[Hanoi5]

If:

2♣-(3♣)-Pa

shows a better hand (a King better) than:

2♣-(3♣)-X

Then:

2♣-(3♣)-3♠

Shows at least 5 spades but not necessarily as good as in 2♣-(Pa)-2♠, right?

What about:

2♣-(3♣)-Pa-(Pa)
X

2♣-(3♣)-Pa-(Pa)
3NT

2♣-(3♣)-Pa-(Pa)
X-(Pa)-3♠

2♣-(3♣)-Pa-(Pa)
3NT-(Pa)-4♠?

[ArtK78]

I would say that 2♣ - (P) - 2♠ and 2♣ - (3♣) - 3♠ would be bid on the same hands. All of the hand where you have values and would bid 2♦ waiting (especially if you play 2♥ as an immediate negative) would be covered by the pass over interference, and all of the hands that are immediate negatives would double.

The hand that bid 4♠ in response to a reopening double is just showing long spades and minimal values (but not a bust, as the hand would have doubled first). The reason for bidding a delayed 4♠ is that if opener's 2♣ opening were based on a long suit responder was willing to allow opener to play in that suit. A direct 4♠ by responder over the interfering 3♣ call would show very long spades and almost no values - certainly nothing that would be of any value in a non-spade contract.

Opener's 3NT rebid should be based on tricks, not power.

Otherwise, his reopening double is based on a wide variety of hands, all of which are willing to play in the opponents'suit doubled if responder has the right hand. When responder takes out the double, he is just bidding his hand naturally.

[nigel_k]

Maybe there is some logic in turning everything upside down because we are in a force, but I find it so much less confusing to just bid normally, i.e. double is for takeout and a bid is stronger than a pass.

[gnasher]

Hanoi5, on 2012-October-18, 10:21, said:

2♣-(3♣)-3♠

Shows at least 5 spades but not necessarily as good as in 2♣-(Pa)-2♠, right?

Yes. Compared with an uncontested auction, there's less room to describe your hand, so more hands have to be covered by the immediate spade bid. The increased risk of preemption also argues for showing your suit while you can.

Quote

2♣-(3♣)-Pa-(Pa)
X

For takeout, but not closely defined.

Quote

2♣-(3♣)-Pa-(Pa)
3NT

Natural. Opposite known values, I think this should be a balanced hand unsuitable for a takeout double. With the source-of-tricks type, it's sensible to bid your source of tricks.

In the sequence where responder has denied values, 3NT might be either hand-type. Likewise if the overcall was 3♦.

Quote

2♣-(3♣)-Pa-(Pa)
X-(Pa)-3♠

Natural, could just be a four-card suit.

Quote

2♣-(3♣)-Pa-(Pa)
3NT-(Pa)-4♠?

I'd assume that it's to play, with six weak spades.

[PhilKing]

I don't think double by the 2♣ opener should be take-out.

[Codo]

Hanoi5, on 2012-October-18, 10:21, said:

If:

2♣-(3♣)-Pa

shows a better hand (a King better) than:

2♣-(3♣)-X

Then:

Possible, yes and 3 NT shows around 6-9 with a stopper.

Quote

2♣-(3♣)-3♠

Shows at least 5 spades but not necessarily as good as in 2♣-(Pa)-2♠, right?

I have no idea how you play 2 ♠ uncontested, but 3 ♠ is 5 or more and GF

Quote

What about:

2♣-(3♣)-Pa-(Pa)
X

2♣-(3♣)-Pa-(Pa)
3NT

[/

I have a quite strong opinion that the first show the typical NT hand for this opening without a stopper and the second the same hand with a stopper- or another hand which is confident to take 9 tricks- maybe a running suit and some bits and pieces....

Quote

2♣-(3♣)-Pa-(Pa)
X-(Pa)-3♠

2♣-(3♣)-Pa-(Pa)
3NT-(Pa)-4♠? [/

The first shows a GF with 4 spades- with the same hand and 5 spades you had bid first round.
The second looks impossible to me.

[gnasher]

PhilKing, on 2012-October-19, 07:05, said:

I don't think double by the 2♣ opener should be take-out.

