Gazzilli: A description
(Page of 3)

In my previous article, I gave a discursive look at why you might want to play 6M-Gazzilli. The general system context includes: five-card majors, strong NT, non-forcing responding NT. While Gazzilli methods can be applied in alternative structures, I will not make further mention of those frameworks. To summarize the system structure from the first article:

• 1M - 1NT - 2 = a normal 2M rebid, or any hand with 17+ HCP
• 1M - 1NT - 2M = replacement 5M + 4 bid
• 1M - 1NT - 2 - 2 = artificial game force opposite the strong option.

Although I've said that this is enough to start playing with (and it is), in the end if you want to adopt a method, you might as well play a decent version, or at least have some agreements about followups. In this article, I will describe the follow-ups that I use (with suggestions offered by Jay Barron).

Someone kindly wrote to me that my previous article was sometimes confusing as to which context I was discussing. To help clarify, in this article, whenever I compare (for surely I will need to compare) to standard natural sequences, I will refer to the standard sequence in italics, e.g. 1M - 1NT - 2 would be a natural club rebid. I hope the context is reasonably clear.

In competition

While your meta-rules might differ, I usually play systems off in competition. I find this easier to remember.

Gazzilli is off the moment the opponents intervene before opener's rebid:

• 1M (Dbl), or
• 1 (1), or
• 1M (Pass) 1NT (Dbl).

If the Gazzilli 2 bid is doubled,

• 1M (Pass) 1NT (Pass)
• 2 (Dbl) ?

Again I stick with my meta-rules: the redouble replaces the artificial values-showing 2 call -- but while showing values, it does not ask to play in 2 XX. Pass suggests that 2 X is a good spot, and 2 is natural weak. In general, if the opponents double an artificial call that has a follow-up artificial ask, then the redouble replaces the artificial ask.

Note that doubling 2 is unlikely to be particularly profitable for the opponents (perhaps as a lead against an eventual 3NT). If opener holds the weak option, he has a six-card major and is likely to play in 2M, in which case the doubler has told himself what to lead.

If the artificial Gazzilli 2 response is doubled, redouble says they made a mistake.

Constructive bidding

Gazzilli is primarily a tool for opener to describe his hand in constructive auctions. In my first article, I tried to motivate the whole discussion by separating opener's hand-types based on general strength, and shape. Here are opener's immediate rebids, where L stands for a lower suit (e.g. 1 - 1NT - 3L is clubs, diamonds, or hearts):

1M - 1NT - ?

• We already know the 6M-Gazzilli bid, and the 2M (= 5-major + 4-clubs) replacement for clubs
• 2 = 4+, limited
• 2 = 4+, limited
• 2NT = ?
• 3L = 5/5, intermediate
• 3M = intermediate 6-card suit.

As a remark, playing Gazzilli, you can (and should) insist that the 1 - 1NT - 2 reverse promise 6+ hearts, while starting with the Gazzilli-2 limits the hand to five hearts if it also contains four spades. If you think about it, this is a Very Good Thing.

I hate to specify high-card values above. Use your judgment as appropriate. Partnership style, suit-concentration/quality, and vulnerability matters (for instance, how aggressively do you bid on 5/5 hands?) and there's no "right answer". But the first time you pick up AK10xx KQJxx xx x, somehow I think you'll figure out which choice to use.

Another improved sequence is the basic limited sequence 1 - 1NT - 2: responder doesn't need to go crazy, when opener has denied 17+ HCP or a hand good-enough to jump-rebid 3. This avoids some of the hopeless 3/4 contracts spurred on by courtesy raises in a standard context.

The 2 NT rebid

Spot the bold question mark above. Now I know what I like to play the 2NT rebid as: 6-of-the-major and 4-of-a-lower-ranking suit (which is also the common usage), intermediate values **. However, I want to point out one of the possible pitfalls. If you hold an 18-19 balanced hand, you have to express that somehow, perhaps by starting with the Gazzilli 2 bid and then rebidding 2NT. However, this treatment would in turn lose an artificial game-force opposite a weak responding hand, which the standard strong-jump shift capably caters for. Here is the problem sequence in psuedo-bidding:

** Edited -- I didn't state the strength. See reply to Adam Parrish's first comment for yet further discussion on this topic.

