[gwnn]

wyman, on 2012-August-02, 07:05, said:

Maybe I was thinking about this wrong, but:

* it seems like it's right to balance here almost anytime LHO has a weak NT, since I doubt that if we hold > half the deck that we're going to get rich playing for 50s

* if LHO holds a 18-19NT, he'll dbl. Do we have a place to run? RHO is unlikely to hold a 4cM (am I wrong about this? Especially white, it seems like people are responding on air pretty frequently). I have tolerance for clubs and hearts, and the only way pard would run out to spades is if he has 5+ (although admittedly if he has 5 spades this could be ugly since his hand is bad enough that he didn't overcall, the fact that he didn't overcall 1S also makes it less likely that he has 5 spades, so this could be kind of moot).

* if LHO holds 12-17 with diamonds, it also seems like we're likely to have a fit and around half the deck.

The -100 v -90 argument is compelling, but it's kind of a narrow range. I wish I had a good sense of the probabilities here. e.g., if I pass it out, what's my expected score? if I bid 1N, what are the probs that LHO hits it and it floats? that LHO hits it and pard bids? that it floats? that LHO passes and pard bids? And what we expect to score up in each case. But I definitely don't think I do. Hell, I don't even know what partner's expected shape is.

I'm really interested in comments.

Suppose for a second that we could prove that on this particular hand, 1NT would work well if everyone at the table followed with their best possible double dummy action. That still doesn't mean that we should do it. What would your 1NT mean then? 10-15? 10-14? I'm not saying that 10-14 is irrational, maybe it's good, but there's definitely a tradeoff here. So even if pass here is "wrong", you have to think about the whole system imo. If you routinely bid 1NT here on 10 counts, it could well be that you will be hurt more when you bid 1NT here on 14 counts.

[wyman]

gwnn, on 2012-August-02, 07:25, said:

Suppose for a second that we could prove that on this particular hand, 1NT would work well if everyone at the table followed with their best possible double dummy action. That still doesn't mean that we should do it. What would your 1NT mean then? 10-15? 10-14? I'm not saying that 10-14 is irrational, maybe it's good, but there's definitely a tradeoff here. So even if pass here is "wrong", you have to think about the whole system imo. If you routinely bid 1NT here on 10 counts, it could well be that you will be hurt more when you bid 1NT here on 14 counts.

Actually, I tend to mark 10-15 for balancing NT (15s mostly happen over 1M -- they have a real suit, so dbl+1N seems like it should show more than a mediocre 15 with K3(2) in their suit imo). My attitude toward balancing NT is that -- especially at MP -- it's less of a constructive "maybe we have a game" measure as it is a "let's score better than defending" measure. And I don't balance with all 10s, but this seemed to me like a reasonable balance (again, at MP; I'm fine passing and taking +50 or +100 or -90 if that's what's coming our way at IMPs).

Back to the 10-15 range: there aren't a whole lot of 14-15 counts that can't act over 1D but where partner can balance 1N (fewer still where partner has ♦Axxx), so game is definitely not my first concern. Every partner I've played with and I have had the "don't hang partner for balancing" discussion. And even if you occasionally bid that 22-23 point 3N (at IMPs), at least the rest of the deck is pretty well marked.

[wank]

protecting here is criminal. you deserve for the opps to get to 4M.

over minors you should have a decent hand to protect because partner can happily double directly with weak NT type hands.

[gnasher]

wank, on 2012-August-02, 07:59, said:

protecting here is criminal. you deserve for the opps to get to 4M.

I think the risk of that is very small. When they try to stop in 1♦ the chance that they can make 4M is already very low. Our possession of two aces makes it lower. Our length in diamonds coupled with partner's failure to bid over 1♦ makes it unlikely that LHO has lots of shape, so even less likely that they can make game.

Quote

over minors you should have a decent hand to protect because partner can happily double directly with weak NT type hands.

That makes it less likely that he has an opening hand, but arguably it gives us more reason to bid. If the expectation is that he is in the 9-11 range, that makes it unlikely that he'll raise us.

[gwnn]

wyman, on 2012-August-02, 07:35, said:

Back to the 10-15 range: there aren't a whole lot of 14-15 counts that can't act over 1D but where partner can balance 1N (fewer still where partner has ♦Axxx), so game is definitely not my first concern. Every partner I've played with and I have had the "don't hang partner for balancing" discussion. And even if you occasionally bid that 22-23 point 3N (at IMPs), at least the rest of the deck is pretty well marked.

