[easy]

One of the forum members mentioned that they doubled the opponents in 3H on a hand and the hand was made. Some have expressed the opinion that his double was a no brainer. I dont see it as such. So i thought i'd start a new thread to discuss the pro and cons of doubling the 3h contract.

Before getting to the discussion of the hand let's recognize that sometimes doubles are made for pschological reasons as opposed to just basing the double on your hand and the information received during the bidding. You r playing against a person you know always overbids, or plays the hands poorly or some other intangible. Putting those issues aside let's examine the merits of double with this hand

the hand i believe was   AQT   K987  AKxx xx

You have what looks like 4 tricks in your own hand, 4 annoying trumps and pard has forced the opponents to the 3 level with a balancing bid in your short suit. Surely pard can contribute 1 trick on the defense. Thus it must be clear to double 3H ......correct?
Well i'm not so certain about that. in the face of the fact that rho has bid all by themselves to the 3 level without any support from partner would give me pause for thought.Remember rho knows he's missing those aces and kings also. What do i know about rho? is rho a sound player? Expert?Novice?world classs?The better my rho is the more i'm going to consider not doubling. I have a maximum nt yes but what has pard? Was pard just trying to push them a tad higher or does he have decent cards.
My hand is not an uknown to pard, i did open 1nt . They know that as well as i. Perhaps i should pass and see what pard has to say.It seems to me when you have limited your hand the double decisions rest in pards hand.
I have eaten lots of penalty doubles in my bridge career but the ones i hated the most were when i had to return to my teammates and read off -530.

What do we gain from the double. +100/300 as opposed to +50/100. yet we are risking -530/-630. My general rule of thumb is that if i dont think they r going down at least 2 i leave them alone.

Maybe im just  a chicken but i wouldnt double.

[keylime]

I'll append one of my axioms from my "rules of the road" page on my website (http://www.geocities...eylimeprecision) when it comes to this:  NEVER double a partscore IMP contract without strength AND a trump stack. It's the fastest way to demoralize your partner/team when returning to the table with a 530/730 to cross imp with a 140....that's a 9-11 IMP loss that in short matches might make the difference between a win and a loss.

Nothing wrong with competing, but double VERY VERY carefully.

[inquiry]

I was going to leave this thread alone, really.  As an aside, I don't remember anyone saying that a double was a no brainer, but what the heck, I will now. I thought the original poster of the given hand:

                  You
                  AQ10
                  K987  
                  AKxx
                  xx

could more than defend his action. However, since the only two post argue against the double, I think it is time for an alternative view.  

First the auction:

You  LHO  Pard  RHO
1NT  Pass  Pass  2H
Pass Pass  3C     3H
DBL  ALL PASS

Before one picks on the double, one has to decide what 3C means and if partner will take the 3Hx as takeout or penalty.

I will address the last point first. Partner could, as pointed out in other post, bid 2NT over 2H to try to get to either minor, and (they way I play, a double of 2H by partner in the balance seat here is also takeout, I realize that this is not standard american). So partner is showing a long suit club suit. We may have not that great a club fit, but I think on this auction, the assumption is that we have a club fit, so double of 3H can not be takeout. The rule I live by is that doubles are for takeout until we have found a fit, and showing better defense than offense after we found a fit for the giving bidding.

So let's look at the hand and auction. The 3C bid a prior establishes our fit given my 1NT opening bid. How can I have a better defensive hand. Small xx in partners suit, Four good trumps, and wonderful values in both side suits. Our hand is all defense values and virtually no special offensive value. This is a clear, unambigious double the way I play.

The second question is what kind of hand can the 3C bidder have? Certainly not a Heart void (would not pass 1NT). In fact with long clubs and enough value to bid 3C over 2H, I think partner would have bid 3C (using 4 suit relay) directly over 1NT. Thus, I believe the odds are long that partner has at least two hearts. This is further fortafided by partner could double with short hearts and willingness to pull, say 2S to 3C. The only problem with such doubles is if he is afraid you (opener) might pass the double with this type of hand. But if he is that weak, why is he bidding?

