3S - X - 4S - ? All vul

[MrAce]

I agree with P.K. Expected imp win for bidding 5♣ being 5.3 is way too much imho. Also as he said there will be hands where we can't convince pd to stop at 5 if we bid direct. Though that might be a wash with the hands that we pushed them to non making 5♠.

[gszes]

5c

 

These problems are not always thought about the right way. There is

far too much concern for "making" something and insufficient worry

about the opps making their contract. At MP getting a positive result

will usually be good for some reasonable amount of mp. At IMPS one can

quickly go broke using that same philosophy.

 

What do we think our chances are of setting 4s? I think they are

probably over 60%. Does that mean we should pass since we also estimate

making game our way (opposite most min tox by p) would appear to be

less than 30%? If the opps make 30% of the time we are going to go broke

VERY fast letting them play there as they score up 1860 for every 800+ we

score on defense (over the course of say 10 hands). Even if we x 4s and get

1600 vs 800 we are still losing in the long run (and the opps will get

even more from making when we x). How about the reverse?

 

If we bid 5c and make 30% and go down 70% over the course of ten hands

we are -700 and plus 1800 even if the opps x and get 1400 we are still ahead

of the game. These assumptions assume p is close to min when they can be far

stronger.

 

I would opt for x playing MP but at IMPS I think 5c has much better long term

potential. By no means do I consider this hand a WTP pass since it is difficult

to imagine many minimum tox by p where we don't have a sound 60+% chance of setting 4s.

 

Bidding 5c now gives us opportunity to not only get the opps to bid 5s

(a contract I am much happier defending) but if p is a bit above average

we might easily make or go down only 1 (maybe 2 at worst). If p has a good

hand we might even be laying the groundwork for us getting to slam (probably

impossible to do if we pass 4s).

[suokko]

You need to treat the hands where partner would double again somewhat differently. On a couple of them, a direct bid of 5♣ would get raised to six in my world.

 

If you just had hands where partner would not double again the sim would at least have some validity. As it stands I think it is somewhat meaningless.

 

Also, looking through the example hands, I was surprised there were no 0(5)44 shapes. Also there were no 2452s, and partner seemed to have 5 clubs extraordinarily often.

 

Those hand types are so rare that they don't dominate scores. But far more important error is that 5♣ won't be doubled ever. But I updated the original simulation with requested parameter changes and increased production limit to thousand deals. I also added statistics for north holdings with these limits.

[cherdano]

There are also loads of hands with 6 diamonds that would double, hoping for 4H or 3NT. 1363, 1462, many (6322). I am also of the view that 2542 would double frequently.

[mike777]

I passed but I thought x was a close second going down one or two vul often and I was a bit too chicken.

[mikeh]

I did my own simulation, using Dealmaster Pro. My constraints were that opener held 5-10 hcp, 7 spades, fewer than 4 hearts, fewer than 5 in either minor. Doubler held 0-1 spade, and no 6 card suit, and at least 3 cards in the non-spades, along with at least 13 hcp..

 

I generated 300 hands and then visually inspected them. I rejected hands on which it seemed clear that N would reopen (maybe I should have constrained N to fewer than 18 hcp, as one example). I also rejected some hands on which I didn't think No would double initially....typically minimum hands with stiff H in spades and poor hearts.

 

There is a considerable subjective element in this, but I tried to be middle-of-the-road, and since I am innately conservative, I included for N some hands on which personally I might well have passed. I also included a few hands on which as N I would have reopened, but suspect that some here might not, since I may be slightly aggressive on the reopening. In other words, I tried to expand the field of hands beyond my personally preferred parameters, but not by much.

 

The result was 258 acceptable deals. I don't have the time to waste on this, so I only looked at making/failing stats.

 

4♠ passed out failed 72% of the time, making 28%. This was dd, since the programme uses deep finesse. I didn't take the time to analyze how much the dd aspect factored in, but while I am not a big fan of using dd analysis in most situations, I think it is a reasonable approximation on sims like this, where declarer, on both sides, is likely to have a fair amount of accurate information.

