[P_Marlowe]

Hi,

yes, partner made a trap pass, expecting
you to reopen the bidding in case East did
not raise.

With kind regards
Marlowe

[luke warm]

MickyB, on Sep 18 2005, 02:49 PM, said:

Hannie,

Would you not play 1♣-P-1♥-P, 2♥-P-P-X as takeout?

han might, but i don't think i would... my lho bid 1C, my partner passed(!) my rho bid 1H and i passed(!), my lho raised to 2H, partner again passes, and i double? what do i have? spades and diamonds? wasn't it safer to compete over 1H?

remember, partner could bid nothing over 1C... i am not strong enough, even with the other 2 suits, to bid over 1H... maybe the double should be takeout here, but it seems to me i can double 1H or simply pass all the way

[han]

I think I would like to play double as takeout in that auction, even though I couldn't double earlier. The fact that the opponents have found a fit and are willing to stop at the 2-level gives me a strong incentive to reopen.

[1eyedjack]

I play a fairly simple "meta" system of agreements on doubles with reg p.
This is born out of a realisation that more points are lost through misinterpreting the meanings of doubles than most other misunderstandings. The OP is a case in point. Looking at your own hand you can predict that the double is penalties. But having a system that relies on partner interpreting whether your X was t/o or pen depending on his own cards is a recipe for disaster, however clear it might be on this hand.

The general rules that p and I follow are to answer the following questions in priority:

Q1) Have we expressed a fit? If yes, then X = penalty If no, proceed to Q2
Q2) Are oppo in a fit at the 2 level? If yes then X - t/o. If no, proceed to Q3
Q3) Do any one or more of the following conditions apply:
... a ) Pass is forcing
... b ) Oppo are in NT
... c ) Either of us has expressed or implied length in oppo suit
... d ) Doubler trap passed when he had opportunity to double the same suit for t/o earlier
... e ) Either of us has preempted
... f ) Oppo are at level higher than 4H
If yes, then X = penalty
If no, then X = t/o

There may be some more subtle inferences about (eg) additional shape requirements for the t/o X etc, and there are additional agreements required to flesh it out, eg:

If we have doubled their 1NT for penalty, and they wriggle, pass of 2m is forcing but not of 2M (however long it takes to get there).
If we open 1NT we have NOT implied length in a suit that opponents overcall
If we overcall 1NT then we HAVE implied length in the suit overcalled (but not another), UNLESS oppo support their suit
If opponents give preference at 2 level by bidding, then a fit is NOT implied, but if they give preference by passing then a fit IS implied.

I expect that if I think hard enough there are other little add-ons that we have come to agree on in order not to disturb the basic algorithm.
You might want to add a proviso that if double is the only available game try after we have found a fit, then it is for take-out (personallyI just bid game :-))

We also have a few nasties that b*gger up the flow diagram a bit, for example
(1m) - P - (1-suit) - X we play as 3-suited t/o of opener, and 1N as 3-suited t/o of responder, and artificial X of strong 1C opener.

Anyway, applying this algorithm to the original post, I would have to conclude that the X is take-out, because the question "Are they in a fit at 2 level" comes as a higher priority in the questions than "did partner trap pass earlier". Well, some other posters have suggested that the X should be t/o anyway, so I guess that doesn't render the algorithm totally ridiculous (on this hand, anyway).

[Beto]

I hate to double part-score for penalties... but in this case...

Pass in 1♥ and then double after the raise means that partner would like to double 1♥!!!

If you play this double as takeout, you will get opps only when opener has 16+ pts and doubles for takeout... And with 16+ and a trap pass you probably make 3N anyway...

In this case, if partner want me to takeout he would bid 2NT which is clearly some kind of scrambling. I guess even 2♠ in this bidding should be something ART, because after passing 1♥ partner does not have spades.

[inquiry]

CASE ONE
1♦-(1♥)-Pass-(2♥)
Pass-(Pass)-DBL

versus

CASE TWO
(1♠)-Pass-(Pass)-2♥
(2♠)-?

Case one in penalty. Partner didn't have enough to keep the bidding alive over 1♥, opener has shown nothing extra. IF responder has ♣ and modest ♦ support he can bid 2NT here (scramble). IF he had ♠ he could have easily bid 1♠ over 1♥, or used neg dbl. Dbl can only be, logically, and bridgewise, penalty.

Case two.
If hand behind opener is loaded for bear, here he simply must pass the buck back to re-opener for a second reopening double. 2NT is needed as natural or some modest conventions (sound ♥ raise), so is not as easily explained as takeout. And to pass hands with modest values lets them rob you blind if partner can't balance a second time when you are short in their suit (if you are long, there is good chance HE CAN reopen on distribution if nothing else). BTW, I think 3♣ or 3♦ here instead of double should be fit non-jump. If you don't have enough to bid these suits immediately, you probably still don't have enough without a fit for ♥'s.

[inquiry]

But your partner will bid again, at least on some of these hands... and if he doesn't that will not necessarily be a bad thing. Remember, the more you have, the less your partner will have (yes, yes, you know)... I am particularily worried about the example with three spades and a heart void. I pass that one for sure.

I also pass the last one and trust partner to reopen if it is right. No need for me to be a hero here.

The first two hands are enticing, but I will pass with the first one, with the second, I would bid 3♣... it is not a stylized fit non-jump, partner maybe less than thrilled with the quality of my ♥ support (I had a spade mixed in with them).

Against that a nice fit-nonjump bid leaves you well placed on subsequent competition, something a nebuous 3♣ bid with doubleton honor support, void support for partner and KQ-sixth, and hidden diamond support with no heart support does not.

[1eyedjack]

inquiry, on Sep 19 2005, 09:43 PM, said:

CASE ONE
1♦-(1♥)-Pass-(2♥)
Pass-(Pass)-DBL

versus

CASE TWO
(1♠)-Pass-(Pass)-2♥
(2♠)-?

Case one in penalty. Partner didn't have enough to keep the bidding alive over 1♥, opener has shown nothing extra. IF responder has ♣ and modest ♦ support he can bid 2NT here (scramble). IF he had ♠ he could have easily bid 1♠ over 1♥, or used neg dbl. Dbl can only be, logically, and bridgewise, penalty.

Case two.
If hand behind opener is loaded for bear, here he simply must pass the buck back to re-opener for a second reopening double. 2NT is needed as natural or some modest conventions (sound ♥ raise), so is not as easily explained as takeout. And to pass hands with modest values lets them rob you blind if partner can't balance a second time when you are short in their suit (if you are long, there is good chance HE CAN reopen on distribution if nothing else). BTW, I think 3♣ or 3♦ here instead of double should be fit non-jump. If you don't have enough to bid these suits immediately, you probably still don't have enough without a fit for ♥'s.

Oddly, I play them precisely the reverse way around.

Whichever way you play them there are hands when you will regret the choice.

There will be hands that are either too weak or off-shape for an initial double but which can more safely double when you are protecting, the opponents have announced a fit and denied game interest.

There will be hands when you want to make a penalty double, but partner is still there to make a take-out double for you to convert.

What I don't want to get into is a situation where every sequence has to be decided on the particular merits of that sequence. It may be technically superior, if you have the memory for it, but I do not see it as being SO superior, contrasted with a much simpler and perhaps only marginally inferior algorithm that substantially reduces the risk of a misunderstanding (which would more than outweigh any theoretical merits if the misunderstanding only arose once every 20 occasions or so).