[thepossum]

Hi all

Apologies for posting a topic that is partly GiB related. However it is a general question

In the 2/1 system we play a weak two can be 4-10 HCPs, 6 cards in the bid suit and ideally no 4 card side suit. This is different from other ranges (too strong IMO) which go from 5-11 - thats getting to 1 bid territory

If you have a strong response. Say 15+ points with only 2 trumps opposite opener and bid 2NT - not playing Ogust - I occasionally am disappointed by the "minimum" response when opener is sitting with a nice 6 card suit, 8 points and maybe a singleton

So game can be missed

P

[helene_t]

Without a feature to show and without the solid suit required for a 3NT rebid, GIB doesn't have an ideal rebid.

Maybe it would be better to play 3♣ as showing no feature, and rebid the suit with a feature in clubs. Then after 3♣, responder could try for game with 3♦.

But that's not how the GIB system works.

[DavidKok]

Thankfully I don't have to deal with GIB. I can say that, when playing with humans, the question is under-specified and even partly mistaken. Just having a certain number of HCP and a specific number of trumps does not make a preempt, and conversely those are not strict requirements for an effective weak two. I wrote about this previously in more detail.

I think the honest answer is that, no matter what slight improvements you make to your responses or requirements for preempting, you will continue to miss good games and continue to bid bad ones. You can decide which of the two you want to do more frequently. If you want to significantly improve on both scores I would recommend going into much more detail on what does and what does not constitute a preempt for your partnership.

[mycroft]

I'll go stronger than David: No matter what [ ] your responses or requirements for preempting, you will continue to miss good games and continue to bid bad ones.

Because the third leg of that stool is "how much you stop your opponents from finding their best contract, or their best response to your contract". You absolutely can decide to make your 2 bids so well defined and so beautifully systemic that your constructive bidding is "perfect", and it will likely even help your 1-level constructive bidding. But you will lose every time you have a weak 2 that you can't open, and that will be a lot.

My preference is very strongly on the side of "there are two opponents and one partner. The chance I'm causing one of them problems is worth when I put partner to the guess." And that means when we go +170 or -100 because I preempted partner and she guessed wrong, all I get to say is "hard luck." But I will play any style - as long as I've had the discussion. A comfortable partner is much more important than my preferences here.

Preempts are *gambles*; sometimes you gamble and lose. The more effective they are at preempting the opponents, the more often, and the more dramatic, the losses. Where the line is between your losses and their gains is the perennial discussion.

Of course, if you're bidding with the robot, you don't get a discussion - play its way or suffer the consequences.

[DavidKok]

That's not stronger than what I said, but rather completely missed my point.

[thepossum]

DavidKok, on 2022-May-04, 07:07, said:

Thankfully I don't have to deal with GIB. I can say that, when playing with humans, the question is under-specified and even partly mistaken. Just having a certain number of HCP and a specific number of trumps does not make a preempt, and conversely those are not strict requirements for an effective weak two. I wrote about this previously in more detail.

I think the honest answer is that, no matter what slight improvements you make to your responses or requirements for preempting, you will continue to miss good games and continue to bid bad ones. You can decide which of the two you want to do more frequently. If you want to significantly improve on both scores I would recommend going into much more detail on what does and what does not constitute a preempt for your partnership.

I will add a bit of additional specification

Suppose you think you have 5 or 5 and half tricks, maybe 7-8 losers is that a minimum, average or maximum weak two

I appreciate pre-empting considers vulnerability and other things

Also, I was hoping for a non-GiB restricted discussion

If 8 hcps in a nice major suit, a singleton in a 6331 shape is regarded as minimum, what is a 4 HCP hand

Say, for example I play a 234 type style with 3 and up pre-empts, based on tricks + 2,3, or 4 or consider losers etc. is a minimum weak two up to 9 losers and only 4 likely tricks? My general style (ideally with a human) is light-sound. Rarely very light. Let's say light

I would regard a singleton as some kind of feature much of the time - maybe not as good as a void

[mycroft]

Then I missed your point completely, I'm sorry. Can you explain, please?