 

5♣ failed 77% of the time.

 

I didn't analyze how often it might get doubled, since that really would be far too subjective a process, and I didn't analyze the effect of multiple undertricks.

 

At imps, one needs to know the size of the swing as well as frequency, but when the failure rates are this high, it seems that passing is clearly correct.

 

I recognize that my constraints may not be to everyone's liking. However, I did do a run with N having as many as 2 spades and my quick impression was that East really never seemed to have a raise to 4♠, and often N didn't have much of a double. Obviously giving West as few as 6 spades would change the dynamic, but he is red in 1st chair, and trying to cater to what a 6 card 3-level red 1st seat pre-empt looked like was far too difficult.

 

Btw, I think that the fact that responder raised is a reason to suspect that partner lacks 2 spades. Had East passed, the simulation ought absolutely to include that possibility.

 

Thus I do not claim that this simulation provides a definitive answer, but I suggest that making N have 0-1 spades tilts the results in favour of bidding, and when even then bidding is a big loser, that is useful information.

[MrAce]

I did my own simulation, using Dealmaster Pro. My constraints were that opener held 5-10 hcp, 7 spades, fewer than 4 hearts, fewer than 5 in either minor. Doubler held 0-1 spade, and no 6 card suit, and at least 3 cards in the non-spades, along with at least 13 hcp..

 

I generated 300 hands and then visually inspected them. I rejected hands on which it seemed clear that N would reopen (maybe I should have constrained N to fewer than 18 hcp, as one example). I also rejected some hands on which I didn't think No would double initially....typically minimum hands with stiff H in spades and poor hearts.

 

There is a considerable subjective element in this, but I tried to be middle-of-the-road, and since I am innately conservative, I included for N some hands on which personally I might well have passed. I also included a few hands on which as N I would have reopened, but suspect that some here might not, since I may be slightly aggressive on the reopening. In other words, I tried to expand the field of hands beyond my personally preferred parameters, but not by much.

 

The result was 258 acceptable deals. I don't have the time to waste on this, so I only looked at making/failing stats.

 

4♠ passed out failed 72% of the time, making 28%. This was dd, since the programme uses deep finesse. I didn't take the time to analyze how much the dd aspect factored in, but while I am not a big fan of using dd analysis in most situations, I think it is a reasonable approximation on sims like this, where declarer, on both sides, is likely to have a fair amount of accurate information.

 

5♣ failed 77% of the time.

 

I didn't analyze how often it might get doubled, since that really would be far too subjective a process, and I didn't analyze the effect of multiple undertricks.

 

At imps, one needs to know the size of the swing as well as frequency, but when the failure rates are this high, it seems that passing is clearly correct.

 

I recognize that my constraints may not be to everyone's liking. However, I did do a run with N having as many as 2 spades and my quick impression was that East really never seemed to have a raise to 4♠, and often N didn't have much of a double. Obviously giving West as few as 6 spades would change the dynamic, but he is red in 1st chair, and trying to cater to what a 6 card 3-level red 1st seat pre-empt looked like was far too difficult.

 

Btw, I think that the fact that responder raised is a reason to suspect that partner lacks 2 spades. Had East passed, the simulation ought absolutely to include that possibility.

 

Thus I do not claim that this simulation provides a definitive answer, but I suggest that making N have 0-1 spades tilts the results in favour of bidding, and when even then bidding is a big loser, that is useful information.

 

Mike, as you said without knowing the imp exchange, the failure/success rate does not really mean much. And did you really visually inspect 300 hands? You need a life!!http://www.bridgebase.com/forums/public/style_emoticons/default/tongue.gif

[awm]

I think this is actually pretty close, and my decision would depend somewhat on what I think of my opponents.

