[dboxley]

---

I am trying to learn Gazzilli but keep running into this hand type which I don't know how to bid. Any hand with less than 2 in partner's major and 5-7 hcp and no biddable suit of my own (except possibly diamonds). I have read 4 articles on Gazzilli and no one mentions this hand. In my simulations it comes up quite frequently. I know it's not possible that I am the only one who has noticed this. What did I miss? Do I really have to bid 2S on this hand? If so, Gazzilli sucks canal water. I can't believe the Italians ignore this problem.

Thanks in advance,

[awm]

Yes, this is a problem hand.

Of course this is also a problem in 2/1 without the convention; your options are to pass in what might be a 3-3 fit, to bid 2d (if that's natural and not BART, and in any case partner will expect better diamonds and routinely pass with stiff), or to correct to spades on stiff.

[bluecalm]

This is a problem with Gazilli, it doesn't come up too often though.
As 2N is free bid here you can use it for hands without 5hearts and with 0-1 spades.

Quote

If so, Gazzilli sucks canal water. I can't believe the Italians ignore this problem.

Believe it or not, they more or less do

Here are their efforts recorded on vugraph:
http://www.qfpost.co...e/d?g=SAXYWnYlG
(filtered for 0-1 spades, 0-4 hearts, 0-5 minors and 0-7hcp)

Also I have about 32k hands played by Versace/Bocchi/Sementa and only 7 of them ended up in this situation.

[han]

We bid 2H on this hand.

[gnasher]

bluecalm, on 2012-May-01, 13:47, said:

Here are their efforts recorded on vugraph:
http://www.qfpost.co...e/d?g=SAXYWnYlG
(filtered for 0-1 spades, 0-4 hearts, 0-5 minors and 0-7hcp)

Also I have about 32k hands played by Versace/Bocchi/Sementa and only 7 of them ended up in this situation.

Sorry, but this merits a thread hijack. Are you saying that you have a searchable database of Vugraph deals from which you can produce LIN files? Did you create it yourself, or did you buy it?

[bluecalm]

I created it myself. I've written it for myself in Python and the source is kinda messy but I could make it available on github if there is interest.
It has no interface and you need to know a bit of Python to use it (there is bunch of functions you can invoke from command prompt to read hands from .lin files, filter for players/sequencies etc. and create lins out of it). I've also written tools for cleaning lin files (removing all the junk which is usually stored in them), downloading .lins of chosen players/chosen periods from vugraph archives page etc. etc.
I have my doubts if any non-programmer could use it and I don't have much motivation now to make nice interface for it that's why I didn't publish it anywhere.

[dave_w]

bluecalm, on 2012-May-02, 04:26, said:

I created it myself. I've written it for myself in Python and the source is kinda messy but I could make it available on github if there is interest.
It has no interface and you need to know a bit of Python to use it (there is bunch of functions you can invoke from command prompt to read hands from .lin files, filter for players/sequencies etc. and create lins out of it). I've also written tools for cleaning lin files (removing all the junk which is usually stored in them), downloading .lins of chosen players/chosen periods from vugraph archives page etc. etc.
I have my doubts if any non-programmer could use it and I don't have much motivation now to make nice interface for it that's why I didn't publish it anywhere.

Continuing on the thread hijack - this could be interesting. And I'm sure that there's plenty of programmers who would be interesting. I'm not sure what the rights to the files are like. Are they considered public domain or the property of bbo? If you put it up on github you might find that someone else helps you out with a pretty interface!

[gnasher]

I'd be interested to see it. But before you go too far in this direction, maybe you should consider whether it has any commercial value.

Quote

And I'm sure that there's plenty of programmers who would be interesting.

You mean they would be interesting if they weren't programmers?

[mgoetze]

dboxley, on 2012-May-01, 12:46, said:

In my simulations it comes up quite frequently.

Well ... fix your simulations then?

Quote

I know it's not possible that I am the only one who has noticed this. What did I miss? Do I really have to bid 2S on this hand?

No, you are not allowed to bid 2♠ on this hand. Either you are allowed to bid 2♥, as Han is apparently, or you must bid 2NT.

