[kgr]

You play 4♠ after a short auction (1NT-2H-2S-3NT-4S) and have in ♠:
KJxxx
Axxx
The best chance for no looser is play from the top. But when will you take the finnese instead of playing from top? (assume no other information):
- LHO did not start trump. That will increase the probability that LHO has ♠Q enough to take the finesse.
- LHO did not start trump and did start with a side Ace without having the K. Apparnatly he did have a difficult lead and that will increase the probability that LHO has ♠Q enough to take the finesse.
- You had another bidding that did going back to Spade after partner suggested to play 3NT. Apparantly you have some ruffing values, but LHO did not start S.
- ....

[Free]

Since I have no information if this LHO is a good player or not, no information about both entire hands (since when can't I see a full size dummy when I need to?), and no other relevant information at all (not leading trump could be void, what's ruffing value in a balanced hand, leading an ace could be singleton or doubleton hoping for a ruff,...), I'll take the percentage play...

[pigpenz]

Free, on Nov 11 2005, 07:17 AM, said:

Since I have no information if this LHO is a good player or not, no information about both entire hands (since when can't I see a full size dummy when I need to?), and no other relevant information at all (not leading trump could be void, what's ruffing value in a balanced hand, leading an ace could be singleton or doubleton hoping for a ruff,...), I'll take the percentage play...

the percentage play is actually what?

playing AK is about 46%
Playing A hooking J♠ incorporates more than 50% in fact almost 56% according to pavlicekts calculator.

The new bridge world had just this sort of article in it about restricted choice when opp plays say the 10 or 9♠ the odds change dramatically

[Free]

Ace and hooking the J is less than 50%, because you need a 3-1 split (one of both possibilities, or stiff Q) which is a lot less! Cashing AK is the percentage play...

[pigpenz]

are you really sure about that?
on all 2-2 50% of the time the Q is on side, and hooking jack will incorporate xxx-q
and all qxx-x combinations....so
just basic
3-1 50%
2-2 40%
0-4 10%....


so 50% of 40% is 20%
75% of 50% is 35%

that is 55%

playing ace and then to jack
is 40% plus and combinations where its stiff queen and 4-0 break onside provided we have 10 or 9.

[ArcLight]

Why would you expect the defense to lead a trump against that sort of auction?

>(1NT-2H-2S-3NT-4S)

Implies declarer has at least 3 Spades. H emay have 4 with a minimum too.
If declarer has 8 or 9 trumps I'm sure he would appreciate a trump lead.

[luke warm]

we've been here (sorta) before... which is more likely, a 3/1 split or a 2/2 split? all things being equal, 3/1 is about 50%, slightly greater than 2/2.. i had a hard time understanding "9 ever" for that very reason

it always seemed to me that (using this hand for example) if you play ace and small, and if LHO follows, hooking the jack is the % play since LHO has already shown up with 2 cards and since he's shown 2 the odds favor him starting with 3... but i've been convinced that i'm wrong

[kgr]

Jlall, on Nov 11 2005, 09:39 PM, said:

Free, on Nov 11 2005, 02:25 PM, said:

Ace and hooking the J is less than 50%, because you need a 3-1 split (one of both possibilities, or stiff Q) which is a lot less!  Cashing AK is the percentage play...

There is no percentage play. There is just one in a vacuum. The original poster knows the vaccum percentage play, and is wondering whether or not these inferences swing things in favor of finessing.

Short answer: it depends.

Right. (Glad to see you have some confidence in my bridge knowledge )
I assume that difference in % between both plays is very small (I guess about 2%). So I was wondering what indications were enough to take the finesse instead of playing for the drop.
F.i. I would think that if the bidding calls for a trump start (even slightly) and LHO does not start a trump and you can finesse over him then a finesse is probably better then play for the drop.
------------------------------
Another similar remark:
If I have in trumps:
AJxx
KTxx
and LHO did not start trumps and without any other information then I will take the finesse over LHO. The only information I have is that LHO did not start trumps, and maybe the reason is that he has Qx or Qxx. It that is the only info I have, then I will use that.

[zasanya]

The rationale behind '9 ever rule' is as follows.
When everybody follows to first trick you know the split is not 4-0.So the odds of 10% in favour of 4-0 are meaningless.Next trick one opponent follows so you know if the break is 3-1 the one who followed did not have the singleton .Now the only question is 'who has the missing queen?'.
The opponent who is yet to play has one extra unknown card than the opponent who has followed.Hence the odds that the missing Queen will be in the hand of the opponent who is yet to play is more than 50%.Therefore play for the drop.
The additional information you get with every card played is the rationale behind 9 ever rule.

