Ghestem: Is it a good convention?

[32519]

I can only ever remember playing against Ghestem once (about 5 years ago). The Michaels/Unusual 2NT combination is what is inevitably encountered at the table.

 

What do others think of this convention? Is it good or not?

[Cthulhu D]

http://www.blakjak.demon.co.uk/brx_brn0.htm

 

Pretty much sums it up I reckon.

[RMB1]

Can I nominate this topic as "most amusing combination of title and poster" for the annual awards?

[paulg]

Good convention, played by all the top Italian pairs.

[antonylee]

Standard in France... if you grow up with it you get confused later when the ACBL bridge bulletin has bidding contests where you hold AKxxx x xx AQxxx, the auction starts (1S)-2S-(P), and everybody tries to guess what is happening :-) (not that this wouldn't be a legitimate problem even with Ghestem in the picture of course).

[mr1303]

No

[lowerline]

Read this

 

[Zelandakh]

Another plug for my solution to this since Cthulhu was very kind in recently posting the link in. Ghestem itself is famous most of all for causing TD headaches. It is fine if you are someone who never forgets any system but for the vast majority of players it is simply a bad idea to play it. Arguably, you could also improve Ghestem by moving around the bids a little. For example, Ghestem over 1♦ is

 

2♦ = spades and clubs

2NT = hearts and clubs

3♣ = hearts and spades

 

Surely it has to be better to play

 

2♦ = hearts and spades

2NT = hearts and clubs

3♣ = spades and clubs, weak

3♦ = spades and clubs, strong

 

giving you back your Michaels cue bid for the majors? Or even RCO

 

2♦ = hearts and spades

2NT = spades and clubs

3♣ = hearts and clubs, weak

3♦ = hearts and clubs, strong

 

 

If a partner insisted on Ghestem then sure, I might play it as a compromise. But I would never choose to play it and there are alternatives out there that are (imho) simply better.

[32519]

Ok, so we have a mixed audience at to whether or not Ghestem is a good convention. Now let’s throw Questem into the mix.

 

At the bottom of the above article there is another link comparing UNT/Michaels, Ghestem and Questem.

 

The author of the article favours Questem. Is (s)he right?

[bluecalm]

Either Ghestem or top/bottom cuebids are superior to michaels imo. You don't really gain much in bidding by showing 5-5 without specific 2nd suit but you give away a lot if they end up declaring.

[Zelandakh]

Ok, so we have a mixed audience at to whether or not Ghestem is a good convention. Now let’s throw Questem into the mix.

Is this not the fancy name for one of the methods I mentioned in my post? The loss of the strong hand type in 3♣ is not really an issue at all - you can simply use the jump cue for this.

[Vampyr]

I can only ever remember playing against Ghestem once (about 5 years ago).

 

Ah, but how many times in the past five years have you played against supposed Ghestem where they got it wrong?

[awm]

The forgets and related director calls are funny and surely true, but let's suppose you can remember the methods.

 

Ghestem definitely wins on hands where opponents raise the bidding and advancer has a good fit for overcaller's 2nd suit. This certainly comes up and its not hard to give examples. However Ghestem is still a bad method.

 

1. Frequency. Surely a specific 5-5 shown by 3C is less frequent than a long club suit. The exact same issue (failure to sacrifice or even make when advance fits overcaller's suit) can arise if you are forced to pass the club one-suiter.

 

2. Effectiveness. When there is no fit, forcing to the three level is much more dangerous; Michaels often can stop in 2M. Telling the opponents both suits can help them in both bidding and play. If 3c is forcing (as in original Ghestem) you give them two cues; if not (as in some mods) you reduce frequency further because you can't use the method on very strong hands in case partner passes.

[32519]

Ah, but how many times in the past five years have you played against supposed Ghestem where they got it wrong?

 

The forgets and related director calls are funny and surely true, but let's suppose you can remember the methods.

