[ahydra]

Played this 3NT today (MPs):

---

Lead: S3 (4ths, so looks like 10xxx)

This hand has only 6 top tricks, so a lot of work to do, and you can tell a heart switch is coming the moment you let West get in. What is the standard thought process for analyzing hands where you need to conjure a lot of tricks? It seems very hard to visualize the entire 13 tricks of play to see whether the defence still have enough tricks if you can set up diamonds and run both pointy suits, so one might just blindly hope for that + the club finesse. But I feel like, at teams in particular where people bid games at the drop of a hat, I could really use a more analytical approach to tough declarer problems. Would appreciate any advice from experienced players on this, hence the choice of forum.

Thanks,

ahydra

[The_Badger]

There's a simple maxim I use when presented with this situation, although it wasn't invented by me, and I don't think it was a Bols Bridge Tip either, but some expert/world class player said it at some time or another, can't think who it was off the top of my head. Something like this...

If you find yourself in an impossible (or near to impossible) contract as declarer, visualise the opponents cards as favourable to you as possible and plan your declarer play on that basis.

[mikeh]

Some hands allow you to plan the play with some specificity. Others leave you with so many branching contingencies, and so few inferences, that you are in what I think Kantar called a 'scramble' mode. You form a tentative plan but you are aware that your future steps are really going to be dependent on what happens early on.

Of course one also has to make tentative assumptions. This is one area in which there can sometimes be a real difference between mps and imps. For example, unless you are doubled, you probably don't care too much at imps about how many undertricks you risk by assuming that the cards allow you to make (and it turns out they don't). At mps, however, if you are pretty sure that you can't make, then you may want to adopt the line that results in the smallest minus, even if doing so gives up on making. Of course, if one is in an unlikely contract, where much of the field will be safely lower, then one may well need to risk the undertricks, since playing safe for down 1 is terrible if the field is full of people scoring an effortless 120.

With that in mind: this looks like a scramble to me. I see no reason to think that we cannot make. I see no reason to think that this contract is not at least reasonably common. I wouldn't play safe for 8 tricks, as one example. Of course, in this case I don't have that luxury anyway.

We have 4 spade tricks. The first thing I want to do, if convenient, is find out how many spades LHO has. Since I intend to play on diamonds first, will want to lead the first round from dummy, and I am not going to worry about the stiff diamond King, it is convenient to win the spade Queen.

If RHO plays the 2, then I am pretty sure LHO has 4 spades. He led 4th best from 10xxx(x)...if from 10xxx then he won't usually hold 4 hearts. He may well hold a 4 card minor, since the auction suggests a major lead. Btw, reading RHO's card depends on whether they play standard. If they play upside down, then rho will play the 2 if he has a doubleton. If they play standard, he will usually give count so won't play the 2 with 2, and can't have the stiff 2 if LHO led 4th best.

I play a diamond towards the Q10x. Btw, I would have taken a 'normal' pause at trick one, before playing the Queen. I would then have called in tempo, but (I hope) reasonably quickly for the small diamond. This is perfectly ethical.

Some weak players will give away the position of the diamond King. Most good players will duck in tempo. So I'd need to know what I think of my opps and whether I detect a flinch. That will dictate my play. If I think LHO has 5 spades, that may affect the play as well: I can cater to 4-2 diamonds and score 3 tricks only if RHO has KJxx and I play low to the 10. Of course, I still have to get back to dummy twice more, which means the heart King onside and, likely, the club Ace as well.

All of these are factors that are running through my mind as I assess how to play diamonds, and how to assess my RHO's pace of play at trick 2.

If he flinches, I probably play the Q. But I'd need to be at the table. Against a Rodwell or Rosenberg, etc, I will have no clue at all, and probably would play the Queen. Against most opps, I probably play the 10 if there was no flinch: RHO never has a reason to flinch with Jxx(x), etc nor, usually, with KJxx, but may feel some doubt about Kx(x) or KJx.

If the diamond Queen loses to LHO, I am probably down but miracles happen. LHO may have the heart Queen, and now maybe I can scramble 4 spades, 2 hearts, find diamonds 3-3 or the Jack drops when I cash the Ace and the club King takes a trick.

Going back to trick 1, my first thought is to attack the suit in which, if all goes well, I gain two tricks. So diamonds. The longer I take at trick one, the longer a good opp gets to reverse engineer my problem and the smoother the opp gets to play. In real life, in most mp fields, most opps can't defend smoothly, and even fewer defend smoothly when the opps play quickly.