So what does partner have? It must be fairly one suited with clubs. And he must not be broke. He passed 1NT to make if he is bidding 3C now. A reasonable hands might be:

         S: Jx  H:xx D:Qxx C:KJTxxx

Would he be bidding 3C now with something like
          S:xx H: x D:QJxx C:QJT98x

I would think no. This hand (short, weak in both majors, and a no quick tricks) would surely pull 1NT to 3C to play before they entered the auction. So any really defensive weak hands with long clubs, partner would not leave 1NT in to begin with. And any really weak hand without a smattering of values, partner would pass 2Hs.

So I think 3Hx is going down on this auction almost every single time. With such good diamonds and spades, I might even start a trump. I think the real question is what kind of hand did my partner have for his 3C bid? I think if a mistake occurred, it was with the three club bid not the double.

And to address the issue of only doubling when sure down two. Boy, you must really be missing a whole lot of good results with tame passes. If your opponents are making the occassional doubled contract against you, you can be darn sure you are not doubling enough.  Especially in this modern, apply maximum pressure based upon the Law of Total Tricks kind of bridge played today.

[bglover]

Well. don't forget my original post had to do with being booted for making a double that wasn't going to hold. And my point was only that it was a reasonable double, and certainly I didn't deserve to be booted just because the double didn't work out.

Now, as to my partner's actual hand... He did, in fact, have a heart void, the QJ of clubs 6 times and nothing....NOTHING else. Sure, a good player immediately bids 2S (xfer) and is done with the hand... even half expecting that 3C may make given the shape and the expectation that partner has an honor or 2 in clubs.

Not this guy... So, when he bids 3C at this stage I did expect (1) a smattering of HCPs in clubs that might be useful on defense (remember I held xx) and (2) a hand that, if with no defense, is going to pull to 4C if I choose to double (for, why else would he bid if he didn't expect another action from me?)

I would expect a partner to find me with a heart stack (I had one) and make that bid as a push (he did) and then lo and behold there's that double he was hoping I'd make given that he pushed... Just because he passed 1nt doesn't mean he is totally broke, right?

In other words, Fred, I 100% disagree with you here.. He KNEW I had a stack.. he PUSHED them a level... to me, the whole purpose of his bid was so that I could then WHACK 3H and I did.

[inquiry]

As I expected, steve could defend the double himself AND the flaw in the hand was the 3CLUB bidders hand, not the doubler's hand. Six clubs to QJ, no other face card and a heart void? The 3C action can't be right. Bridge is a partnership game. How do  you expect your partner to bid logically when you can hold

xxx  void  xxxx QJxxxxx    or a more reasonable

Jx  xx  Qxx KJTxxx on this auction?

With first hand, try to bail in 3C to begin with, or let them play 2H. With the second, bid 3C and be delighted when your partner wacks 3Hs.

[keylime]

Not only do I stand by my post, but mathematically speaking it's the right course of action at IMPs. In a knockout match where I have a number of boards to make up a swing of 9-11 IMPS then I can see making a cooperative double and living with any bad aftereffects. However, at Swiss, not in my life would I double - I've been the recipients of too many gifts from opponents doubling my 3 of a major or 4 of a minor contract when it makes. Thusly this type of situation is very sensitive in nature, depending on the nature of the event and its duration.

It has been mentioned that the double makes sense because of LOTT. Well I couldn't disagree with you more. LOTT is an approximation of the lay of the cards. The less variance that all four hands possess the more accurate that LOTT becomes. However, with more distribution and shortages, LOTT principles become mathematically unjustifiable because the variance of the means and deviation of the four hands exceed recognizable limits of error. Statistically speaking, LOTT is not empirical based - it is simple a good approximation along the lines of Newton's method of determining roots of an equation.

I will say this though about the topic: it's certainly one that brings many different views into the mix.

[easy]

Quote

I was going to leave this thread alone, really.  As an aside, I don't remember anyone saying that a double was a no brainer, but what the heck, I will now. I thought the original poster of the given hand:

                  You
                  AQ10
                  K987  
                  AKxx
                  xx

could more than defend his action. However, since the only two post argue against the double, I think it is time for an alternative view.  