Ghestem definitely wins on hands where opponents raise the bidding and advancer has a good fit for overcaller's 2nd suit. This certainly comes up and its not hard to give examples. However Ghestem is still a bad method.

1. Frequency. Surely a specific 5-5 shown by 3C is less frequent than a long club suit. The exact same issue (failure to sacrifice or even make when advance fits overcaller's suit) can arise if you are forced to pass the club one-suiter.

2. Effectiveness. When there is no fit, forcing to the three level is much more dangerous; Michaels often can stop in 2M. Telling the opponents both suits can help them in both bidding and play. If 3c is forcing (as in original Ghestem) you give them two cues; if not (as in some mods) you reduce frequency further because you can't use the method on very strong hands in case partner passes.

 

These are very interesting comments from both of you.

 

When I encountered Ghestem at the table for the first time, after receiving the explanation from the opponents as to what the bid promised, my then F2F partner and I decided we would experiment with it. We dropped it again about six months later for the very reasons that you mention here:

1. We screwed up the bidding the first two times we had an opportunity to use Ghestem.

2. When we started getting it right, the loss of the 3♣ pre-empt over the opponents opening bid hampered effective interference in the auction.

 

We subsequently resorted back to Michaels which is what I still play with my current F2F partner.

 

Feedback from everyone much appreciated. Thanks to all.

[blackshoe]

I admit to some fondness for Hardy's approach in Competitive Bidding with Two Suited Hands: Top and Bottom Cue Bids, ELC doubles, Unusual NT and a couple of artificial jumps over minor openings (specifically (1♣)-2♦ shows diamonds and hearts, usually 4=5, and similarly (1♦)-2♥ shows clubs and hearts. OTOH, I don't have enough experience with it to know whether it's a better method than standard Michaels. Hardy certainly seemed to think so (yes, I know some here don't think much of Hardy's opinions on bidding theory).

[bluecalm]

Surely a specific 5-5 shown by 3C is less frequent than a long club suit.

 

Yes, but some hands with long clubs could be overcalled with 2C and effectiveness of preempt once they open 1M is small (usually they get to their contract anyway and declaring is now much easier). I think this is a good point and surely cost of Ghestem but it's not big.

 

Effectiveness. When there is no fit, forcing to the three level is much more dangerous;

 

It's also more preemptive, not much and it doesn't matter that much but it is.

 

Michaels often can stop in 2M.

 

Yeah, but you can just as well overcall 1S instead of using Michaels, this way you are on 1 level and you didn't give them free information which benefits them every time and your side only if partner has 4-4 fits in minors (and this is what we should compare Ghestem to).

 

My point is that while it's possible that the costs mentioned by you are too much ot make Ghestem > Michaels we have top-bottom cuebid which just dominates Michaels in every area.

[Zelandakh]

Yes, but some hands with long clubs could be overcalled with 2C and effectiveness of preempt once they open 1M is small (usually they get to their contract anyway and declaring is now much easier).

I think there is a good point here, that we should differentiate between times when they open 1m and 1M. This is one of the reasons why my proposed method (the one where the cue is a wjo in a major or top+bottom strong) only applies to a 1m opening. The gains of playing this kind of thing diminish when they open 1M.

 

 

Yeah, but you can just as well overcall 1S instead of using Michaels, this way you are on 1 level and you didn't give them free information which benefits them every time and your side only if partner has 4-4 fits in minors (and this is what we should compare Ghestem to).

The main benefit of playing Michaels for these hands is (arguably) to get them out of other auctions. If you play top+bottom plus ELC you are affecting your auctions after a double. If instead you overcall 1♠ then you either lose the inference that a 1♠ overcall followed by a new suit shows a good hand or you simply give up on those hands. This is even worse when they open spades and we have hearts - do we really think it is a good idea to throw weakish ♥ + minor 2-suiters into a 2♥ overcall?

