[pescetom]

The annual convention cleanup freed up memory and enthusiasm for something new, and having already introduced a convention for strong minor opening hands it seems logical to try Gazzilli to do the same for strong(ish) major opening hands. If that goes well then maybe next year will be the moment to move the remaining 18-19 balanced hands into an artificial opening (2♦ or a 1NT rebid after an unbalanced 1♦). I will miss the fun of opening 1NT with 5 card majors, but I can still do that in other partnerships and it will be interesting to compare the MP results of both methods.

My first doubt about Gazzilli is the minimum HCP in conjunction with 2/1 and 15-17 1NT 20-21 2NT. Everyone seems to use 8+ HCP as the positive by Responder, but in various versions I have seen 15, 16 and 17 all used as minimum for Opener. 15 seems the most logical from the 1NT point of view, but not quite enough to render the positive Game Forcing. I guess 17 is fine if we leave 5 card majors in 1NT, but it does not seem elegant to handle 17-19 balanced as either one or two ranges. 16 looks ideal for unbalanced hands not strong enough to reverse, but awkwards for balanced hands of 15(14+) unless we leave those in 1NT (which would also define them precisely).
The better regional players seem to use 16 as the minimum for all shapes handled and with a balanced hand of 15 just open 1M and muddle through.

Does anyone familiar with Gazzilli have comments and suggestions here?

[DavidKok]

I recently wrote out and sent my Gazzilli notes in full, if you're interested I can share them.

The traditional version has the strong hands start at 16 and the positives start at 8. This lets us get away with a game force - usually one side or the other has a bit extra, and 16 opposite 8 is fine but not terrific for game. It also does a nice job of splitting responder's nominal range of 6-9 in two.
I have seen Gazzilli variants with three ranges - 11-14, 15-17, 18+ - and split these. The idea is that 15-17 opposite a positive can still stop in a partscore (though it's a bit awkward to miss some 17-opposite-8 games). Maybe you're thinking of these?
I've also seen Gazzilli variants with two ranges and a balanced hand - 11-15, 16+, and 14-16 balanced. These occur in conjunction with a Kamikaze (10-13) NT. You could shift this up a point in a weak (12-14) notrump system if you want.
The 17+ approach sounds conservative to me, and dated.

I think that playing Gazzilli to resolve 5M332 inside your 15-17 NT opening range is a bad idea. We're inventing a whole system to deal with a single hand type that already has an adequate, if not straight up better, solution.

[pescetom]

DavidKok, on 2026-January-04, 11:45, said:

I recently wrote out and sent my Gazzilli notes in full, if you're interested I can share them.

Please do, either in a specific thread (would be hidden here) or in PM to me, thanks.

DavidKok, on 2026-January-04, 11:45, said:

The traditional version has the strong hands start at 16 and the positives start at 8. This lets us get away with a game force - usually one side or the other has a bit extra, and 16 opposite 8 is fine but not terrific for game. It also does a nice job of splitting responder's nominal range of 6-9 in two.
I have seen Gazzilli variants with three ranges - 11-14, 15-17, 18+ - and split these. The idea is that 15-17 opposite a positive can still stop in a partscore (though it's a bit awkward to miss some 17-opposite-8 games). Maybe you're thinking of these?
I've also seen Gazzilli variants with two ranges and a balanced hand - 11-15, 16+, and 14-16 balanced. These occur in conjunction with a Kamikaze (10-13) NT. You could shift this up a point in a weak (12-14) notrump system if you want.
The 17+ approach sounds conservative to me, and dated.

I agree that 17+ sounds conservative and dated, the Milan school of bean counters.
No interest whatever in that, but my partners are not ready for Kamikaze NT either
I'm not particularly interested in stopping in a part score, even at MP: KISS.
But I do want to keep the convention as natural as possible and I have both 2NT and 3NT available to say something.

DavidKok, on 2026-January-04, 11:45, said:

I think that playing Gazzilli to resolve 5M332 inside your 15-17 NT opening range is a bad idea. We're inventing a whole system to deal with a single hand type that already has an adequate, if not straight up better, solution.

That was my first reaction too, although I'm not sure it is straight up better than a Gazzilli solution at MP (being in 1NT when the rest of the field is in spades can be uncomfortable).
But yes, I can live happily with 1NT if the Gazzilli 3NT bid means 6332 or whatever.

[mike777]

Gazzilli is very popular, especially in Europe. I have never played it. Don't really know it in any depth.


However I seem to recall reading somewhere that it has a number of bad flaws. I don't recall what these bad flaws were.

In addition I don't believe I can play BART or modified BART with it..

[DavidKok]

mike777, on 2026-January-04, 14:12, said:

Gazzilli is very popular, especially in Europe. I have never played it. Don't really know it in any depth.


However I seem to recall reading somewhere that it has a number of bad flaws. I don't recall what these bad flaws were.