First the auction:

You  LHO  Pard  RHO
1NT  Pass  Pass  2H
Pass Pass  3C     3H
DBL  ALL PASS

Before one picks on the double, one has to decide what 3C means and if partner will take the 3Hx as takeout or penalty.

I will address the last point first. Partner could, as pointed out in other post, bid 2NT over 2H to try to get to either minor, and (they way I play, a double of 2H by partner in the balance seat here is also takeout, I realize that this is not standard american). So partner is showing a long suit club suit. We may have not that great a club fit, but I think on this auction, the assumption is that we have a club fit, so double of 3H can not be takeout. The rule I live by is that doubles are for takeout until we have found a fit, and showing better defense than offense after we found a fit for the giving bidding.

So let's look at the hand and auction. The 3C bid a prior establishes our fit given my 1NT opening bid. How can I have a better defensive hand. Small xx in partners suit, Four good trumps, and wonderful values in both side suits. Our hand is all defense values and virtually no special offensive value. This is a clear, unambigious double the way I play.

The second question is what kind of hand can the 3C bidder have? Certainly not a Heart void (would not pass 1NT). In fact with long clubs and enough value to bid 3C over 2H, I think partner would have bid 3C (using 4 suit relay) directly over 1NT. Thus, I believe the odds are long that partner has at least two hearts. This is further fortafided by partner could double with short hearts and willingness to pull, say 2S to 3C. The only problem with such doubles is if he is afraid you (opener) might pass the double with this type of hand. But if he is that weak, why is he bidding?

So what does partner have? It must be fairly one suited with clubs. And he must not be broke. He passed 1NT to make if he is bidding 3C now. A reasonable hands might be:

         S: Jx  H:xx D:Qxx C:KJTxxx

Would he be bidding 3C now with something like
          S:xx H: x D:QJxx C:QJT98x

I would think no. This hand (short, weak in both majors, and a no quick tricks) would surely pull 1NT to 3C to play before they entered the auction. So any really defensive weak hands with long clubs, partner would not leave 1NT in to begin with. And any really weak hand without a smattering of values, partner would pass 2Hs.

So I think 3Hx is going down on this auction almost every single time. With such good diamonds and spades, I might even start a trump. I think the real question is what kind of hand did my partner have for his 3C bid? I think if a mistake occurred, it was with the three club bid not the double.

And to address the issue of only doubling when sure down two. Boy, you must really be missing a whole lot of good results with tame passes. If your opponents are making the occassional doubled contract against you, you can be darn sure you are not doubling enough.  Especially in this modern, apply maximum pressure based upon the Law of Total Tricks kind of bridge played today.

Wow i thought you guys were asleep. I expected a response to this post much sooner but .......... better late than never

Yes ben you r rite i have missed out on some + 300 on occasion. But on the other hand i dont come back to the table with a frown on my face begging for forgiveness with a minus 530 very often, I would double 3d in a heartbeat at matchpoints but imps......... as i said i just dont double in this auction. Maybe if we were in a 48 board match or way behind, but in an imp pairs or swiss i leave it alone. Sure partner should have bid clubs immediately no question in my mind about that!! but I have a tuff time understanding why when pard gets us to a plus position (assuming 3d was going down) we need to punish him for his efforts. Wouldn't you say that this is a similar situation

1h p 2h 2s
p   p  3h 3s.

Must we double here after pard has done his job and gotten them in a minus position?

[bglover]

OK, one final word here... Declarer got incredibly lucky when his partners only working cards turned out to be KJ of spds in front of the AQ10 killing the defense.

My partner had 5 HCP-- QJ of clubs and the Q of dimonds... Did I not have the expectation that the bidder would be holding the spade cards as well for his frenetic bidding given that his hearts could only account for 7 HCPs?

And, since I have 4 decent hearts, he  SHOULD have some points somewhere. NOW, add into this that my partner now elects to make a club bid...AH more grist for this mill.. He doesnt have club winners...

In other words...that 3C bid gave me USEFUL information for doubling (sure I didn't X 2H but now 3 looks like a balls-on favorite to go down!)

Based on everything that I heard and could deduce (and that 3C bid was a HUGE part of my reasoning) double not only looks good, it looks INSANE not to...

