Suit Combination

[han]

AQ98

 

xxxx

 

(Motivated by a recent comment of hrothgar)

 

edit: must win 3 tricks.

[pigpenz]

how ace first then up to the queen thats where the deciscion is ;)

[han]

I can see three different lines:

 

A: low to 9, and when this loses to 10 or J, low to 8.

 

B: low to 9, and when this loses to 10 or J, low to Q.

 

C: Ace, low to Q (except when J or 10 drops offside).

 

One of these seems clearly superior to me (but I don't know the exact percentages).

[mikeh]

In the cases in which it is possible to win 3 tricks:

 

low to 8 then low to 9 loses to K10, KJ, J10 doubletons offside: 3 holdings

 

low to 8 then low to Q loses to stiff J or 10 offside, and K10 or KJ doubleton offside:

4 holdings

 

Cash A and then low towards Q, playing the 8 if a 10 or J drops on your left, loses to J10 tight, KJ10x (2 holdings) onside or KJ10xx onside or Kx (2 holdings) offside, for a total of 6 holdings

 

The clear winner is low to the 8 then to the 9. Playing the A first is a clear loser.

[kfgauss]

A nice problem. Here's my answer (hidden):

 

 

A is best, but there's a variant of C that's almost as good as A.

 

Line A is clearly better than B because B only beats A for Kxx-J10 precisely, and A beats B for all of the 4-1's (and the 5-0) except J10xx-K (which is picked up by any line).

 

[Edit: Oops, KJ10x-x is picked up by B, so A only beats B for the 5-0 and KHxx-H, where H=J or 10. These 3 cases are enough to beat one 3-2, though.]

 

A is better than 50% since Kx-J10x is picked up in addition to "two of the 3 honors onside", and C only beats A when K10 or KJ is offside (need 2 honors offside and we gain if J or 10 falls and K is there too, or if J or 10 doesn't fall and K is onside... in which case it's just Kx-J10x). A beats C in several cases: J10x-Kx (which makes up for the two losing cases above) and then the 5-0 and two of the 4-1's (KJ10x-x).

 

A better competing line, however is:

D: Ace, then low to the 9. This *never* loses to line C. You'd need the K onside, and more than doubleton, so there are only 2 cards offside. If they're J10, neither line works. If there are one or zero honors offside, you'll find out about the honors onside in time for your 2nd play from dummy.

 

This line D picks up the J10x-Kx in addition to everything C picks up. So we're left to compare: D beats A when K10 or KJ offside, and A beats D for the 5-0 and two of the 4-1's (KJ10x-x).

 

This is close enough that it'll be heavily influenced by outside data. In a vacuum, I think the three "more distributional" holdings eke out a win over the two 3-2's, so A is best.

 

 

 

Andy

[mikeh]

A nice problem. Here's my answer (hidden):

 

 

A is best, but there's a variant of C that's almost as good as A.

 

Line A is clearly better than B because B only beats A for Kxx-J10 precisely, and A beats B for all of the 4-1's (and the 5-0) except J10xx-K (which is picked up by any line).

 

[Edit: Oops, KJ10x-x is picked up by B, so A only beats B for the 5-0 and KHxx-H, where H=J or 10. These 3 cases are enough to beat one 3-2, though.]

 

A is better than 50% since Kx-J10x is picked up in addition to "two of the 3 honors onside", and C only beats A when K10 or KJ is offside (need 2 honors offside and we gain if J or 10 falls and K is there too, or if J or 10 doesn't fall and K is onside... in which case it's just Kx-J10x). A beats C in several cases: J10x-Kx (which makes up for the two losing cases above) and then the 5-0 and two of the 4-1's (KJ10x-x).

 

A better competing line, however is:

D: Ace, then low to the 9. This *never* loses to line C. You'd need the K onside, and more than doubleton, so there are only 2 cards offside. If they're J10, neither line works. If there are one or zero honors offside, you'll find out about the honors onside in time for your 2nd play from dummy.

 

This line D picks up the J10x-Kx in addition to everything C picks up. So we're left to compare: D beats A when K10 or KJ offside, and A beats D for the 5-0 and two of the 4-1's (KJ10x-x).

 

This is close enough that it'll be heavily influenced by outside data. In a vacuum, I think the three "more distributional" holdings eke out a win over the two 3-2's, so A is best.

 

 

 

Andy

Line D loses to J10 tight, x(2 cases) or void offside. If the Jx or 10x are offside, then the other low honour will pop on the second trick: but that could be from J10x or KHx.