[cherdano]

I agree with most of Mike's post but I think he missed the most important reason to play the Q rather than the T (against strong defenders). Say LHO has ♣A and RHO has ♥KTx.
(It's similar if LHO has ♥K and RHO's clubs are strong enough for them to take 3 club tricks, say RHO has ♣AH9.)
Then we can make against clubs being Jxx on the left and K98 on the right by playing the Q, but if we switch the K and J we are always off, as LHO will set up hearts when in with the
(And if RHO has something like K9x in hearts he has to unblock the 9 on the first round.)

Basically, we want to keep LHO off the lead as long as possible, as otherwise they will switch to the round suit in which the finesse is off.

[gszes]

Defense is hard. Just because YOU can see the contract is umm err stammer poor does not mean it is obvious to the opps. I say this assuming you did not vomit or some other obvious act of ughh when the dummy appeared. So anything we can do to keep the defense in the dark the better. Trick 1 We cannot fool rho about lho spade length BUT we might be able to fool lho about rho spade length/power. We also have to take danger hand into consideration and LHO is definitely the danger hand.


Since 3 diamond tricks are almost necessary to make 3n AND we cannot keep lho off lead if they have the dia K it seems we need to play diamonds for 33 and rho the K.
This means I would win the spade K (to try and keep some mystery in spades) and low to the dia A and another dia to the Q since there really isn't any good play aside from 33 diamonds and K with rho and ducking with Kxx is easy anyway.

If the dia Q holds we no longer need mystery in spades so before playing a 3rd round of diamonds I would play a spade to the Q which has a high degree of probability of removing a safe spade exit from rho when they win the K of diamonds. If rho is out of spades they will have to provide an entry to dummy AND possibly your ninth trick.

[ahydra]

 cherdano, on 2019-January-31, 00:40, said:

I agree with most of Mike's post but I think he missed the most important reason to play the Q rather than the T (against strong defenders). Say LHO has ♣A and RHO has ♥KTx.
(It's similar if LHO has ♥K and RHO's clubs are strong enough for them to take 3 club tricks, say RHO has ♣AH9.)
Then we can make against clubs being Jxx on the left and K98 on the right by playing the Q, but if we switch the K and J we are always off, as LHO will set up hearts when in with the
(And if RHO has something like K9x in hearts he has to unblock the 9 on the first round.)

Basically, we want to keep LHO off the lead as long as possible, as otherwise they will switch to the round suit in which the finesse is off.

I think you meant s/clubs/diamonds/

- but this is the kind of insight I was hoping for, thanks. The point is that we can only let opps in once, with the CA. So we need to play a diamond to the Q, not the T like I did at the table. The T lost to the J and they insta-switched to hearts.

---

mikeh's advice is good as well. I always pause at trick one to plan the play as declarer or defence. Opps were unfortunately one of the better pairs in the club, and there was no discernable hitch or anything to go on when I led the small diamond. (Actually I've had a number of cases where less experienced people in the club hitch involuntarily and I misread the cards, so I'm never sure how much faith to put into reads like those!)

ahydra

[gszes]

 ahydra, on 2019-January-31, 16:38, said:

I think you meant s/clubs/diamonds/

- but this is the kind of insight I was hoping for, thanks. The point is that we can only let opps in once, with the CA. So we need to play a diamond to the Q, not the T like I did at the table. The T lost to the J and they insta-switched to hearts.

---

mikeh's advice is good as well. I always pause at trick one to plan the play as declarer or defence. Opps were unfortunately one of the better pairs in the club, and there was no discernable hitch or anything to go on when I led the small diamond. (Actually I've had a number of cases where less experienced people in the club hitch involuntarily and I misread the cards, so I'm never sure how much faith to put into reads like those!)

ahydra

note that if u start by playing low to the Q when u go back to the ace rho can unblock the K thus creating the dreaded entry to lho starting with the ace then toward the Q allows them no such luxury:)

[nige1]

 gszes, on 2019-February-11, 18:34, said:

note that if u start by playing low to the Q when u go back to the ace rho can unblock the K thus creating the dreaded entry to lho starting with the ace then toward the Q allows them no such luxury:)

Conceivably, a brave or desperate RHO might drop ♦K under ♦A.
Also, you are short of dummy entries, in case ♦ are 3-3.
As the cards lie, it might be better to win the opening ♠ lead in hand, to play ♦A and another
(although your chances in ♦ are severely reduced)