But what off what I read: no matter what slight improvements, gross improvements, or complete overhauls you make to your weak 2s, you will have hands that will catch you out. You can decide how they're going to catch you out, but "not missing a good game when I preempt" and "not bidding a bad game after I preempt" (never mind "games that make" and "games that don't", respectively) are just not achievable goals, and even to prioritise those you have to have a preempt style that will let the opponents run roughshod over you.

If what you're saying is "given your preempt style, your followup system can only be on a sliding scale between 'missing good games' and 'bidding bad games', but with a serious look, not just tweaks, you can make that sliding scale smaller ", then, sure. But eventually you run into "to get any better than this, we have to change our preempt style".

[thepossum]

mycroft, on 2022-May-04, 09:36, said:

Then I missed your point completely, I'm sorry. Can you explain, please?

But what off what I read: no matter what slight improvements, gross improvements, or complete overhauls you make to your weak 2s, you will have hands that will catch you out. You can decide how they're going to catch you out, but "not missing a good game when I preempt" and "not bidding a bad game after I preempt" (never mind "games that make" and "games that don't", respectively) are just not achievable goals, and even to prioritise those you have to have a preempt style that will let the opponents run roughshod over you.

If what you're saying is "given your preempt style, your followup system can only be on a sliding scale between 'missing good games' and 'bidding bad games', but with a serious look, not just tweaks, you can make that sliding scale smaller ", then, sure. But eventually you run into "to get any better than this, we have to change our preempt style".

I don't think anyone missed my point

All I was wondering was if you have, say, AKJTxx,txx,xxx,x is that a minimum Weak Two

To me after my 2NT positive strong bid I would expect something like QJxxxxx, Kxx, xx, xx - edit maybe the King is wrong for a minimum (maybe a Jack)

I accept with my nice 2NT hand that I considered game irrespective of the discouraging 3S

Oops. sorry. Additional specification. Playing IMPs against a rather cautious/disciplined(?) field - where 3+1 only scored -1 IMP

[akwoo]

thepossum, on 2022-May-04, 09:42, said:

I don't think anyone missed my point

All I was wondering was if you have, say, AKJTxx,txx,xxx,x is that a minimum Weak Two

To me after my 2NT positive strong bid I would expect something like QJxxxxx, Kxx, xx, xx

This is a matter of partnership agreement.

I've played in a partnership where a weak 2 promised 1.5 quick tricks, so the minimum weak 2 is AKxxxx xxx xx xx. I didn't like it.

I play in a partnership where QTxxxx x xxxx xx is a weak 2 NV. (Okay, for that partnership, AKQxxx xxx x xxx is a 1 bid.)

Decide on your agreements with your partner. Different styles have pluses and minuses.

(My preference is wide-ranging weak 2s with Ogust. But if partner wants I'll play sound weak 2s without Ogust.)

[mycroft]

Sorry, possum. I should have quoted. David said "I think you completely missed my point", and I said I'm now confused. Wasn't aimed at you at all.

[DavidKok]

mycroft, on 2022-May-04, 08:45, said:

I'll go stronger than David: No matter what [ ] your responses or requirements for preempting, you will continue to miss good games and continue to bid bad ones.

Because the third leg of that stool is "how much you stop your opponents from finding their best contract, or their best response to your contract". You absolutely can decide to make your 2 bids so well defined and so beautifully systemic that your constructive bidding is "perfect", and it will likely even help your 1-level constructive bidding. But you will lose every time you have a weak 2 that you can't open, and that will be a lot.

My preference is very strongly on the side of "there are two opponents and one partner. The chance I'm causing one of them problems is worth when I put partner to the guess." And that means when we go +170 or -100 because I preempted partner and she guessed wrong, all I get to say is "hard luck." But I will play any style - as long as I've had the discussion. A comfortable partner is much more important than my preferences here.

Preempts are *gambles*; sometimes you gamble and lose. The more effective they are at preempting the opponents, the more often, and the more dramatic, the losses. Where the line is between your losses and their gains is the perennial discussion.

Of course, if you're bidding with the robot, you don't get a discussion - play its way or suffer the consequences.

mycroft, on 2022-May-04, 09:36, said:

Then I missed your point completely, I'm sorry. Can you explain, please?