 

So you must decide to play the Queen, losing to Kx offside, or duck, losing to Jx/10x offside. Assume you always pop the Q: you have 6 losing layouts (the 4 set out above and the two Kx combos). Assume you play RHO for one of those Kx combos and thus decide to duck, you lose to Jx or 10x offside, also 6 losing layouts.

 

Thus line D is markedly inferior to line A

[kfgauss]

Line D loses to J10 tight, x(2 cases) or void offside. If the Jx or 10x are offside, then the other low honour will pop on the second trick: but that could be from J10x or KHx.

 

So you must decide to play the Queen, losing to Kx offside, or duck, losing to Jx/10x offside. Assume you always pop the Q: you have 6 losing layouts (the 4 set out above and the two Kx combos). Assume you play RHO for one of those Kx combos and thus decide to duck, you lose to Jx or 10x offside, also 6 losing layouts.

 

Thus line D is markedly inferior to line A

Yes, that was pretty silly, wasn't it. Thanks for catching it.

 

Andy

[P_Marlowe]

The double finesse is your best

shot, at lossing only 3 tricks.

 

AQT9 opposite xxx

 

=> the double finesse will work 75% of the time,

 

AQ98 opposite xxx

 

its less, but still above 50%.

 

In short, if only one card is missing to a double finesse

layout, in your case the ten, treat it as double finess position.

 

With kind regards

Marlowe

[fred]

AQ9

 

xxxxx

 

Consider this related suit combination. How do you play for 4 tricks.

 

The answer is not something that I would characterize as being suitable for "Beginner and Intermediate Bridge Discussion"...

 

Fred Gitelman

Bridge Base Inc.

www.bridgebase.com

[Guest Jlall]

To answer fred's, I think low to the 9 then low to the Q. Because of the spots, we cannot pick up any 4-1 breaks, so 3-2's are the only relevant cases. A then low to the 9 if T/J drops wins against KT and KJ off, but loses to Kx(2 cases) off, and also JT off. So low to the 9 is better by 3-2 in terms of cases (could look up percentages).

 

You may be wondering why A then low up to the Q9 loses to Kx offside since you could duck, but if that was your plan you'd lose to Jx and Tx off.

[fred]

To answer fred's, I think low to the 9 then low to the Q. Because of the spots, we cannot pick up any 4-1 breaks, so 3-2's are the only relevant cases. A then low to the 9 if T/J drops wins against KT and KJ off, but loses to Kx(2 cases) off, and also JT off. So low to the 9 is better by 3-2 in terms of cases (could look up percentages).

 

You may be wondering why A then low up to the Q9 loses to Kx offside since you could duck, but if that was your plan you'd lose to Jx and Tx off.

Suppose you start by playing low to the 9 (the correct play). This loses to the 10 or J. On the second round, LHO follows with the only remaining small spot card.

 

You sure it is right to play the Queen now?

 

Fred Gitelman

Bridge Base Inc.

www.bridgebase.com

[Blofeld]

The principle of restricted choice says that it isn't.

 

Playing the Queen picks up Kxx onside.

 

Playing the ace picks up (Jxx / Txx) onside.

[Guest Jlall]

Suppose you start by playing low to the 9 (the correct play). This loses to the 10 or J. On the second round, LHO follows with the only remaining small spot card.

 

You sure it is right to play the Queen now?

 

Fred Gitelman

Bridge Base Inc.

www.bridgebase.com

Oh wow, youre right... that's amazing...supposed to play the ace restricted choice...

 

I wonder if I'd figure it out at the table? I hope so.

[fred]

Suppose you start by playing low to the 9 (the correct play). This loses to the 10 or J. On the second round, LHO follows with the only remaining small spot card.

 

You sure it is right to play the Queen now?

 

Fred Gitelman

Bridge Base Inc.

www.bridgebase.com

Oh wow, youre right... that's amazing...supposed to play the ace restricted choice...

 

I wonder if I'd figure it out at the table? I hope so.

Are you sure you are supposed to play the Ace?

 

I do know the right answer, but I have not stated what it is yet. So far I have just been asking misleading questions :P

 

Fred Gitelman

Bridge Base Inc.

www.bridgebase.com

[cherdano]

Not so fast guys. If we play the Ace after LHO follows with the remaining spot, he will change his strategy and always play the J or T from an original holding of Jxx or Txx. If we play the Ace in these cases, we lose to KJx or KTx onside. Since KJx and KTx are 6 holding altoghether, we can never afford that.

As long as LHO "falsecards" (this is so obviously safe that I am not sure this should be called a falsecard at all) at least half the time by playing the J or T from Jxx/Txx, it is still right to play the Queen when he follows with a low spot.

