With only 7 total trumps and a 4-2 split being more likely, I had
reasoned that LHO's 9 on the first round gave a slight edge to trying
to catch A9 tight on the left over 10 9 on the right (some measure of
restricted choice might have caused LHO to play the 10 the first
time). Was I all wet?
Playing to the J and back to the K makes 4 tricks 38.8% of the time.
Finessing the 8 on the way back makes 4 tricks 33.3% of the time if LHO
false cards from 109x all the time but is nearly equal if he never false
cards.
However if you need 3 tricks the finesse is slight favourite.
This isn't a pure restricted choice position, since LHO might have T9x and
RHO Axx. But that requires an interesting choice of card from LHO and a
weird duck from RHO. I think the finesse of the 8 is very reasonable,
catering for A9 (and a less likely bare 9) and losing to AT9, T9 and T9x. Of
the losing cases, only AT9 must have defended like this.
Tiggrr
Most defenders will play the 9 from AT9 rather than the 10. Playing
the ten risks fooling partner, and might cause him to miscount the
hand. It might be different if the auction placed declarer with
exactly five spades, but that's not the case. I wouldn't put much
stock in restricted choice here.
There are only two relevant holdings: AT9 and A9. Assuming you have
zero information about the distribution (seldom the case), these are
about equal in probability. I'd say you made a bad guess.
However, the fewer HCP your side has, the more inclined you should be
to play LHO for AT9, because he didn't make a takeout double. The
opening lead and play to this point would have a bearing, as would the
actual N-S hands.
Tim
begin:
J5
KQ863
3s, 9s, Js, 2s
now:
5
KQ86
missing cards: A, 10, 7, 4
16 ways these cards can lie
From RPbridge.net I get:
1 AT74 -- 1 110 4.78
2 AT7 4 1 143 6.22
3 AT4 7 1 143 6.22
4 AT 74 1 156 6.78
5 A74 T 1 143 6.22
6 A7 T4 1 156 6.78
7 A4 T7 1 156 6.78
8 A T74 1 143 6.22
9 T74 A 1 143 6.22
10 T7 A4 1 156 6.78
11 T4 A7 1 156 6.78
12 T A74 1 143 6.22
13 74 AT 1 156 6.78
14 7 AT4 1 143 6.22
15 4 AT7 1 143 6.22
16 — AT74 1 110 4.78
I will assume RHO would have played the Ace if he had it, so I
eliminate 9-16. #1 nothing works. #2,3,5,6,7 LHO doesn't play the 9.
Your line works with distribution #8, fails with #4... making it a
coin flip.
If you don't assume anything about RHO having the Ace...
#9 we eliminate because it doesn't make sense.
Your line works for #14,15... but LHO probably wouldn't have played
the 9.
But not for #10,11,12
Everything works for #13
You cant make 4 tricks with #16
Seems to me like you should not have finnessed. I did this out the
long way... so you or anyone else can tell me my error.
Rich Regan
I dismissed a duck by RHO from Axx as impossible, as it would throw
away a trick whenever LHO has KT9. This is why I said that only AT9
and A9 are relevant. So, the decision is very close, and OP didn't do
anything terrible. Singleton 9 is an interesting possibility, but I'd
guess most defenders would take the ace from AT7xx.
Tim
Basically, if you finesse you gain against S A9 and pay off to S
A109.
But LHO might have played the ten or the nine on the first round so
you can't count the S A109 situation fully where the nine was played.
Where as the S A9 counts as a full situation. So I think the finesse
is a better shot.
Eric Leong
T9x is also relevant because a false card would be automatic, to give
declarer a losing option with KQ8xxx.
David
It's certainly close. Assuming sound defense the pertinent
cases are AT9, T9, A9 and AT. (Even though AT is impossible
it must be included for restricted-choice reasons.) First two
are slightly more likely, so you went against the odds. Not
all wet... just a bit damp.
Note that T9x is illogical (RHO would have to win the ace else
pay off to KT9). So is a blank T or 9; RHO should win, as your
next play of the king is clear-cut (LHO may have T9x); but if
he ducks, it's close (as above).
--
Richard Pavlicek
Web site: http://www.rpbridge.net
East cannot have Axx as it would be an egregious error to duck. West
could have KT9 from East's POV.
Tim
But also playing LHO for exactly 109 doubleton is illogical when RHO
wins with the ace.
You gain with S 109 but you play off to three positions S 109x. So I
would think since 109 is eliminated and A10 is impossible the only two
situations to consider is A109 or A9.
Eric Leong