In addition I don't believe I can play BART or modified BART with it..

It's imperfect but the flaws aren't too terrible. As with any convention there are costs and benefits. I think it's good to not exaggerate the effects in either direction. The convention is probably good, and probably a net positive, but it's not going to make you rich.
Indeed you cannot combine it with BART, LISA or any modified versions thereof.

[pescetom]

mike777, on 2026-January-04, 14:12, said:

Gazzilli is very popular, especially in Europe. I have never played it. Don't really know it in any depth.


However I seem to recall reading somewhere that it has a number of bad flaws. I don't recall what these bad flaws were.

In addition I don't believe I can play BART or modified BART with it..

I doubt that either Gazzilli or BART have bad flaws, otherwise they would not have survived at national level.
But I agree with Davidkok that any convention has flaws, of course, and modifications are also the rule rather than the exception. In my limited experience of Gazzilli (as an opponent or TD) I have seen Experts playing it with conviction, but also lesser pairs struggling with a moderately complex convention they can barely understand and (above all) with little insight or agreement into how it impacts the rest of the natural system.
I don't undervaluate that, but I think I am ready and understand the risks.

[mw64ahw]

I am familiar and looked at playing Gazzilli (15/16/17) when first fleshing out my 1M openings. Needless to say the benefits (after many simulations) didn't quite stack up when the alternatives were considered. Playing in 2M may be a benefit, but I can cope with 2N and won't miss the games others are in. Finding the optimal partial is often a boon.

[pescetom]

mw64ahw, on 2026-January-04, 16:13, said:

I am familiar and looked at playing Gazzilli (15/16/17) when first fleshing out my 1M openings. Needless to say the benefits (after many simulations) didn't quite stack up when the alternatives were considered. Playing in 2M may be a benefit, but I can cope with 2N and won't miss the games others are in. Finding the optimal partial is often a boon.

I am a bit surprised to hear you (of all people) say this, and would be curious to know what you simulated.
It strikes me that the downsides are mainly in terms of system complication and memory load, which you rarely seem to worry about.
In particular, how were you reusing the natural sequences freed up by Gazzilli?

[mw64ahw]

pescetom, on 2026-January-05, 16:16, said:

I am a bit surprised to hear you (of all people) say this, and would be curious to know what you simulated.
It strikes me that the downsides are mainly in terms of system complication and memory load, which you rarely seem to worry about.
In particular, how were you reusing the natural sequences freed up by Gazzilli?

Say I have 1m simulations and x% are 1♠ openers. Now I have Y sets of possible continuations which I run through my bidding simulator and then compare to DD ending with a score for each set.

I think I originally started with this idea, but it may have been one of the Gazzilli variations, and aim to fold in Useful Space Principles.

I don't seem to have an issue with memory load (apart from occasionally forgetting what partner I'm playing with) and enjoy complexity. This stems from repetition and using generic techniques. You could say it's like learning a language, once you know 3/4 the new ones become easier. I had double digit fluency in computer languages at one stage and sold my first games in the late 70s/early 80s.

The hardest task is getting a partner to try alternative approaches and then implement sufficiently; 'it takes 2 to tango'.

[pescetom]

I too reached double digit, although only by including a few assembler languages. I'm more proud of the fact that my first program compiled and worked first time (and it wasn't "hello world" either).
Agree about the difficulty of finding someone to tango with. Of course we get older ourselves too.

My point about the reused natural sequences is that they constitute a large part of the value of Gazzilli, just like they do with XYZ. You have all the jump bids available to define two suiters of various lengths and strengths and this in turn adds definition to the hands that do go through Gazzilli. It's something I undervalued at first sight, at least.

[mw64ahw]

pescetom, on 2026-January-06, 04:34, said:

I too reached double digit, although only by including a few assembler languages. I'm more proud of the fact that my first program compiled and worked first time (and it wasn't "hello world" either).
Agree about the difficulty of finding someone to tango with. Of course we get older ourselves too.

My point about the reused natural sequences is that they constitute a large part of the value of Gazzilli, just like they do with XYZ. You have all the jump bids available to define two suiters of various lengths and strengths and this in turn adds definition to the hands that do go through Gazzilli. It's something I undervalued at first sight, at least.

Gazzilli is one way of using the jump bids available, but there are others, for example a transfer approach over 1♠.
1♠ - 1N - 3m can show 6x(4x) with intermediate values
1♠-1N-2m/♥-?-2N upwards show stronger hands together with shape

You can also stop in a ♥ partial if opener isn't short
1♠ - 1N (NF) - 2♣ 52(4x)
.. 2♦ pref.
.... 2♥ 52x4
.. 2♥ 5♥
.. 2♠ pref.
.. 2N x5x5 INV - can make 4♥ when opener is 53(4s) & others miss
.. 3♣ 1x26 Weak
There are many ways to 'skin a cat', but I aim to maximise shape/strength combinations as per your comment above.