 

Even using Mike's stats, it must be close to 50% that "either 5♣ or 4♠ is making" (23% + 28% minus a bit for the chance both games make). It will be very hard for opponents to double us in 5♣ on this auction, where they have a big spade fit and we have more than half the values -- they cannot know that I have a very flat hand if I bid in tempo. This means I'm probably losing 5 when both contracts fail and winning 11 when one of the two contracts makes (which is half the time). Even if we occasionally get doubled or occasionally one of the contracts goes down two, there is enough of a cushion here to compensate.

 

Of course it's also possible that partner pushes me to 6♣ (which will normally be a hand where he would have made a second double). But some of the time 6♣ is making too. And it's also possible that opponents take a push to 5♠ or even 6♠ (which will often be wrong on their part).

[mikeh]

I think this is actually pretty close, and my decision would depend somewhat on what I think of my opponents.

 

 

 

And it's also possible that opponents take a push to 5♠ or even 6♠ (which will often be wrong on their part).

Where do you find these opps? Red v red, they make us guess what to do at the 5-level, and then they assume not only that we guessed right, but that they have a good 5-level save?

 

Put it another way, if a good opp bids 5♠, I would expect them to make it quite often...they aren't saving on this auction if they are any good, at least imo.

[mike777]

fwiw I emailed this hand around to some good top players including WC.

 

All of them thought that this was a wtp double.

 

They expected pard to pull with a void or very offshape hand but sit very often with 1=4=4=4

 

However if the 3s bidder is 7-4 they concede this may make.

[whereagles]

WC standards are so low these days...

 

(joking lol)

[PhilKing]

However if the 3s bidder is 7-4 they concede this may make.

 

This seems like a horrible assessment of the chances of 4♠ making.

[mike777]

This seems like a horrible assessment of the chances of 4♠ making.

 

I take the blame for a poor choice of words or paraphrase.

 

here was the quote

 

 

"...Of course if 3S bidder is 7-1-4-1 or 7-2-4-0 4S makes..."

 

I was a bit surprised that is was a 100% unanimous response of double.

[PhilKing]

I take the blame for a poor choice of words or paraphrase.

 

here was the quote

 

 

"...Of course if 3S bidder is 7-1-4-1 or 7-2-4-0 4S makes..."

 

I was a bit surprised that is was a 100% unanimous response of double.

 

Yeah I guess they realise it makes plenty of other times as well. B-)

[MrAce]

OK, thanks all. So it wasn't clear what to do, at least I wasn't on another planet when I found it hard to decide without huddling for 5 minutes.

 

I first thought "Goodie, I've got stuff!", then I tried to picture what I need for game/slam to make. I then realized partner needs to have quite a perfect hand for us to make something 5th or 6th level, and then, instead of passing, I figured I have to tell pd I've got something, so I doubled.

 

I didn't expect double to be 100% penalty, and pd was a very good player, he wouldn't play me for a trump stack with opps bidding and raising spades against us. But he had the most boring 1-4-4-4 T/O double so he left it in, and opps made 4SX.

 

Other table bid 5C which went down 2 undoubled.

 

What went wrong with my judgement though was that I kinda took it for granted that we own the board, and didn't consider the possibility of them making as a strong alternative, in which case maybe I should have thought of bidding just because we don't know who makes what and we've got a good fit.

 

 

Can you please provide full hand?

[diana_eva]

[hv=pc=n&s=SJ4HAJ4D862CQJ854&w=SAK97652HQD53CT97&n=STHK982DAT97CAK32&e=SQ83HT7653DKQJ4C6&d=w&v=b&a=3SD4SDPPP&p=CAC6CQC9STS3S4SAD5DAD4D2H2H6HAHQSJSKC2S8S2&c=10]399|300[/hv]

[mike777]

Thanks Diana for a very interesting thread. I enjoyed reading everyone's posts. As a nonexpert player it looks like everyone had their call and I would have estimated both sides often go down one or even two tricks.

[eagles123]

it's a bit unlucky that it makes imo, I mean you just have to swap a heart and a club in the EW hands and they're probably bidding the same way and it's going -2