But what off what I read: no matter what slight improvements, gross improvements, or complete overhauls you make to your weak 2s, you will have hands that will catch you out. You can decide how they're going to catch you out, but "not missing a good game when I preempt" and "not bidding a bad game after I preempt" (never mind "games that make" and "games that don't", respectively) are just not achievable goals, and even to prioritise those you have to have a preempt style that will let the opponents run roughshod over you.

If what you're saying is "given your preempt style, your followup system can only be on a sliding scale between 'missing good games' and 'bidding bad games', but with a serious look, not just tweaks, you can make that sliding scale smaller ", then, sure. But eventually you run into "to get any better than this, we have to change our preempt style".

I do agree with all of this, it's just a different point from the one that I was trying to make. There are many hands that are neither too strong nor too weak for a preempt, but that are simply not suitable. Earlier this evening I picked up ♠Ax, ♥Axx, ♦9865432, ♣J at everybody vulnerable, second hand (RHO passed). A few games before that I picked up ♠x, ♥9865, ♦AQJ9xx, ♣Qx at everybody vulnerable, third hand (partner and RHO passed). The first hand has got an extra diamond and is arguably stronger, and the second hand has the classic flaw of a 4-card major side suit, but I passed the first and opened 3♦ with the second and stand by those calls.
If you adopt a preempt style that is more nuanced than "am I too weak to open + how many cards in my longest suit?", you will create correlations between how many HCP you can have and what type of hand you can have. In sufficiently extreme examples, the link between HCP and playing strength is almost entirely gone. I think this is related to the old theory of 'cover cards' - the observation that, when we've got a long real suit (and therefore one or more somewhat short suits), aces and kings are worth way more than their point count traditionally suggests, and queens and jacks might well be near worthless. As some fabricated examples, if I pick up ♠AQJTxx, ♥x, ♦xxx, ♣xxx I will not only open 2♠ when vulnerable, first seat, but also tell partner I have a maximum if partner bids 2NT. Change the hand to ♠QT9xxx, ♥AJ, ♦Qx, ♣xxx and I would still preempt (with distaste) but now I'm very tempted to show a minimum if partner asks (again, vulnerable and first seat). Swap the jack of hearts for the queen in that hand and I'm opening 1♠. The playing strength of that hand is just much lower, even if it holds more HCP.
Put differently, a 2NT asking bid (no matter which flavour you prefer) is a pretty crude tool to win a few more bits of information from partner before we have to commit to a contract. It is only possible to make sensible decisions if your agreements about opening 2X were doing a lot of hard work for you in the first place. It's really not as simple as saying "two opponents, one partner, we bid a lot and shrug when it goes wrong" - I'll bet I bid more weak two's than most here, but I still consider my style aggressive but disciplined. And part of those partnership agreements include determining what sort of hand partner is looking for with an asking bid, and to what degree we have that hand on offer. And of course that is a two-way street - partner also needs to consider whether, in your partnership agreement on preempts, their decent-looking balanced 15 HCP hand is suitable for making a game try in the first place.

[sfi]

DavidKok, on 2022-May-04, 16:40, said:

I do agree with all of this, it's just a different point from the one that I was trying to make. There are many hands that are neither too strong nor too weak for a preempt, but that are simply not suitable.

I agree with this point (assuming I'm not missing it as well ). There is a fairly strong school of thought that a 9-11 point hand with a six-card major (and no extreme shape) is either a one-level opening or a two-level opening - there is no gap between the two ranges. I've never bought into this because I think about the two bids differently. One-level openings are all about general values and shape, and they're intended to be constructive. Two-level openings are about preemptive value and trick-taking potential with that suit as trumps. The two things aren't IMO particularly part of the same spectrum.

Therefore, a hand like A97632 Jx Axx xx doesn't look to me like either of those things and I would prefer to pass it rather than try to squeeze it into either range. On the other hand, a hand like AKQJxx xx xxx xx kind of fits both descriptions, and I would happily open it 1S or 2S depending on partnership style. So only some hands bridge the gap between one-level and two-level openers.

[thepossum]

Hi all

I need to make a small correction (and apology) for my scenario (my memory is not good these days)

The spade suit was not accurately reported. It was AQJxxx so only 7 points but I regretted not joining the immediate 4S high scorers with my beautiful 2NT

...and 3+1 scored 0

It worked out because the singleton was opposite my weak suit - so if it could have been shown as a feature it would have been good!