The problem is not partner having a 15 count, the problem is him having an 10-11 count and being interested in game if we have 15 but wanting to stop (very) low when we have 10. It has nothing to do with hanging partner, it's just that too wide ranges will lead to some bad contracts (as will too narrow ranges, because they will make some the range of some other action unduly wide).

[wyman]

gwnn, on 2012-August-02, 08:26, said:

The problem is not partner having a 15 count, the problem is him having an 10-11 count and being interested in game if we have 15 but wanting to stop (very) low when we have 10. It has nothing to do with hanging partner, it's just that too wide ranges will lead to some bad contracts (as will too narrow ranges, because they will make some the range of some other action unduly wide).

Fair point at IMPs for sure, but my instinct is that this is really a non-issue at MPs. I know I'm not getting pushy with a 10 count if partner balances 1N, but it wouldn't surprise me to learn that either this is a style thing or that it's bad bridge and I should be getting more aggressive with these hands. I think either way the edge is probably small, and I don't have a big enough sample to evaluate this.

I would typically be inviting with the good 11-12 hands, in which case we're still (pretty much?) fine if balancer has 10-11, especially since we have an opening bid (and lead) to place HCP and shape.

In any case, like I said, I definitely don't know enough/have a big enough sample to be confident in what I'm writing -- I'm just sharing my understanding and wondering what the heck I'm thinking the wrong way about if the rest of the bridge-playing world thinks I'm nuts for getting involved here.

If the argument is that you can't balance -- not because it's a -EV action with this hand, but because it's a -EV style that will hurt you in other situations, I guess I can accept that. But it still feels just like a leap of faith.

[gnasher]

At matchpoints I think it's OK to play the 1NT bid as about 10-15, but I wouldn't raise it with 11 or 12. I'd just pass, expecting to miss game occasionally but to get it back on the hands where we play 1NT instead of defending 1♦.

That doesn't mean I'd overcall 1NT on this 10-count at this vulnerability. It's not a very offensive hand, and the risk of losing 100 seems to great. I'd do it if we were non-vulnerable, though.

[CSGibson]

For me, you lose by balancing 1N: If you go minus instead of plus, if you push them into a better part score/game as a result, or if your +90 or +120 is not enough to compensate for +100 or +150.

In general, one of the improvements I've made in scoring recently is being less aggressive balancing over 1x-P-P auctions. Inevitably, they are not in the best contract, and LHO has a moose, sometimes distributional, sometimes balanced. For example, in this one he could be 5-1-6-1 or something with a game forcing hand that decided to open 1D because of the difficulties in opening 2C and describing the hand. If I always reopen light, he can get away with those tactics, if I don't he gets punished for his decision. I'm not saying never balance, but when I have length in their suit, a major suit hole, and a partner who has not taken any actions despite having an auction where it is fairly easy to enter, that's a big warning sign for me.

[Quantumcat]

No one seems to have noticed that you have length in the opponent's suit.

Bidding can only lead to bad things:
1. Partner think you have more HCP that you do and you go down in a contract.
2. The opponents play in 2♠.

Wouldn't you much rather score plus 50 or 100 than minus 50, 100 or 110?

[wyman]

Quantumcat, on 2012-August-02, 19:01, said:

No one seems to have noticed that you have length in the opponent's suit.

Bidding can only lead to bad things:
1. Partner think you have more HCP that you do and you go down in a contract.
2. The opponents play in 2♠.

Wouldn't you much rather score plus 50 or 100 than minus 50, 100 or 110?

1. You don't need to underline; it detracts from your point.

2. Everyone sees the length in diamonds.

3. Sentences like "bidding can only lead to bad things" are wildly hyperbolic if not incorrect.

4. Why are you convinced opps have a spade fit? I'm not.

5. I don't understand why partner will think I have more points than I have if our agreement allows me to balance with some 10s.

[Quantumcat]

1,2,3 Ok, I understand, all fair points.

4. Everyone thinks it likely that partner has a good hand. If the opponents have a diamond fit and not a spade fit, then partner would have made a take-out double, or overcalled 1♠ or 2♣ (when he has 5 clubs 4 spades and 2-3 bad hearts so he doesn't want to make a take-out double). If it turns out that partner does not have a good hand (so he might not bid with the shapes described), then they probably have a better contract available which they'll find, if it turns out they do not have any good fits then we don't either, they can punish us in a bad-fitting trump suit or punish 1NT if it's their hand.