So, ya I fully expected to find the K of spds in bidder's hand and thus expected 2 spd trix. OK, that didn't work out... I NEVER expected the AK of clubs to be there... and if partner had kept his mouth shut (or, better had xfered to clubs like he should have then shut up showing weakness), I NEVER would have doubled. I think you and Dwayne are being very narrow minded.

[inquiry]

Dwayne,

I don't mind reading your math if you want to post it... but even if not, let me address two points. First, the only person to mention LOTT in this thread was me. So let me address what I said, and what you said "someone" said.

"It has been mentioned that
the double makes sense because of LOTT.

Actually, at the end of my post addressing a comment by easy, I actually said:

"And to address the issue of only doubling when sure down two.
Boy, you must really be missing a whole lot of good results with tame passes.
If your opponents are making the occassional doubled contract against you,
you can be darn sure you are not doubling enough.  Especially in this
modern, apply maximum pressure based upon the Law of Total Tricks kind
of bridge played today."
This comment was well after the logic of doubling with this hand was over and was a generic remark concerning the general principle not to double unless sure down two. And alluded to the fact that pressure bidders using LOTT will often push you higher in highly comptetitive auctions (With LOTT driving their decisions, not yours). This comment had nothing to do with this hand, and my discussion never touched upon the law of total tricks at all. So no need to apply newton's method or standard deviations or any such stuff here. The justification for the double did not espouse LOTT on this hand, period.

Second, as for how lucky you are that people have handed you gifts. I think it has more to do with misunderstandings between your opponents. You are looking at them doubling you. That is the view from the other side of the table (what the 3H bidder must be thinking on this hand). We are dealing with rather we should double or not, and we assume our bridge logic is better than some random opponents who don't bid as well as we do. Just because you have been given gifts because of bad doubles, doesn't mean you have to try to avoid all but bone crusher doubles. All you have to do is make good doubles based upon bridge logic (not LOTT).  Let me give you a hand from just today.

You hold:
     S: T9xx
     H: Void
     D: AT8x
     C: KQ9x

If the bidding goes (EW VUL)
You  LHO  Pard  RHO
P     1C     1H     1S
P     2S      3H     3S
?

What do  you bid?

How about if it goes
You  LHO   Pard   RHO
P      1C    2H      2S
P      3S    P         P/4S
?

In the first case I would double 3S every time. In the second, I would pass 3S and also pass 4S. My hand hasn't changed, and partner has shown likely 6 to 7 hearts in both auctions. So while LOTT hasn't changed a bit on this hand, my bidding depends upon the reasonable and EXPECTED nature of my partners hand (one with defensive tricks or not). How do I determine this difference? By the auction of course. In the first case, partner bid 1H then to 3H by himself. With no defense, he would have made a WJO. So he is showing he can win some tricks in addition to his long hearts. In the second he made a WJO and may not be able to provide even one trick to the defense. But as far as LOTT goes, it is the same on both hands. One has to use bridge logic more than simple trying to apply a "four trump" double. If my partner bids the as in the first example with
              S:void
              H:KQTxxxx
              D:xxx
              C:xx

my doubling 3S will be very painful for us (as what happened to me today). And if that happens, take your partner aside after the game and explain how bidding really works. No doubt your "success" occur most often when your opponents participate is such sillyness But with any partner who knows how to bid, a double of 3S on the first auction will be very profitable with the given hand and vulnerabiity.

I have no wild desire to convince you of the correctness of this double, and no religious belief that double is the only option. I just think logically, this hand is the precise hand I would double with on this auction.

[keylime]

Mathematics eh? Well I can certainly provide that easily - it's making it understandable for the masses that might be the hard part.

LOTT is not exactly Larry Cohen's creation. The first writings about this subject was penned in 1969 by Jean Rene Viernes in the Bridge World, but really came forward to the masses in Cohen's various books on the theory of LOTT.

In essence the theory of LOTT is based on 4 principles: probability of two mutual inclusive events occurring at the same time, Baye's theorem, the principle of randomness, and a form of evaluation that is consistent for the target population.