[zasanya]

Correction
Wherever I said 9 ever read 9 never :-)

[Winstonm]

Here's an odd side thought on this - so what else is knew, right?

Could there be instances where a trump lead is the stand out but restricted choice disallows a singleton lead? More likely in a 5/3 fit I would think.

[DelfinoD]

pigpenz, on Nov 11 2005, 02:43 PM, said:

so 50% of 40% is 20%
75% of 50% is 35%

that is 55%

Well in my opinion 20%+35%=48%

[P_Marlowe]

Hi,

playing for the drop is the percentage
play, but the difference is only very
tiny. (*)

If you would try to make your living
from this difference, I would say that
there are easier ways to earn money,
i.e. if you are always finessing, you wont
loose a lot.

Any additional information, e.g. knowledge
about the existence of a 5 card side suit
will make, that the percentage play is the
finesse.
I am not sure, if a 5 card suit is suffienct,
or if both plays are equal in this case.

With kind regards
Marlowe

(*) One specific 3-1 split is less likely than the 2-2
split, but both uneven spilts are more likely that the
single 2-2 split.
This sentence holds for all splits involving an even
number of cards, except the case that two cards are
missing, i.e. 1-1 is more likely than 2-0 splits.

[kenberg]

pigpenz, on Nov 11 2005, 02:43 PM, said:

are you really sure about that?
on all 2-2 50% of the time the Q is on side, and hooking jack will incorporate xxx-q
and all qxx-x combinations....so
just basic
3-1 50%
2-2 40%
0-4 10%....


so 50% of 40% is 20%
75% of 50% is 35%

that is 55%

playing ace and then to jack
is 40% plus and combinations where its stiff queen and 4-0 break onside provided we have 10 or 9.

Everyone starts with ace and a small spade, there is only a decision to be made if everyone follows small. Stiff queens don't enter into this calculation.

In theory, the odds slightly favor the finesse, at the time you must choose, on the grounds that you know lho has at has two spade spots and rho has one spade spot, making a little more room for the queen. Practically anything can cause you to reject the finesse. However, the non-trump lerad doesn't do much for me here. On this auction I, as defender, am reluctant to lead a trump from xx since it often helps declarer play the trump suit for no losers.

It depends on a lot of things but I probably, in most cases, just bang down A then K.

[P_Marlowe]

kenberg, on Dec 10 2005, 04:50 PM, said:

Everyone starts with ace and a small spade, there is only a decision to be made if everyone follows small. Stiff queens don't enter into this calculation.

Hi,

this maybe true, but deciding to play
the Ace or King is already a decision,
which opponent to play for 3 cards.

Marlowe

[kenberg]

P_Marlowe, on Dec 10 2005, 05:04 PM, said:

kenberg, on Dec 10 2005, 04:50 PM, said:

Everyone starts with ace and a small spade, there is only a decision to be made if everyone follows small. Stiff queens don't enter into this calculation.

Hi,

this maybe true, but deciding to play
the Ace or King is already a decision,
which opponent to play for 3 cards.

Marlowe

The given hand has Axxx opposite KJxxx. It's true that with A9xx opposite KJxxx you can guarantee no more than one loser by starting with the king but I did not take that as the point of the question. With Axxx opposite KJxxx and an intention to pick up the suit, you start with the ace, lead an x and if you have seen three spots you choose. IF (very big if) you ignore all other clues, there are 12 unknown cards to your right, 11 to your left, and 12/23 is .521739. Hence the 52-48 odds you see in books. Gambling casinos make a great deal of money repeatedly getting folks to take the short end of close odds.


The poster asks if I would decline the finesse because of a non-trump lead. I would not. Perhaps I should, but no I wouldn't. Not with the lead of an unsupported ace either.

[DelfinoD]

kenberg, on Dec 10 2005, 04:50 PM, said:

In theory, the odds slightly favor the finesse, at the time you must choose, on the grounds that you know lho has at has two spade spots and rho has one spade spot, making a little more room for the queen.

That's not true. Odds ar slightly against finesse.

The time when you choose doesn't make any difference, because you didn't get any usefull information.