5. If you have a well-defined range that includes 10 - say, 10-12 - then you have no bid with balanced 13-15's, and those are the hands you are not going to want to pass out, unlike this one. If you have a wide range, like 10-14 (and double and rebid NT with 15-18 or so), then partner always has to guess what end of the range you're on and you will at least sometimes miss games or get too high. So either you are outright lying (with the usual 12-14 range) or (with a 10-14 range) you just hope that today partner is depressed and feels very pessimistic about everything. And you also hope next hand you don't pick up a nice 13 count and have the same sort of auction!

[wyman]

Quantumcat, on 2012-August-02, 21:10, said:

4. Everyone thinks it likely that partner has a good hand. If the opponents have a diamond fit and not a spade fit, then partner would have made a take-out double, or overcalled 1♠ or 2♣ (when he has 5 clubs 4 spades and 2-3 bad hearts so he doesn't want to make a take-out double). If it turns out that partner does not have a good hand (so he might not bid with the shapes described), then they probably have a better contract available which they'll find, if it turns out they do not have any good fits then we don't either, they can punish us in a bad-fitting trump suit or punish 1NT if it's their hand.

I guess I just think it's super likely that spades are LHO 4, CHO 4, RHO 3, since RHO didn't bid 1M and since CHO didn't overcall.

And if it's a giant misfit hand (which I think is pretty likely -- 4234?) and partner has some values (also pretty likely), we can just play 1N.

I just don't see why the only options are: they find their spade fit / we decide to bid into a misfit over 1N and get punished / they rip us in 1N.

[CSGibson]

wyman, on 2012-August-02, 21:48, said:

I guess I just think it's super likely that spades are LHO 4, CHO 4, RHO 3, since RHO didn't bid 1M and since CHO didn't overcall.

And if it's a giant misfit hand (which I think is pretty likely -- 4234?) and partner has some values (also pretty likely), we can just play 1N.

I just don't see why the only options are: they find their spade fit / we decide to bid into a misfit over 1N and get punished / they rip us in 1N.

Why would you assume that RHO only has 3? I wouldn't automatically bid with a 4 card major unless I had a hand where I could deal with partner responding 2N. A 4 card major is a MUCH different prospect than a 5 card major in my opinion. Some people won't respond with "inadequate" values even with a 5+ card major, I don't think OP specified the strength of the opponents with anything more than adequate - hard to know whether adequate means aggressive response or not.

[wyman]

You can stand 2n with like 3hcp though. So the times rho has a 4cM he has 0-2 hcp so pard has 9-12. Even then LHO might not have a fitting major.

Maybe I'm wrong -- I just felt the "they will play 2s" was way overstated.

[gnasher]

Quantumcat, on 2012-August-02, 21:10, said:

4. Everyone thinks it likely that partner has a good hand.

I expect partner to have something in the 9-14 range. I'm not sure if that's what you mean by a good hand.

Quote

If the opponents have a diamond fit and not a spade fit, then partner would have made a take-out double, or overcalled 1♠ or 2♣ (when he has 5 clubs 4 spades and 2-3 bad hearts so he doesn't want to make a take-out double). If it turns out that partner does not have a good hand (so he might not bid with the shapes described), then they probably have a better contract available which they'll find, if it turns out they do not have any good fits then we don't either, they can punish us in a bad-fitting trump suit or punish 1NT if it's their hand.

The idea that he'd routinely make a vulnerable overcall with a 4225 shape fills me with horror. Do you really expect him to bid over 1♦ with Axxx Kx xx KJxxx? Similarly, a 4324 10-count isn't worth a takeout double.


Regarding the possibility that they have a spade fit, even if they have one it doesn't seem at all likely to me that they'll find it. The hand that passed 1♦ won't be backing in with 2♠, so opener will have to act.

One possibility is that opener doubles 1NT with a balanced 18-19 and responder bids 2♠ with a five-card suit. But if they're going to have that type of sequence, the 1NT overcall will gain at least as often as it loses, because we can expect to pick up a penalty on the hands where they have that sequence and our partner happens to have four spades.

Of course there will be occasional hands where opener has the strength and shape to bid 2♠ by himself, but those will be very rare, especially given our diamond length. I don't think that's a significant factor in deciding whether to balance on this hand.

[JLOGIC]

wyman, on 2012-August-02, 07:05, said:

Maybe I was thinking about this wrong, but:

* it seems like it's right to balance here almost anytime LHO has a weak NT, since I doubt that if we hold > half the deck that we're going to get rich playing for 50s

* if LHO holds a 18-19NT, he'll dbl. Do we have a place to run? RHO is unlikely to hold a 4cM (am I wrong about this? Especially white, it seems like people are responding on air pretty frequently). I have tolerance for clubs and hearts, and the only way pard would run out to spades is if he has 5+ (although admittedly if he has 5 spades this could be ugly since his hand is bad enough that he didn't overcall, the fact that he didn't overcall 1S also makes it less likely that he has 5 spades, so this could be kind of moot).