Remarkably it even made 3+1 with a 4-1 trump break and the King being offside

[Stephen Tu]

sfi, on 2022-May-04, 19:57, said:

I agree with this point (assuming I'm not missing it as well ). There is a fairly strong school of thought that a 9-11 point hand with a six-card major (and no extreme shape) is either a one-level opening or a two-level opening - there is no gap between the two ranges. I've never bought into this because I think about the two bids differently.

I thought that the school of thought was there's no gap *when holding a good suit*. Not with junky suits without internal solidity.

[thepossum]

Just to double check. You are on lead as follows

---

Try to forget everything above. What would you lead against 3 or 4 spades?

Double dummy just makes 3

[sfi]

Stephen Tu, on 2022-May-05, 00:01, said:

I thought that the school of thought was there's no gap *when holding a good suit*. Not with junky suits without internal solidity.

That's certainly a more reasonable school of thought, but not the only one. If there truly is no gap, then you get to have an agreement that a new suit by a passed hand shows a fit in a bunch of auctions. For example:

P - (1H) - 2C - (2H)
2S

Now can show 5 spades and some clubs, since a hand with 6+ spades and values would have acted in first seat (and bidding without tolerance would be too risky). As I said, not something I agree with. But there are definitely people who think like this and teach these ideas.

[DavidKok]

thepossum, on 2022-May-05, 00:08, said:

Just to double check. You are on lead as follows

---

Try to forget everything above. What would you lead against 3 or 4 spades?

Double dummy just makes 3

Ace of diamonds.

[DavidKok]

Stephen Tu, on 2022-May-05, 00:01, said:

I thought that the school of thought was there's no gap *when holding a good suit*. Not with junky suits without internal solidity.

I still don't think this is accurate, although it is of course better. For example, second hand unfavourable, you pick up ♠QJxx ♥KQT9xx ♦x ♣xx. Or, if you frequently bid a weak two on only five cards as I do, first hand nobody vulnerable you pick up ♠JTxxx ♥AKTxx ♦xxx ♣- (I would open this 1♠ if the jack was swapped for a queen. If you wouldn't, feel free to make that swap instead).

[nullve]

DavidKok, on 2022-May-04, 16:40, said:

As some fabricated examples, if I pick up ♠AQJTxx, ♥x, ♦xxx, ♣xxx I will not only open 2♠ when vulnerable, first seat, but also tell partner I have a maximum if partner bids 2NT. Change the hand to ♠QT9xxx, ♥AJ, ♦Qx, ♣xxx and I would still preempt (with distaste) but now I'm very tempted to show a minimum if partner asks (again, vulnerable and first seat). Swap the jack of hearts for the queen in that hand and I'm opening 1♠. The playing strength of that hand is just much lower, even if it holds more HCP.

You seem to be claiming that

1) QT9xxx AJ Qx xxx has much lower playing strength than AQJTxx x xxx xxx.

2) While QT9xxx AJ Qx xxx is barely maximum for a Weak 2♠, QT9xxx AQ Qx xxx is worth a 1♠ opening.

If so, what do you mean by 'playing strength'? If something like trick expectation, then the Binky values for

AQJTxx x xxx xxx

and

QT9xxx AJ Qx xxx

suggest you're wrong about 1).

Also, if both 1) and 2) are true, is the former hand actually closer to being worth a 1♠ opening bid?

[DavidKok]

Yes, I claim both 1) and 2). I think that the AQJTxx hand is much more likely to take a lot of tricks opposite the type of hand where partner would ask with 2NT, and is therefore worth a positive response. With decent support partner will almost always blast 4♠ with a good hand, so 2NT typically shows either a slam try, three small or a good doubleton in support (keep in mind my weak two's often show only five) or a GF hand in a different suit altogether, where quick tricks are vital.
I think you've very accurately pinpointed the heart of the issue, thank you. Linking to some hand evaluators (especially the Binky one - fascinating research no doubt, but why should I care about a statistical additive hand evaluator in this scenario?) or HCP or some other metric can wildly miss the point regarding what is being asked on a preempt-into-asking-bid scenario (and it is unfortunate that the standard answers are labeled 'minimum' and 'maximum', I would much rather call them 'sign-off' and 'positive').