[Phil]

Dummy: AK8754

You: T6

Entries are not a big deal. Best chance for five tricks.

[Hanoi5]

Letting the ten run? If it is covered winning and another towards the 8.

[HighLow21]

Lemme see how accurate this is:

• 5 tricks are child's play if they divide 3-2. This is 68%.
  
• Cashing AK loses to any singleton on either side, however. Either Q9, J9, or QJ will remain for 2 defensive tricks.
  
• In theory, a double finesse against QJ is better than the drop, but it fails to pick up some QJ combinations with West, and obviously loses to QJ or QJx with East.
  
• If East has QJxx there is no chance.
  
• If West has QJ9x(x), leading the T, covered by W/N, then finessing the 8, works. But this is less than just playing for the drop. West has QJ9x(x) only 2% + 14% = 16% of the time.
  
• If East has the singleton 9 (3%), leading the T intending to run it forces a cover from West. But it fails to pick up the single Q or J with East. This is 6%.

So I would cash the Ace then run the T as a safety play. This makes 5 whenever the suit is split 3-2 OR East has the singleton Q or J.

6% + 68% = 74%.

[HighLow21]

Also I think if the suit were AK876 opposite T54, A then low to the T is best.

[masse24]

Hanoi5, on 2012-April-03, 21:34, said:

Letting run the ten? If it is covered winning and another towards the 8.

Yup. This one.

[nigel_k]

If we run the ten, we gain when East has a small singleton. If we cash a top honour then run the ten, we gain when East has singleton Q or J. These are equally likely. Both lines also work if East has singleton 9.

[Fluffy]

Ace and run the 10 is what I'd do, inmeadiate fineses leave you with impossible guesses next round.

[gnasher]

Fluffy, on 2012-April-04, 02:30, said:

Ace and run the 10 is what I'd do, inmeadiate fineses leave you with impossible guesses next round.

If you start by running the 10, you don't have any guess - if it loses to an honour, you have to play the suit from the top.

As Nigel says, that makes the two lines equally good in theory. The auction might be relevant. If dummy hadn't shown this suit, LHO might have led one from QJ9x, so we'd prefer to cash the ace.

[gszes]

LHO holding QJ9x will be a lot more unhappy not scoring
2 tricks then they will be happy when they score 2 tricks
holding Q9xx or J9xx so run the TEN:))))))))))))))))))

There is also a chance a tired or bored LHO
holding Q9xx or J9xx will cover the Ten.

[Phil]

gnasher, on 2012-April-04, 04:22, said:

If you start by running the 10, you don't have any guess - if it loses to an honour, you have to play the suit from the top.

As Nigel says, that makes the two lines equally good in theory. The auction might be relevant. If dummy hadn't shown this suit, LHO might have led one from QJ9x, so we'd prefer to cash the ace.

Dummy had bid diamonds twice, so I'd ignore this inference.

[kayin801]

Suppose that declarer opts to lead the 10 first (regardless of whether it's right or wrong):

In line with the last thread like this, should LHO play the 9 from H9xx as well as from 9x to protect QJx offside? Of course the defense can always take 2 tricks when LHO has H9xx and playing the 9 would cost one of those tricks double dummy but in practice will declarer ever hook again when the 9 is dropped on the first round?

Or is it something along those lines? I think that last argument makes no sense because with the 9 comes up at round 1 that's then an indication that declarer should opt to go up instead? So do you have to play the 9 from QJ9x too then as LHO to protect yourself where the double finesse is ACTUALLY working?

Sorry commentary is disorganized but I'm too lazy to try to summarize and actually put it together.

Edit: I guess this all favors cashing the A and running the 10?

[gnasher]

kayin801, on 2012-April-04, 09:59, said:

Suppose that declarer opts to lead the 10 first (regardless of whether it's right or wrong):

In line with the last thread like this, should LHO play the 9 from H9xx as well as from 9x to protect QJx offside? Of course the defense can always take 2 tricks when LHO has H9xx and playing the 9 would cost one of those tricks double dummy but in practice will declarer ever hook again when the 9 is dropped on the first round?

Or is it something along those lines? I think that last argument makes no sense because with the 9 comes up at round 1 that's then an indication that declarer should opt to go up instead? So do you have to play the 9 from QJ9x too then as LHO to protect yourself where the double finesse is ACTUALLY working?

Sorry commentary is disorganized but I'm too lazy to try to summarize and actually put it together.

Edit: I guess this all favors cashing the A and running the 10?

It favours taking no notice of LHO's antics. If you started by leading the 10, your original plan was:
- Run the 10.
- If that loses to an honour, lead from the top.

When LHO plays the nine, you should stick with your plan.

[wank]

running the 10 is better as you get 3 winners for 0 losers. to score the 6th one you still need to lose 2 admittedly, but the extra flexibility may come in useful elsewhere in the hand.

[Phil]

Had a nice discussion in kibchat about this. The actual layout was stiff J on your right. Declarer mindlessly banged down the AK which gave him no chance.

I should have mentioned something, because its relevant. On the actual hand, the defense is poised to take three winners once you lose the lead. The entire hand looked like this:

---

The lead was the lowest outstanding spade to the J and K. What happened was declarer led low to the K and saw the Jack. If he returns to hand with a heart and runs the ♦T, LHO can duck. Whatever he chooses to return to dummy with, it sets up a 5th winner for the defense, so leading the T is a necessary unblock in any event.

As far as the static combo is concerned, you can never afford to double hook because you lose to silly holdings like J9 and QJ9, so ignore the 3-2's as highlow states. You also can't help 5-0's. The choices are:

1. T to the A, then low toward K8
2. Run the T, then K
3. Low to 8 and then A
4. Low to the A

Running the T loses to stiff Q/J. Low to 8 loses to stiff Q/J or 9. Ten to the Ace loses to the QJ9x(2) onside. Low to the Ace really doesn't seem to pick up anything the T to the A does, and screws up your entries in the process.

So it seems running the T and the T to the Ace are equivalent with two losing cases. However, LHO doesn't know that you have Tx! If you might have stiff T, RHO needs to cover some of the time to preserve his two diamond tricks with QJ9x and take three spades and two diamond tricks. Therefore, T to the Ace, and then lead toward the remaining K8 looks best.

As an aside, leading the T is also a farsighted play holding Txx for the same reason.

[HighLow21]

Phil, on 2012-April-04, 12:37, said:

...
and screws up your entries in the process.
...

You said originally that entries were not a concern... anyway, T to the A is the equivalent of low to the A then running the T if entries and timing are adequate, if I'm not mistaken...?

[johnu]

HighLow21, on 2012-April-04, 15:50, said:

You said originally that entries were not a concern... anyway, T to the A is the equivalent of low to the A then running the T if entries and timing are adequate, if I'm not mistaken...?

If the opponents play double dummy, they look to be the same because QJ9x will never cover the 10. For 10 to the ace in real life, you gain by whatever non-zero percentage of the time they cover the 10. They are unlikely to misplay if you lead low.