[han]

We're talking about the case where responder has specifically a 1-4-4-4 or 1-4-(53) distribution, and fewer than 8 points. Partner needs to open 1S and rebid 2C. It's not that common. If partner bids 2H over my Gazilly 2C bid I would usually pass with a doubleton and pull with a singleton.

I'm going to make a short commercial for Gazilli. We started playing it about a year ago, and we have probably played about 1500 hands together as a pair since then. Of course the method is not perfect, you pretty much can't play in 2C or 2D after Gazilli so it will hurt you sometimes. However, as far as I remember the 4-card heart suit issue hasn't come up yet. Our decision to start playing the convention came after missing a game two weeks in a row that would have been straightforward to bid had we been playing Gazilli. We decided that it was more important to us not to miss those games. By playing Gazilli opener describes his strength more accurately, you will miss fewer games. On gamegoing hands you also have more room to describe your hand as opener, so you will more often be in the best game. The downsides are that you will sometimes play an inferior partscore (although sometimes you'll stop in 2D/H instead of 3D/H), and that you will always be in game with 16+ opposite 8+, which may work out poorly on misfitting hands.

Here are three further suggestions if you are interested in playing Gazilli, in order of priority:

1- Play artificial follow-ups. If you are going to learn this convention, you might as well play it well. I often see people play natural Gazilli, and they waste a lot of space. Maybe it is possible to play it naturally and play it well, but I haven't seen it yet.

2- Play that after 1S - 1NT - 2C - 2D, opener shows the minimal hand by bidding 2H instead of 2S. I remember one time when I was playing on BBO and this came up, responder passed with hearts. Rogerclee was kibitzing and went crazy, claiming it was the first time ever that he saw a succesful Gazilli auction. Obviously you lose 1 step on some strong auctions, but actually the method becomes easier (assuming you play something artificial, see 1) as you have exactly the same steps as after 1H - 1NT - 2C - 2D.

3- I am not convinced that it is good to play Gazilli after 1H - 1S. It certainly is a lot more complicated than after 1M - 1NT.

To summarize, I don't think Gazilli is suitable for a pair that doesn't want to spend much time on system, but if you play it well I think it is a considerable improvement over "standard". Of course you don't need Gazilli when you play a strong club, in that case these auctions automaticallly become much better.

[Zelandakh]

gnasher, on 2012-May-02, 06:38, said:

I'd be interested to see it. But before you go too far in this direction, maybe you should consider whether it has any commercial value.

There is an online Chessbase site which seems to have survived reasonably well. My guess is that a Bridge Database site could probably attract enough traffic (through Google Ads for example) to cover the server costs if the interface was good enough. It may be that BBO might be interested in a fully working front end to their Vugraph database too since they are probably in a position to market such a utility more effectively.

[bluecalm]

I will think about it. As it is it lays on my hard drive for almost 2 years now.
I am amateur programmer and task of putting together nice front end scares me a bit and code itself probably would need some restructuring too (it's not even proper db, I just use huge list of bridge hand objects which is fast (imports 250k hands under a minute from lin files) but probably not good solution for a site.
Anyway, thanks for interest I will make available one way or the other soon.

Quote

No, you are not allowed to bid 2♠ on this hand.

You could try downloading a file I attached to see that Italians indeed bid 2S with stiff sometimes (and sometimes choose 2N).

[gwnn]

To my understanding, standard Icelandic gazzilli ("Hexan") addresses this by having:

2♦=8+ any
2♥=hearts, weak, 0-1 spade
2♠=2(3) spades, weak
2NT=minors, weak, 0-1 spade, perhaps 1444
3m=5+ in the minor, 0-1 spade

So 2N with 1444 and 3m with 1435/1453. Not perfect, but usually 2NT will have 54+ and 3m will have 6+.

[affe82]

Admittedly in a precision based structure but this is how we solve it:

2♦: accepting invitation or a weak 3suitedhand (short in opener´s suit) that will pass on p´s rebid
2♥: (if opening 1♠, weak 0-1♠, 5+♥)opener is allowed to raise with fit, even if minimum
2♠(over 1♥-1n-2♣): both minors with better ♦.

I also strongly prefer to play 2♣as strong or 6+M, and 2M as M+♣. It makes it a whole lot easier to end up in correct part score exept in a spade sequence where you have a 5/6-3 ♥ fit.

