Suit combination

[vuroth]

[hv=n=sj543hdc&s=sak7hdc]133|200|Scoring: MP

No outside entries to dummy[/hv]

 

While I misplayed this at the table (assuming, that is, my opponents were counting, too), I wondered if there was a preferred/percentage way to approach this suit, in isolation. Should I try low to the board early, before opponents can get count, or should I just slap the ace king down and try to drop the queen?

 

I'm guessing that, with outside entries, I can play AK, and if both follow and the Q does not drop I can try for 3/3.

[Simplicity]

If you lay down the AK you'll make 3 tricks when there is:

 

stiff Q either side - 2 combinations

Qx either side - 10

 

If you play small to the jack first up you've a chance of scoring 3 tricks when LHO has:

 

Qxx - 10 combos

Qxxx - 10

Qxxxx - 5

Qxxxxx - 1

 

Of course the likelyhood of LHO selecting the Q from these holdings is rather dependent on things like: what he knows about your hand, the rest of the deal, how good they are etc.

[vuroth]

Simplicity - you're using Pascal's triangle?

[TimG]

If you lay down the AK you'll make 3 tricks when there is:

 

stiff Q either side - 2 combinations

Qx either side - 10

You also make three tricks when the suit is 3-3, or when the Qxxx, Qxxxx or Qxxxxx is onside (because after the AK don't fetch the queen, you'll still be leading towards the Jack).

[TylerE]

Why not cash AK then lead towards the jack? Then you pickup stiff Q or Qx either side, any time Q onside, and any time the suit is 3/3.

[vuroth]

You also make three tricks when the suit is 3-3...

Nope. In this case, I didn't have an entry to cash the 4th spade if they split 3/3.

 

I *think* AK is clearly the line if I do have an entry, because I essentially get a second chance.

[Simplicity]

If you lay down the AK you'll make 3 tricks when there is:

 

stiff Q either side - 2 combinations

Qx either side - 10

You also make three tricks when the suit is 3-3, or when the Qxxx, Qxxxx or Qxxxxx is onside (because after the AK don't fetch the queen, you'll still be leading towards the Jack).

I believe the poster is well aware that with entries to dummy AK and another spade is clearly correct, he's asking about the merit of playing a low one up when no outside entry is available...

[Finch]

You also make three tricks when the suit is 3-3...

Nope. In this case, I didn't have an entry to cash the 4th spade if they split 3/3.

 

I *think* AK is clearly the line if I do have an entry, because I essentially get a second chance.

You are quite right that if you have another entry, the best line is A, then K, then lead towards the jack which makes three tricks whenever RHO does not have the Queen to four or more.

 

When there is no outside entry, your only genuine chance is that the queen is singleton or doubleton. As you say, you have two ways to play the suit:

i) Hope the queen is singleton or doubleton

ii) Lead a low one and hope LHO has the queen and does not play it.

 

I can't tell you which is better, or right. It's more likely (everything else being equal) that LHO has Qxx or Qxxx (or Qxxxx) than that either player has Qx or singleton Q. But it depends on the full hand and on your LHO how likely it is that LHO will play the queen on the first round.

[Simplicity]

Simplicity - you're using Pascal's triangle?

Esssentially yes, obviously this method does not take into account the distributional bias, but it does allow for very quick at the table calculations when comparing different lines.

[kenrexford]

I don't understand this at all. The ONLY percentage action, given the parameters, is to cash the Ace and King and then see if the Queen drops.

 

If the question concerns the possibility of sneaking the Jack past Qx(x)(x) in front of the Jack, I have no idea what the odds of a duck might be. The longer LHO's suit, perhaps the better the odds. Adding in a troubling card like the 9 or 10 also probably increases the odds, as would LHO's knowledge of what is going on.

 

I also know from seeing it that there is a chance, small though it may be, that RHO might also duck the Jack. This seems impossible, except that I just saw it happen. So, if it ever happens, there is a remote chance of it happening, per force.

 

All of this said, context usually is quite important. The odds, for example, of a drop are increased late in the hand because of pseudo-squeeze possibilities. Throw-in plays may solve the problem, as well. Heck, small to the Jack might also yield another trick in the remote and strange possibility of a stepping-stone developing, such as where one person has Q or Qx and the other later is endplayed into the entry. Could happen.

