[MarkDean]

Scoring: IMP

---

1♠ (5♣) ?

What is your guess?

[rogerclee]

Double, take the money. But I was watching at the time.

[jdonn]

I double because that's what the one who was watching at the time does.

[whereagles]

hum... it's either dbl or 6♠ for me. Can't decide.... I'm leaning towards 6♠, but I must say I haven't been in this situation often enough to have a good enough experience on the subject.

[hotShot]

The choice is between 6♠ and dbl.
They are white we are red, how many down do we need for 1400?

I go for 6♠. If partner really opened with:
Jxxxx
Jx
KQJx
Ax
We will be down 1 or 2.

[pclayton]

Even I will double with this collection. Move my ♣K somewhere else and I might try something.

[neilkaz]

X as I think 6♠ is going down too often on a quick ♣ ruff or bad splits or whatever. 5♠ is OK but what if that goes down opposite PD's dead min and bad splits, or what if he doesn't bid 6♠ on hands where it's cold ?

I prefer X to 6♠ inspite of the colors.

[Halo]

6♠

Who knows, so let's be bold.

[karlson]

No one wants to be the hero and bid 6N? wtf

I would double too. If I thought there was some chance of getting to 7 opposite the right hand, I would bid.

[Halo]

karlson, on May 20 2008, 05:08 PM, said:

No one wants to be the hero and bid 6N? wtf

I would double too. If I thought there was some chance of getting to 7 opposite the right hand, I would bid.

Irony?

Lot of rule of 19 hands make slams. Why is +500 such a great target.

[mikeh]

I ran a small simulation. I assumed that RHO held 8 clubs or more and that partner held 11 hcp or more and 5+ spades, with spades his longest suit.

It will be no surprise that it took 1,000,000 hands to generate 24 that fit the constraints, so my apologies for the small sample size.

And I appreciate that there are those full-blooded preemptors who might bid 5♣ white v red with only 7 clubs (especially if 74 in some way).

I will check on that shape if I have time.

As it is, on the constraints as above, NS made 13 tricks 5 times, 12 tricks 12 times, and doubling was better 7 times (I didn't analyze for 5♠ compared to double.... I decided that we would either defend or bid slam) I didn't analyze the hands on which grand was makable, in any effort to see if it was biddable.

So far, the case for bidding slam seems to be about 17 - 7... which is a mild surprise to me. Incidentally, most of the hands on which slam failed had opener with 2 clubs, and the defence played club A and a ruff. One hand lost 2 trump tricks to A1098.

BTW, this is the laborious simulation type of which I wrote in a recent post, not a double dummy analyzer

I would have doubled, but maybe my gut is wrong.

Edit: I simulated another 49 hands, allowing overcaller to hold as few as 7 clubs. Some of them were hands that I doubt anyone would bid 5♣ on, but the vast majority were within the bounds of action I have seen or read about by reputable players.. and of the really borderline or sub-borderline hands, 2 would be hands on which slam is best and 2 were hands where double was best.. in those cases, as was generally the case where double was best, opener held 2 clubs, lacking the Ace.

Those 49 hands resulted in bidding slam prevailing 30 times and doubling being best 19 times. It is not surprising that allowing a 7 card overcall reduced the chances of slam by a small amount.

As it is, combined we still have a strong indication that slam is better than doubling: by 47 to 26. I will leave it to the mathematicians to argue whether these samples are large enough to give rise to statistically significant data.

[karlson]

It's imps, right? So we need to know how much 5c is going for on the hands where we can make slam. 1400 vs 1430 is a rather smaller swing than -100 vs 1100. Or am I missing something?

[benlessard]

My first insctinct would be to bid 5Nt PAS. Over 6C ill play it safe and stop in 6S over 6D ill bid 6S to suggest 6Nt or 6S and over 6H ill pass knowing i wont lose a slow trump that i might have lost in 6S.

But my main reason is I dont like to endup in 6S -1 when 6Nt is cold.

[Fluffy]

mikeh, I think that the fact that we double doesn't mean that we cannot end up in 6♠, at least to me double is take out, and we are gonna raise if partner is not balanced enough to pass.

[mikeh]

Fluffy, on May 20 2008, 07:49 PM, said:

mikeh, I think that the fact that we double doesn't mean that we cannot end up in 6♠, at least to me double is take out, and we are gonna raise if partner is not balanced enough to pass.

This is a very minority position. This is not like doubling a 4 level overcall. Partner will virtually never pull... at least, that is my view

[twcho]

mikeh, out of the 24 hands you generated (preemptor with 8+ clubs), how many boards 6NT is makeable while 6♠ fails?

I am thinking that 6NT maybe a better contract than 6♠.

[FrancesHinden]

mikeh, on May 21 2008, 12:40 AM, said:

I ran a small simulation. I assumed that RHO held 8 clubs or more and that partner held 11 hcp or more and 5+ spades, with spades his longest suit.

It will be no surprise that it took 1,000,000 hands to generate 24 that fit the constraints, so my apologies for the small sample size.

You need a better simulator. I have run exactly the same simulation, it took about a minute to produce 50 hands consistent with these constraints (about 1.5 million hands).

I stuck to the 8-card club suits; when I included 7-4s the vast majority of the hands generated didn't look like 5C bids.

Out of those hands

it's right to defend 5C on 16 on them
it's right to play a slam on 26 of them
partner will bid on if you double on 8 of them (and you will reach a making slam)

I didn't look at whether you were making 12 or 13 tricks because you are basically never going to be raised in 6.

5C is pretty unanimously going for 800 when you are making slam (7 club tricks), and quite often, with the occasional 500, when you aren't.

6S is making with 6NT off more often than the other way around. 6H is often better than 6S, but not often enough that we should bid it unilaterally.

We don't know what team-mates will be doing so it's hard to estimate the imps, but let's look at the aggregate scores. Mike's and my estimates indicate that bidding slam will result in making say 2/3 of the time of the time it matters, and going off one third of the time (ignoring hands where we double and partner bids). Doubling will be +800 2/3 of the time and +500 1/3 (say).

So double = +700
6S = +920

This is close enough given all the guesses involved that I don't know what the right answer is.

I've certainly been convinced that double is less obvious than I thought it was when I first saw the hand.