I play the same when playing 2-1 and find it working except the pass out version of 2♦ is a lot less frequent since partner can be so strong you strive to find an alternative that is descriptive enough. In one partnership we have 2N after 2♦ as a GF bid catering to this and 2♦ works as description above again.

[Free]

Depending on suit quality the 1-4-5-3 can be bid in several ways. Usually I consider my options in the following order:
- with ♠A, K or Q, just bid 2♠
- with a decent ♥ suit, bid 2♥
- with a good ♦ suit, bid 3♦
- with a good 7-pointer, bid 2♦
- if all else fails, bid 2NT

On this hand I'd probably bid 2NT. This obviously depends a lot on what the weak options in 2♣ can be.

The 1-4-4-4 is similar, but has fewer options (don't bid 3♦ and don't upgrade).

[Free]

affe82, on 2012-May-03, 08:15, said:

Admittedly in a precision based structure but this is how we solve it:
~snip~

Why on earth would you ever play Gazzilli in a precision type system? You don't need to show the difference between 10-12 and 13-15 so fast.

affe82, on 2012-May-03, 08:15, said:

I also strongly prefer to play 2♣as strong or 6+M, and 2M as M+♣. It makes it a whole lot easier to end up in correct part score exept in a spade sequence where you have a 5/6-3 ♥ fit.

Really dislike this approach after 1♠ openings. But also after 1♥ openings in a regular 2/1 GF framework this is pretty awful imo. The point of Gazzilli is the possibility to play in responder's suit. When playing 6+M or strong, responder will usually bid 2M so the other bids are pretty much wasted. When playing 2M as M+♣ responder needs some bids for invitational hands, so he can hardly run to a decent spot. Moreover when you're playing a forcing or semiforcing 1NT response, opener would have to bid 2M on a 5M-3♣ hand.
I rather play Gazzilli as 2+♣ or strong, and keep opener's 2♦ rebid as 4+♦. I've been playing Gazzilli this way for several years now, and I've had very few problems, even with the problem hands.

[bluecalm]

For what's it worth in limited 1M context (polish club, so 11-17) I play in one partnership:

1S - 1N
2C* - ???

*-5-3-3-2 or strong one suited and then:

2D = weak hand or hearts
2H = hearts, decent hand, say 9-11
2S = 9-11, two spades

This way we end up in 2S frequently on hands where others has to play at 3 level (or 2N instead of 2S) as well as we avoid dreaded jumps to 3S on 6S-3H hands thus missing hearts. I like it but it's not suited for 11-21 openings imo.

[affe82]

if you open good 10s and can have an excellent 15 it makes all the sense to split the range for the version of gazilli I play in precision.
we don't play forcing nt either so the comment about 5-3 M+clubs is semi void.

I find gazilli a great tool for judging when to invite/bid thin games.
admittedly we play inv+ trf follow ups on 2/1 so a lot of the good hands in 1n are gone and responders 1n is most likely a very descriptive bid. Finding the right partscore when opener is balanced is a lot easier when ranges are split and most likely the hand with suitlenghts interested in not playing 1n.

where i play a version of gazilli to precision systems is standard. greater minds then my i trust. there are few problem hands in the version suggested.

[Bende]

affe82, on 2012-May-03, 08:15, said:

I also strongly prefer to play 2♣as strong or 6+M, and 2M as M+♣. It makes it a whole lot easier to end up in correct part score exept in a spade sequence where you have a 5/6-3 ♥ fit.

An idea to address that issue is to play 1♠ - 1NT; 2♦ as 53(41). Of course that comes with some other issues instead.

[32519]

dboxley, on 2012-May-01, 12:46, said:

---

I am trying to learn Gazzilli but keep running into this hand type which I don't know how to bid. Any hand with less than 2 in partner's major and 5-7 hcp and no biddable suit of my own (except possibly diamonds). I have read 4 articles on Gazzilli and no one mentions this hand. In my simulations it comes up quite frequently. I know it's not possible that I am the only one who has noticed this. What did I miss? Do I really have to bid 2S on this hand? If so, Gazzilli sucks canal water. I can't believe the Italians ignore this problem.

Thanks in advance,

This post and this thread contain some random thoughts on Gazzilli.