 

In this respect I think it is misleading to say that the only gains from Michaels are when partner has 4-4 in the minors. It is certainly reasonable to suggest that the losses from ELC are less than the losses of the minor being unknown in the cue bid. I have not seen any statistics but enough good players have switched to using ELC that it would be difficult to escape that conclusion. On the other hand there are still plenty of good pairs using Michaels which also suggests that both methods are reasonable.

[blackshoe]

Michaels is a reasonable method. If it wasn't, it wouldn't be so popular. But that very popularity makes it suspect — how many players (even experts, although less so in their case, I think) play it because "it's standard", or "everybody plays it", or "I've always played it", and not because they've tried all other methods and found them wanting?

[Cthulhu D]

Michaels is a reasonable method. If it wasn't, it wouldn't be so popular. But that very popularity makes it suspect — how many players (even experts, although less so in their case, I think) play it because "it's standard", or "everybody plays it", or "I've always played it", and not because they've tried all other methods and found them wanting?

 

There is a second point here - one of the big advantages of Michaels is that it doesn't have much opportunity cost - a natural 2NT overcall and a 'natural' cue are bids you can toss without even noticing. If you want specified two suited over-calls you either need to play some bids multiway which is a large degree of complexity to sort out with partner (after (1C)-2D!-(X) where 2D! was a a pre-empt in an unspecified major, what exactly do your bids mean? It's obvious that 2S should be a paradox advance, but how high does that authorise the 2D bidder to go and what hand types should go?) or sacrifice weak jumps - a relatively common handtype.

 

Despite the recommendations to play specified two suited bids from Partnership Bidding at Bridge guys, and how helpful knowing what the second suit is in a two suited overcall is to competitive bidding generally, this cost of complexity or oppotunity is, imho, a key reason Michaels remains very popular. It's simple AND effective.

[gnasher]

If you want specified two suited over-calls you either need to play some bids multiway which is a large degree of complexity to sort out with partner ... or sacrifice weak jumps - a relatively common handtype.

There are at least two other solutions, both of which seem better to me:

- Play only two specific two-suiters.

- Use a jump-cue bid as a two-suiter.

[ColdCrayon]

I have the opposite problem; I don't play Michael's, but I'm much more comfortable with the Qhestem/Questem or comprehensive two-suited systems. So when I play with another partner, and they make a Michael's bid I'm constantly bolloxing it up.

 

Anyway - momentarily forgot my point - Questem avoids Ghestem's ambiguity by making the 3c jump show top-and-bottom hands. So clubs are likely to be one of them anyway, and the bottom suit is usually the stronger one - otherwise you'd just bid the major.

 

I prefer Max Hardy's system - UnNT shows bottom two, a cuebid shows top-and-bottom, and a takeout double shows top-two. The advancer will at first treat it like a convention T/O; if he bids the one suit you don't have support for, you correct them to the lower of the top two that you have support for. If you move the higher suit, that shows a hand too strong for an overcall.

 

The occasions when you want to make a double showing an 18-19+ hand, AND advancer picks the suit directly below your suit are the only times you'll have ambiguity. To show the rock-crusher rather than the top-two, you have to jump.

 

If it sounds unwieldy it shouldn't. If you can master the whole negative/responsive/takeout/support/overcall/proper balancing/direct-seat/passed responder/take-out/for penalty/forcing pass system, adding a tiny wrinkle to it is nothing. [edit] I'd like to go on record by saying that bridge books really need to devote more time to the above, and just competitive bidding in general than they do. Typically most systems have about 9 chapters to show the most elaborate systems for uninterupted auctions for every one chapter on competitive auctions. It should be closer to 50/50, if not totally reversed.

[nige1]

With a tiny change, Michaels handles more 2-suiters. No change over 1M:

• 2N = 5+5+ minors.
• 2M = 5+ OM & 5+ minor.

Over 1m:

• 2N = 5+ ♥ & 5+ om as usual, but
• 2m = 5+ [sP} & 5+ another unbid.

