneficial for 2NT to be nat NF.

I also played some kit over 1S:1N that bore some resemblance to the 1H:1S stuff, but that was when I was playing a 2/1 as GF or good preference, there wasn't room to sort out the 5-10 range through the 1NT kit. Now 1S:1N includes good preference the only artificiality is 1S:1N, 2m:2H as bad pref or various invites [which is free because 1S:2D = 6+hearts any strength or 5+hearts INV+].

I should probably think about 1S:1N, 3C gadgets, but I can't believe it won't have a negative impact on the spade+club GFs.

[han]

I am surprised.

[PhilKing]

MickyB, on 2012-October-31, 08:45, said:

I should probably think about 1S:1N, 3C gadgets, but I can't believe it won't have a negative impact on the spade+club GFs.

You can switch 3♣ and ♥ over 1NT.

In order to handle spade/club hands sufficiently, you have to open 1♣ on game-forcing hands with 5-5 in the blacks. Now everything flows, since opener's only possible shapes are 64, 65 and 5422 (5431s can never jump to 3♥ and hope to survive). Responder bids 3♠ waiting if he needs clarification:

3NT 5224
4C 65
4D/H 64 with void
4S 64, no void

Of course, the main point is to have all that extra room when opener has a game force with both majors, and so now you can do it on 5431s.

[Zelandakh]

You can also solve this by using transfers. There are several implementations of this, for example (after 1♠ - 1NT):
2♣ = diamonds or big balanced
2♦ = hearts
2♥ = spades (one-suited)
2♠ = clubs, weak
2NT = clubs, inv
3♣ = clubs, GF

is a simple possibility.

[MickyB]

I'd thought briefly about switching 3C and 3H, but losing the natural equivalent of 1S:1N, 3C:3H seems non-ideal, even if responder has denied six hearts. I'm afraid I've not quite bought into the idea of rebidding 2m on a GF hand because it's the wrong shape, yet. Maybe 1S:1N, 3C as hearts or 5314 is the way forward.

I don't like transfers on this auction at all, I am far from convinced that the benefits when I am not dealt clubs outweigh the costs when I am. 1S:1N, 2S as 5S4C 11-16 is behind on just about every auction possible.

1S:1N, 2C:P
1S:1N, 2C:2D [natural]
1S:1N, 2C:2S [good preference]
1S:1N, 2C:3C [constructive raise]
1S:1N, 2C:2H!, 2S:bids [various invitational raises of clubs]

This is assuming that you play 1S:1N, 2S:2N as natural invite not a club invite. You really need both of them.

Plus I feel like I have a horrible guess every time I am tempted to pass partner's unlimited transfer rebid.

[MickyB]

han, on 2012-October-31, 08:58, said:

I am surprised.

In the UK this is actually considered quite sophisticated. The old guard still play it as F1, indeed it was suggested that we play it like this at a junior squad training session a few years back.

[bluecalm]

Quote

I realize that most Americans don't play Gazzilli (although I am not sure why) nor 2NT as GF, but I thought that this 1S - 1NT - 3C gadget to distinguish 5-4 from 5-5 was popular among American experts. I certainly learned it from Americans.

Y, I am not sure if any world class pair play that. Even Zia-Hamman played 3C thing (either 5-4 majors or spades or some hands with clubs). Why would you play that instead of Gazilli 2NT GF containing some of those hands is a mystery to me.
Not only you don't distinguish between 5-5 and 5-4 (and it goes for any suit but obviously hurts the most with hearts) but also you don't have any bid for GF hands with spades available (you never have that after 1/1 in "standard") so you lose your 5-3,6-3 heart fits and any way to sensibly stop in 3NT with 6-1 spades too.
Losing slam bidding is not that bad as slams are very rare with one suited which didn't open 2C facing limited hand with no support.

Quote

Bluecalm, doing it separately for 5-4 hands is good because then you know partner won't correct to (for example) 4S. He might with some 6-4's, and certainly with many 6-5's.

Also, 5-4 17-counts wouldn't usually rebid 3H and 6-4 21-counts wouldn't open 1S.

I should be more careful with my assumptions, true. I remember trying several before but maybe I was sloppy if your results are so different. I will run some more later and see.

Quote

I don't like transfers on this auction at all

Me neither. Buras Narkiewicz use it so maybe there is something to it though.

Quote

The old guard still play it as F1

Interesting. Lauria-Versace and Sementa-Duboin play jump shift in a minor as f1 (and 5-5 as 5-4 is covered by Gazilli). I like this. They don't do that with majors though which would be worse as there are no operational bids between 3H and 3S.

[JLOGIC]

bluecalm, on 2012-October-31, 06:19, said:

Maybe I made a mistake somewhere, will rerun later and update this post as those layouts are slow for dd. I tried 17-21 from I remember and 5-6spades.



It sucks but it's standard in America from what I gather.

I run dd simul again:
a)17-21 with 5-6spades-4hearts:
3NT 62%
4H 56%

b)17-21, 5-6spades-5hearts:
4H - 95%
3N - 72%

I could improve a bit maybe by removing 5-5 "pure" hands but I doubt it affects overall score by much.

So when Partner is 54 3n will be much better. When he is 64 or 55 when he will sometimes pull 4m will be better and when he is 65 and will always pull 4m will always be better. Seems like bidding 3n is right. This is especially true Given that double dummy simulations overrate 43 Fits which often have multiple close lines and at tough to play in real life

[JLOGIC]

And has been mentioned 54/ will have a stronger range to jump shift 54 17 counts usually wont

[JLOGIC]

Bluecalm i guess where we disagree is that you seem to think 3n will end the auction. Depending on suit quality I will Expect partner to keep bidding with 64 or 55 a reasonable amount of time, and I expect a 43 heart fit to ply terribly when he passes... your Simulation even has3n better opposite 64 and 54 when surely he is bidding with great spades and 64 and Hez not jump shifting with 54 17 counts

[JLOGIC]

And fwiW Hamman and i play a 3c jump as 4h or clubs or onesuiter. I Think showing four vs five hearts is important. I like 2n natural because its not forcing and sometimes you have 17 opp 3 or 4 or whatever. GaZilli Is not bad but one of thE biggest advantages of semi or non forcing not is the minor bids showing four so I don't like t give that up for gazilli

[bluecalm]

I will come back to this thread with more careful simuls. I agree about the point of simuls overstating chances of 4-3 contract as well as understating chances of 3NT contracts in general so I see that it well might be that I am simply wrong here.
I am probably biased too much by thinking it's so terrible to miss 5-3 hearts that it clouds my assessment of overall EV (so that I am not even motivated to put careful assumptions for dealt hand to avoid pure 5-5's/good 6-4's)

Btw, intuitively what would you bid if clubs were Axx instead of AJx ?

[JLOGIC]

With axx of clubs I'd be tempted to pass. I probably would not if I were scared of being fired, guess I'd just bid 4h and hope for the best

Maybe its just me but partner is the one to expected to bid with 55 or 64 and great suits, he's the one that knows those are suit orienteD even opp a misfit not us unless we are a freaky 11 which he shouldn't worry about IMO