 

Arend

[Guest Jlall]

Oh...lho could play the T or J if he started with Jxx or Txx...OK back to playing the Q. Not fair to trick me when im playing poker :P

[Blofeld]

Sometimes I think this game is too tough for me.

 

Great situation!

 

[incidentally, does this imply that against a LHO you think would always play small-small from (J/T)xx, you should still be going up with the ace?]

[fred]

Sometimes I think this game is too tough for me.

 

Great situation!

 

[incidentally, does this imply that against a LHO you think would always play small-small from (J/T)xx, you should still be going up with the ace?]

Thanks. I am glad you liked this problem.

 

Don't worry. You are in good company. This game is too tough for everyone!

 

Yes, against an LHO who is not capable of making this mandatory falsecard (probably over 95% of the LHOs you will ever meet) you should play the Ace on the 2nd round.

 

Fred Gitelman

Bridge Base Inc.

www.bridgebase.com

[Stephen Tu]

[against opponents who have no idea that they are supposed to false card with (j/T)xx, following with x second round]

 

Go up with the ace ... only if you need 4 tricks (not max tricks) or are *sure* they falsecard < ~1/6. Since you lose an extra trick to KJxx/KTxx onside, your edge by going up with ace vs. non-falsecarders is very slim if you are looking for max tricks, only a bit over 1% vs. someone who never falsecards.

 

Is it really 95% of players? I often have trouble with deciding how much credit to give opponents. At what level of competition does it stop being right to deviate from the "book" line to cater to bad play when encoutering an unknown opponent? 1st day Blue Ribbon, 2nd day, 3rd day? Bracket 1 in a typical regional? Bracket 2?

[Trpltrbl]

AQ98

 

xxxx

 

(Motivated by a recent comment of hrothgar)

 

edit: must win 3 tricks.

The correct technicall play is small up to 9.

Followed by small to the 8, that will give you 56,7826 % for 3 tricks.

 

GBB :)

[Trpltrbl]

To answer fred's, I think low to the 9 then low to the Q. Because of the spots, we cannot pick up any 4-1 breaks, so 3-2's are the only relevant cases. A then low to the 9 if T/J drops wins against KT and KJ off, but loses to Kx(2 cases) off, and also JT off. So low to the 9 is better by 3-2 in terms of cases (could look up percentages).

 

You may be wondering why A then low up to the Q9 loses to Kx offside since you could duck, but if that was your plan you'd lose to Jx and Tx off.

Suppose you start by playing low to the 9 (the correct play). This loses to the 10 or J. On the second round, LHO follows with the only remaining small spot card.

 

You sure it is right to play the Queen now?

 

Fred Gitelman

Bridge Base Inc.

www.bridgebase.com

Yes, even though it is only 44,0870% of getting 4 tricks.

But playing the Q is correct play now.

 

GBB :)

[han]

Please provide more digits next time so that we can learn something from it Mike. :)

[fred]

Is it really 95% of players? I often have trouble with deciding how much credit to give opponents. At what level of competition does it stop being right to deviate from the "book" line to cater to bad play when encoutering an unknown opponent? 1st day Blue Ribbon, 2nd day, 3rd day? Bracket 1 in a typical regional? Bracket 2?

I was trying to be conservative when I used 95%.

 

I think that if you consider all of the world's bridge players, the number would be well over 99%.

 

I would suggest that you never assume that anyone has made a play like this unless:

 

1) You have heard of this person before

2) You have reason to believe that they are strong player

3) They appear to be paying attention

 

Or unless you know the person habitually gives remainder count of course :)

 

Probably most of the players who make it to the last day of the Blue Ribbon Pairs are capable of this play, but it is the sort of play that anyone could miss if they were not paying attention. I suspect that at least half of the players who made it to the last day of the Blue Ribbons would miss this play more often than not in practice.

 

You have to judge each situation as it arises. Some book plays are obviously harder than others and the very notions of "hardness" varies from person to person. In my experience the principle of "don't give your opponents a lot of credit unless they have given you reason to think otherwise" has worked well.

 

I think you will find that mistakes are a lot more common than good plays when you are playing against strangers.

 

Fred Gitelman

Bridge Base Inc.

www.bridgebase.com

[Guest Jlall]

On an amusing note, Marc Jacobus was playing against my dad once. He had to guess in the endgame whether my dad had made a REALLY good play, or a bad one that he shouldn't have made. He immediately played him to have made the REALLY good play, and was wrong lol. I informed him my dad had become senile :)

 

Even great players are probably at best equally likely to have made a great play as to have made a stupid one, so I don't generally give opponents a lot of credit.

[kfay]

Good thread