Thursday's suit combination

[luke warm]

imo it seems to be impossible to make if rho has H,10,x i think you have to play as if he doesn't... neither can you make if rho holds 10,x because lho holds K,Q,x and will rise on first trick... you can't make against H,x on the right because lho will then duck from H,10,x

 

so to me it seems you have to play in a way that makes the most against distributions that give you a chance to make *at all*... that means playing lho for either: H,10 or H,x or 10,x (it seems to me that you lose against any other distribution)

 

so low to the board... if lho plays an honor, lead the J next time to pin his 10... if lho plays the 10, play the J and then play the ace next time (it either wins else you were down all the time - i think)

[Guest Jlall]

well, i was going around in circles and mainly playing spy vs spy with myself. I finally decided, perhaps wrongly, to ignore what they play and quit thinking about what % of time they play the T etc etc because that gave me nightmares. So I decided that i was just going to pickup Tx, stiff T, or KQ tight on left meaning...

 

If they play an honor i will play them for KQ tight.

If they play the ten or low, i will play for Tx or stiff T.

 

This is slightly better than cashing the ace first

[han]

For the moment I agree with Jlall's solution.

[fred]

J987

 

A432

 

Jlall is indeed correct. The best you can do is to plan to pick up the following holdings with LHO:

 

10x

KQ

10

 

Although several people thought of the possibility of the 10 falsecard from 10x (well done to even consider this in my view!), as far as I can tell nobody mentioned what a bizarre play this is.

 

To me this falsecard flies in the face of one of the basic concepts of card play that we have all become programmed to understand. To me it is amazing to think that playing the 10 from 10x with J987 is the dummy is actually a good play!

 

Fred Gitelman

Bridge Base Inc.

www.bridgebase.com

[Fluffy]

ok I messed up, somehoe I forgot to remembe 10x is not a case but 2.

 

Anyway, I don't undestand why lowto 9 is better than the Ace, both seems to succed in 4 cases:

 

Ace succeds against:

 

K1065 - Q

Q1065 - K

KQ - 1065

1065 - KQ

 

While small to 9 and ace succeds against

 

10 - KQ65

105 - KQ6

106 - KQ5

KQ - 1065

 

 

Of course not taking into account that LHO didn't lead its 4th best on our suit :).

[mpefritz]

ok I messed up, somehoe I forgot to remembe 10x is not a case but 2.

 

Anyway, I don't undestand why lowto 9 is better than the Ace, both seems to succed in 4 cases:

 

Ace succeds against:

 

K1065 - Q

Q1065 - K

KQ - 1065

1065 - KQ

 

While small to 9 and ace succeds against

 

10 - KQ65

105 - KQ6

106 - KQ5

KQ - 1065

 

 

Of course not taking into account that LHO didn't lead its 4th best on our suit :).

While you can count the same NUMBER of combinations for which each line of play works, you really need to factor in the likelihood of each particular combination AND the likelihood a particualr player will play a certain card from that holding.

 

Again, in general, each individual 3-2 split is more likely that each individual 4-1 split, which in turn is more likely that each individual 5-0 split.

 

In your scenario, after cancelling out equal splits, you are left comparing a 4-1 split to a 3-2 split. The specific 3-2 split is a priori is more likely than the specific 4-1 split. So "small to 9, then A" wins over "A, small to 9".

 

fritz

 

p.s. I have not fully analyzed this combination, but am replying to this particular question.

[Stephen Tu]

Anyway, I don't undestand why lowto 9 is better than the Ace, both seems to succed in 4 cases

 

All cases are not created equal. Each particular 3-2 break represents 3.39% of hands, but a 4-1 break only represents 2.83%. So the intrafinesse is a sliver better.