* if LHO holds 12-17 with diamonds, it also seems like we're likely to have a fit and around half the deck.

The -100 v -90 argument is compelling, but it's kind of a narrow range. I wish I had a good sense of the probabilities here. e.g., if I pass it out, what's my expected score? if I bid 1N, what are the probs that LHO hits it and it floats? that LHO hits it and pard bids? that it floats? that LHO passes and pard bids? And what we expect to score up in each case. But I definitely don't think I do. Hell, I don't even know what partner's expected shape is.

I'm really interested in comments.

If partner has values, he likely has diamond length (as in, 3+ diamonds). Gnasher has gone over the cases where this might not be true, a 4324 hand that is too light to double (let's say a 10 count), or 4225 with 14 (though some of those would overcall if they had either very good clubs or very good spades, still pass would be normal).

The other times, partner has either 3+ diamonds, or few values (less than 10). For my money when I have no game, I'd rather defend 1D when I have 7 or 8 diamonds than declare 1N. This kind of comment:

Quote

And if it's a giant misfit hand (which I think is pretty likely -- 4234?) and partner has some values (also pretty likely), we can just play 1N.

is typical but I don't see why it's not considered a loss that we failed to defend 1D and ended up declaring. It might still be right to play 1N, for instance if we have 8 tricks in NT and the same 8 tricks on defense against diamonds, 120 will beat 100, and if we have 7 tricks against both 90 will beat 50. But I would guess that we rate to take an extra trick with diamonds as trumps when we have 7 of them, and certainly when we have 8 of them. Not to mention we have to be making 1N to begin with which doesn't seem like a lock, especially if they make a good lead. If we are going down in NT then certainly passing will work better, and if we are making 1N I think partner will sometimes raise to 2. Passing 1N with an 11 count would be a strange decision to me (maybe because I do not bid 1N that often with a 10 count?), and certainly with a 12 count. It's not like our hand is that good.

And if partner has short diamonds with too few values to bid (which is not that unlikely, LHO could just have 18-19 bal and often will when I am this weak and patner passes imo), then that will really be bad if we are going down in 1N whether they double or not.

I guess if you think that declaring 1N or 2N will be better on average than defending 1D on these "normal" hands then bidding 1N is fine, I don't.

[gnasher]

Part of the reason for wanting to bid is that one-level partscores are quite hard to defend. There's often some uncertainty about why we're defending, what our trick target is, and which side-suits we should be playing. In contrast, 1NT with all the opposing high-cards in one place is usually fairly easy.

[JLOGIC]

I agree, on the other hand playing 1D with 5 trumps between you is usually hard.

I guess it depends on RHOs tendencies for passing w/r. Personally I don't do that, but I guess it's possible he would not pass with a stiff but would with a doubleton which would change the likelihood partner has 4 diamonds.

Even with 3 diamonds and an opening bid I think partner would often have bid unless he has a doubleton heart, but that is also a style thing. I agree that the less often partner will make takeout doubles, the more aggressively we should balance.

Playing 1N with very few HCP is also hard, it is still definitely live that we have 17-19 HCP with no fit (except diamonds?), and where most leads they make will be favorable for them. I would bet on them beating us in those cases, regardless of how hard it will be to defend.

Lastly, playing 1D with all the HCP in one hand is also hard, and defending accurately becomes much easier (part of the hard part of defending 1 level partials is having no clue what to lead, but seeing a very weak dummy makes this much easier). I still like our chances against diamonds when we can make 1N if we have 7+ diamonds.

I dunno, I'm still unconvinced about balancing but there are definitely a few issues that are cyclical (one person can say partner won't invite with 11 since they balance often in this spot since they think it's right, another might say partner will invite with such hands since they don't balance with this hand type, one person might say partner will often bid in direct seat with short diamonds or with 3 diamonds and an opening hand, another might say partner doesn't have to since they will get in the bidding including balancing aggressively...etc etc). On top of that we might make assumptions about RHOs pass tendencies that could fit either argument when really we don't know. Against a random I'd think they'd have less than 6 and most hands (or less than 5 and all hands), against a non random I might think they wouldn't pass with a stiff diamond, or that they wouldn't pass with like 0-3 points, whatever.