What do you think it should be? And what do you do if you're dealt a takeout double?

[PhilKing]

gnasher, on 2012-October-19, 09:20, said:

What do you think it should be? And what do you do if you're dealt a takeout double?

1. Strong balanced, unwilling to commit to 3NT.

2. I would bid, of all things, the suit I intended to when I opened.

[mycroft]

I agree with Phil. Pass says "partner, I heard you, but I'm too weak to commit to game." Double by opener should say "okay, let's take our plus", shouldn't it? Where do I want to be with 22 BAL opposite a flat crappy hand? trying to take 5 tricks, or 9?

I would suggest that 2♣-3♣-p!-p; X-p-3♥ is perfectly legitimate and shows something like xxx J9xxxx xxx x; but with a flat yarborough, I'd be passing, thanking them for getting us out of trouble.

If I have a big takeout double, either it's balanced, or I shouldn't have opened 2♣ in the first place - three-suiters are a real pain to show.

[gnasher]

PhilKing, on 2012-October-19, 09:47, said:

1. Strong balanced, unwilling to commit to 3NT.

2. I would bid, of all things, the suit I intended to when I opened.

If I'd opened 2♣ on a 4342 shape, I would have been planning to bid 2♣-2♦;2NT, but after they overcalled 3♣ I'd make a takeout double. If I were 4234 with a club stop, I'd rebid 3NT instead of doubling.

Are you saying you would double on both, or if not where would you draw the line?

[PhilKing]

gnasher, on 2012-October-20, 03:16, said:

If I'd opened 2♣ on a 4342 shape, I would have been planning to bid 2♣-2♦;2NT, but after they overcalled 3♣ I'd make a takeout double. If I were 4234 with a club stop, I'd rebid 3NT instead of doubling.

Are you saying you would double on both, or if not where would you draw the line?

Not sure. I'll generate some hands and see if any rule of thumb suggests itself.

In my case 2♣ is GF and double of 3♣ would be 5+ no clear bid, positives ultra light in comp.

.....

Didn't spend that long on it. On a small sample it was massively best to just double basically all strong balanced and float the double if balanced.

[gnasher]

PhilKing, on 2012-October-20, 12:04, said:

Not sure. I'll generate some hands and see if any rule of thumb suggests itself.

Are you telling us that when you wrote "strong balanced, unwilling to commit to 3NT", you didn't actually know what you meant by it, so you had to do a simulation to find out?

Quote

Didn't spend that long on it. On a small sample it was massively best to just double basically all strong balanced and float the double if balanced.

That sort of result has very little probative value, because you haven't told us what the criteria were, or shown us any of the hands.

[lalldonn]

PhilKing, on 2012-October-20, 12:04, said:

On a small sample it was massively best to just double basically all strong balanced and float the double if balanced.

That's what my experience tells me as well. I have seen a number of penalties of very aggressive overcalls achieved this way.

[PhilKing]

gnasher, on 2012-October-20, 16:29, said:

Are you telling us that when you wrote "strong balanced, unwilling to commit to 3NT", you didn't actually know what you meant by it, so you had to do a simulation to find out?


That sort of result has very little probative value, because you haven't told us what the criteria were, or shown us any of the hands.

I am pretty sure I knew what I meant by the phrase "strong balanced, unwilling to commit to 3NT." I just didn't have in mind an exact percentage of hands I would double with and pass the double with. But so what? At the table I would have had to choose, albeit without access to the best modern research (or a slapdash 30 minute excercise in this case). I ran the sim partly because it struck me as unusual that you would recommend bidding four-card suits up the line over the double, which had not occurred to me.

I didn't write down any results, and because I don't use a "proper" simulation, the criteria are almost irrelevant. I just set a wide overcalling range (in the playbridge hand generator) so as not to miss any inferences (eg there was one 5-6 hand) and then came to my own conclusions as to whether each hand was a candidate 3♣ overcall. I didn't bother analysing more than 32 hands, because I don't like the sight of blood.

The chances are I will just double and pass quicker having looked at a few hands. And I'm only bidding 3NT with a 5+ card suit (which could be a major).