Over the ambiguous cue-bid:

• Raise known suit = pre-empt.
• Cue = High card raise in known suit.
• 2N = Asks for other suit.
• New suit usually = Pass/Corrent.

Later, when your two-suits are known, opposite a silent "advancer", you can show:

• Extra shape by bidding a long suit,
• Or bidding no-trump with 6-6.
• Extra power by doubling.
• Both power and shape, by cue-bidding again.

When advancer shows a suit, then

• With a weak hand, pass or correct
• With a strong hand and a fit, cue-bid
• With a strong hand and the other suit, bid notrump.

Similarly with extra shape and extra power, when it isn;t clear which unbid suit you hold, then

• cue-bid shows higher ranking
• notrump shows lower ranking

Other refinements:

• When RHO bids a new suit that might be one of partner's suits e.g. (1♥) 2♥ (3♣) ?? then double = pass/correct.
• When RHO pre-empts e.g. (1♥) 2♥ (4♥) ?? then double = support for higher unknown suit - here ♦..
• When LHO pre-empts and partner passes e.g. (1♥) 2♥ (4♥) _P; (_P) ?? then I suppose you might agree that double shows extra power and ♣.

[Free]

There are probably 32519 different ways of showing 2-suiters and you let us choose only 3 of them :D I like Kuala Lumpuestem

[P_Marlowe]

Hi,

 

first you should decide the answers to some question:

 

#1 Do you want to get in with a cont. range, or only with weak or

strong hands?

#2 Do you want to get in with 54 shapes as well?

This includes also the cases, where you go in, with 54,

pretendng to show 55, and afterwards saying,sry p, I mixed

up a card, a club was among my spades.

 

If you answer one or both qustions wih yes, than Michaels is

better, because it keeps the auction lower, i.e. you have more

room to figure out, what is going on.

 

With kind regards

Marlowe

 

PS: For whats it worth - we play Michaels.

[mikestar13]

With a tiny change, Michaels handles more 2-suiters. No change over 1M:

• 2N = 5+5+ minors.
• 2M = 5+ OM & 5+ minor.

Over 1m:

• 2N = 5+ ♥ & 5+ om as usual, but
• 2m = 5+ [sP} & 5+ another unbid.

Over the ambiguous cue-bid:

• Raise known suit = pre-empt.
• Cue = High card raise in known suit.
• 2N = Asks for other suit.
• New suit usually = Pass/Corrent.

Later, with the two-suited hand , opposite a silent partner, you can show:

• Extra shape by bidding a long suit,
• Or bidding no-trump with 6-6.
• Extra power by doubling.
• Both power and shape, by cue-bidding again.

Other refinements:

• When RHO bids a new suit that might be one of partner's suits e.g. (1♥) 2♥ (3♣) ?? then double = pass/correct.
• When RHO pre-empts e.g. (1♥) 2♥ (4♥) ?? then double = support for higher unknown suit - here ♦..

 

This is exactly what my favorite partner and I play, but you have better-defined followups (which we will probably adopt). Ghestem is likely to get you too high, and in any case I don't think

5-5's are frequent enough to use three bids on, especially the devastating (1♦ )-3♣ sequence. If I really insisted on an unambiguous method, perhaps this:

 

(1♣)-2♣ = ♠ and ♦.

(1♣)-2♦ = ♠ and ♥.

(1♣)-2NT = ♥ and ♦.

 

(1♦)-2♦ = ♠ and ♥.

(1♦)-2♥ = ♠ and ♣.

(1♦)-2NT = ♥ and ♣.

 

(1♥)-2♥= ♠ and ♦.

(1♥)-2♠ = ♦ and ♣.

(1♥)-2NT = ♠ and ♣.

 

(1♠)-2♠ = ♥ and ♣.

(1♠)-2NT =♦ and ♣.

(1♠)-3♣ = ♥ and ♦.

 

This make the non-cue bid suit show two other suits, so it can also be used on strong hands. I still don't like to give up a WJO, but if I must, let it be the cheapest.