[TimG]

You also make three tricks when the suit is 3-3...

Nope. In this case, I didn't have an entry to cash the 4th spade if they split 3/3.

 

I *think* AK is clearly the line if I do have an entry, because I essentially get a second chance.

Sorry, I was focused on this part:

I'm guessing that, with outside entries, I can play AK, and if both follow and the Q does not drop I can try for 3/3.

My point being that after AK and no Queen, you are not dependent upon a 3/3 split; you still can take a 3rd trick when Qxxx is onside, for instance.

 

I don't have any great feel for how often LHO will duck if you lead up immediately i the no entry situation.

[Walddk]

I can't tell you which is better, or right.

Playing for the drop is a legitimate chance, hoping that LHO follows small from Qxx or longer if you lead a low card is a psychological chance. How can you ever compare the two? It's like asking how far there is from Budapest on horseback compared to the distance between Budapest and Vienna. It makes no sense.

 

Sorry if this sounds rude, but cashing AK is the only percentage action. You can't put percentages on psychology.

 

Roland

[vuroth]

If the question concerns the possibility of sneaking the Jack past Qx(x)(x) in front of the Jack

Yup, that's what it was.

 

Ok, it sounds like my uncertainty at the table was far from unique. Nobody had disarded, so everything was flat so far. RHO had shown up with enough points that I was confident LHO had the queen.

 

Of course, if LHO had been counting, he could have been certain that his partner didn't have the AK, so dropping was probably the right play at the table.

 

Fortunately, our bidding had already guaranteed a 50%+, so giving up an overtrick wasn't overly painful.

 

Thanks, guys.

 

V

[vuroth]

Sorry if this sounds rude, but cashing AK is the only percentage action. You can't put percentages on psychology.

I'm a bit confused. On defense, isn't putting the Q on the 7 when dummy is J543 sometimes correct bridge play, and sometimes incorrect bridge play? Or are you saying that it all comes down to what LHO reads as the card layout on the hand?

[kenrexford]

Sorry if this sounds rude, but cashing AK is the only percentage action. You can't put percentages on psychology.

 

Roland

I'm not sure that this is entirely accurate, even if I agree on the sentiment. Psychological ploys, however, are more difficult to establish as far as percentage likelihood. Rather than simply running mathematics, you would have to analyze actual occurrences, which seems rather difficult.

 

The idea, however, has some merit in exploring. I wonder if there is a way to run a program to analyze data from prior events to check on this sort of thing. Suppose, for instance, that you programmed a program to check for instances where a specific known psychological ploy was used, such as "fake splitting" KQ when Declarer leads up to AJ9. You might then be able to search the recap data on BBO for instances where dummy has AJ9 and LHO K10x+ or Q10x+ in the suit. You could then limit the database sample to those hands where LHO played the King or Queen at trick one in the suit, after Declarer played small toward dummy. You would then run data as to what Declarer played next in the suit when LHO plays small to the second trick. From that data source, you would get "percentages" as to how effective the fake-split maneuver is. You could even compare the win-loss from that maneuver with the percentage of time when Declarer does not play the 9 after the normal small from LHO. Something like that.

 

From this sort of case-specific analysis, an unreliable but perhaps somewhat persuasive percentage approximation table might be generated for certain high-frequency ruses. These could be updated periodically, and additional instances could be added over time. Parameters could be added to provide better data, such as analyzing some sort of data limitation as a function of the apparent talent level of the opposition.

[TimG]

Sorry if this sounds rude, but cashing AK is the only percentage action. You can't put percentages on psychology.

I'm a bit confused. On defense, isn't putting the Q on the 7 when dummy is J543 sometimes correct bridge play, and sometimes incorrect bridge play? Or are you saying that it all comes down to what LHO reads as the card layout on the hand?

Seems to me that hopping Queen would be a bad play if the suit were laid out like this:[hv=n=sj543&w=sq98&e=skt&s=sa762]399|300|[/hv]There will be clues from the bidding and play up to the point the suit is broken, but I don't think anyone would claim that LHO will always get this right in both cases.