The form of evaluation is of course some point counting scheme, but here lies the first of LOTT's many problems. Cohen and Viernes independently state that when a double fit exists, the possession (of lack there of) of honors in the bid suits, and/or the remaining distribution of the cards require the user to adjust up or down at least one trick (in some cases, two). Furthermore, double fits are much less likely to occur when the total number of trumps AND the degree of fits are ODD. The reason for this is simple: each hand holds an odd number of cards, and thus there are less opportunities to possess an odd number holding than there is to hold and even one due to Baye's theroem on conditional probability. Furthermore, your hand patterns contain either three odd numbers and an even one, or 3 even numbers and an odd one. Thusly there are multiple avenues to achieve an even numbered fit, thus most double fits and LOTT evals will involved an even number.

Randomness is assumed if the deck of cards is shuffled 7 times. The simplisitic reasoning is that 7 is not only a prime number, but a number distinctively different than other divisors. It is very difficult to determine whether a number is divisible by 7, unlike 2, 3, 5, and so on. With 52 cards in the deck each shuffle assuming it's perfect, with change the positions of the cards one position. As you shuffle against, those positions are dynamic, and it's been proven using many different means that 7's numerical attributes insure randomness (revolves around modular arithmetic and other topics).

Baye's theorem in the modern game is also known as theory of restricted choice. Basically, if declarer plays a card that is deem significant, it is assumed by conditional probability that they were forced to play that card and didn't have another option. Relating to LOTT, this means that when degrees of fit are found, they are directly contingent on how the other suits lie (in essence, fits are forced). When there exists sufficient shape, the percentages of probability become so distinct that no model of prediction can accurately predict the outcomes. This is why, and NOT the lack of trumps, that both Viernes and Cohen tell us to leave the 5 level to the opponents. That concept is not empirically based; it's based on the wildness of Baye's.

Lastly, let me remind you that LOTT isn't a bad concept. However like all treatments, I think we would agree that those treatments function at its most optimum level when applied correctly. Alas, we are humans, and thus we err, and err often. With the above hand, the problems started with the club holder who didn't appreciate the distribution did not guarantee success over a 3H furthering because the hand has enough error in variance to render a logical guess iffy at best.

I realize this is a boiled down version, but hopefully sufficient.

[nadroj]

given 1) your p didn't xfer to clubs originally and 2) your p freely bid to 3C, then i believe your double was at the very least understandable... given the same auction, and if you had the same hand 100 times, i'd expect you to come out the imp winner in the long run ...

you had every reason to expect 4 tricks in hand, given rho's bidding, and your pard's bidding doesn't come close to suggesting a total bust... i wouldn't be surprised if he showed up with both black kings

[easy]

Quote

OK, one final word here... Declarer got incredibly lucky when his partners only working cards turned out to be KJ of spds in front of the AQ10 killing the defense.

My partner had 5 HCP-- QJ of clubs and the Q of dimonds... Did I not have the expectation that the bidder would be holding the spade cards as well for his frenetic bidding given that his hearts could only account for 7 HCPs?

And, since I have 4 decent hearts, he  SHOULD have some points somewhere. NOW, add into this that my partner now elects to make a club bid...AH more grist for this mill.. He doesnt have club winners...

In other words...that 3C bid gave me USEFUL information for doubling (sure I didn't X 2H but now 3 looks like a balls-on favorite to go down!)

Based on everything that I heard and could deduce (and that 3C bid was a HUGE part of my reasoning) double not only looks good, it looks INSANE not to...

So, ya I fully expected to find the K of spds in bidder's hand and thus expected 2 spd trix. OK, that didn't work out... I NEVER expected the AK of clubs to be there... and if partner had kept his mouth shut (or, better had xfered to clubs like he should have then shut up showing weakness), I NEVER would have doubled. I think you and Dwayne are being very narrow minded.

Relax Steve we are here exchanging ideas. Your opinion is certainly as valid as mine. Maybe based on this discussion you will be more reluctant to double next time and i'll be more apt to double. Isnt that what an exchange of ideas is about Or maybe we'll ask ourselves who is sitting on the other side of the table from us. lol

My goal here has been accomplished.... a nice discussion about the pro and cons of a bid!!