• 1M - 1NT
• Gazzilli-2 (good hand for once) - bad hand response with <8 HCP
• Want to force *

As far as I can tell, there are various reasonable solutions (notice that this problem is similar to that arising with the Meckstroth adjunct, to which Kenneth Rexford offered a possible "solution" to in the commentary following my first article):

• Have an opening bid to show 18-19 balanced (e.g. Mexican 2 or 2)
• Use the bold-type 2NT as a Meckstroth-type artificial game force
• Stick with Gazzilli-2-then-2NT as 18-19 balanced, but make further suit bids (at *) forcing for 1 round
• Stick with Gazzilli-2-then-2NT as 18-19 balanced, and just don't have an artificial game-force.

My own solution is obviously #1, but in practice, I think the fourth option is just as good. Think about hand-types you might hold for a standard game-forcing strong jump shift. If it's semi-balanced, you always have a Hamman-3NT available at *, while if you had a fake 1-suited jump shift nothing's stopping you from bidding 4M (or 3NT) at *. Finally if you have a genuine powerhouse two-suiter, you can still re-affirm your second suit (possibly by jumping in it).

More simply, the Gazzilli 2 bid is itself 100% forcing, and so in the worst case scenario, you can suck it up and bid game the following round.

Moreover, the 6M+4L handtype simply does occur and offers powerful playing potential. Standard methods starting 1M - 1NT - 2m risk being being passed out in 2m with a good 4M contract being on (being good precisely because of the secondary minor fit). You also add some definition to the single-suited 3M rebid (it's genuinely one-suited).

So now that the doubt is vanquished, 2NT = six of the major + four of a lower-ranking suit, intermediate values. The followups are:

1M - 1NT - 2NT - ?

• other bids are natural and non-forcing (particularly 4M).

Since you would like to be able to play in 3-of-the-lower-ranking suit if possible, opener's third bid over the 3 ask is "revolving natural", that is:

1 - 1NT - 2NT - 3 - ?

• 3 = hearts + diamonds
• 3 = hearts + clubs, while

1 - 1NT - 2NT - 3 - ?

• 3 = diamonds
• 3 = hearts
• 3 = clubs

As ever, the club suit gets the shortest thrift, but we're used to that by now right?

After Gazzilli 2M

1M - 1NT - 2M - ?

• 3 = to play
• others natural (in particular including 2NT, 3M-invite, 4-you-hit-it, and 4M-to-play)
• impossible 2 when the opening major is hearts.

Whether 3L here is to play or invitational depends on your 2/1 and 3/1 systemic agreements, and are not particular to the Gazzilli setup.

After Gazzilli 2

1M - 1NT - 2 - ?

• We already know the Gazzilli 2 bid, which is game-forcing opposite the strong option
• 2M = weak preference
• After 1 - 1NT - 2, 2 is a weak offer to play
• 2NT = both minors, weak
• 3L = weak jump shift *
• 3M = limit raise **

* My own preference is that this weak jump shift should still be constructive (opposite the 17+ option), and is another reason why the 6M-Gazzilli is (in my opinion) superior.

** Opener will nearly always accept with the 6-card suit, and any other call is a cuebid for the strong option.

After a weak response

By example,

1M - 1NT - 2 - 2M - ?

• Pass = the weak option, or a downgraded strong option
• 2NT = 18-19 balanced
• suits = natural, strong

As discussed in depth above, in a Mexican 2/2 context, you can play 2NT as an artificial game force. Otherwise once responder has shown a bad hand, it's up to opener to bid game, or make a clearly forcing bid.

Next time

I've run out of steam, and so I'm going to leave you hanging. Next time, I will discuss the follow up to the positive Gazzilli 2 response (in particular, the 3+ other major bid, and simple shape ask). I will wrap this series with a (hopefully brief) discussion of the promised modifications after 1 - 1.