[bglover]

I think perhaps you are missing my point... My hand's defensive values improved markedly by partner's unsolicited bid when we are about to be passed out in a part score (remember he took his bid in the passout seat).

That bid should be meant to show some values and meant as a push... If he is pushing, then it is with the expectation that i SHOULD double 3hts in this situation... that he has working cards for me.

You and Dwayne have tried to put this into a purely mathematical discussion-- doubling opponents into game results in high scores when it fails. ergo it is improper to do so mathematically.

My point is and has been, it partner takes an action, it is for a purpose. If, as here, someone decides to take his 1st call at the 3 level (after already annoucning he has fewer than 8 points), there must be a great reason for it. Since that reason can't be to compete for the part score (this is imps, not matchpoints), then the only logical conclusion is he is pushing them to the 3 level EXPECTING me to double given the right hand.

Well, I had the right hand... I got a tad unlucky in the spade suit but gosh, how much better should my hand be to double here based on this reasoning?

If you NEVER double partscores you are doing something very wrong. On the other hand, if your partner is doing his job in assisting you (as here, where I reasonably assumed he pushed them for a purpose and didn't just make a plain stupid bid to hear his own voice), then you can double frequently with a good expectation of positive imps.

[keylime]

I agree the double was right IF AND ONLY IF partner hadn't done what they did.  However, for partner to not transfer to get into a playable place, THEN balancing in....that's not exactly good partnership bridge.

[inquiry]

Easy said "Maybe based on this discussion you will be more reluctant to double next time and i'll be more apt to double." and now Dwayne said... I agree the double was right IF AND ONLY IF partner hadn't done what they did."

I think sharing ideas even without converting people is great, and easy said it best when he added...

    Isnt that what an exchange of ideas is about

[bglover]

Actually this brings up an intersting point. If I hold something like:

109x
9xx
x
AQJxxx

and my partner opens 1nt, I am more likely to pass and come in later with my club suit than to immediately transfer to clubs. I don't know, maybe I have played too many MP games in my life, but I have found that this hand usually plays better as a NT hand than a suit hand.

If the opponents do come into the bidding, I am not embarassed to bid 3c opposite at least 2 clubs and 15 points (and a quite possible King of clubs in partner's hand).

On the other hand, change my ace to the 10 and I transfer every time. I have no source of trix without clubs as trumps and when I do a minor-suit transfer its because I have garbage. Maybe everyone wont treat this hand the same way, but I have bid this hand this way for as long as I've played transfers, and I do this ONLY for the minor suits (change the clubs into spades and I texas the hand, expecting partner has a good chance to make 4).

This certainly was the type of hand, or suit, anyway, that I expected in my partner's hand when this occurred.

[keylime]

The exchange of ideas, a novel concept in our world today indeed...

Look at it this way, bridge is very much a game of interpretation.

[inquiry]

I disagree with Steve on the following hand.

109x
9xx
x
AQJxxx

He said he would pass opposite a 15-17 hcp. Not me, not today, not ever.

First you are laydown for 3NT opposite a 15 hcp minimum such as Axxx Axx Axx Kxx

The problem is if you play the transfer to 3C as absolute signoff.  An interesting line is to play 2S as transfer to 3C and 2NT as transfer to 3D (BTW, this is not how I typically play this, but my method requires longer discription). With "extra values" or "extra fit" you can bid the inbetween strain. For instance (1NT-2S-2NT), now you can then bid 3NT with this hand. I remember the good old days where 1NT-3C showed exactly this kind of hand. But you should have a way to bid these hands descriptively.

If I lacked a clever way to invite 3NT with this hand, I will simply make a normal 3NT invite without specifically showing the long club suit.

[easy]

I think i must be getting a little weak in the brain. Surely with all the computer geniuses here we can figure this problem out in no time.

Isnt it possible to take the hand in question and run a simulation to see what the most prevelant result will be for the given auction. Giving the club bidder a range of expected hands?

oh no i guess not this isnt a case of frequency it more a case of potential gain vs potential loss. Or is it.

ok im listening