BBO Discussion Forums: How do you play the heart suit? - BBO Discussion Forums

North

♠ Q 8 7 5
♥ A K J 10 8
♦ A
♣ A 8 5

South

♠ K 10 9 6 4 2
♥ 5
♦ 5 2
♣ 9 7 4 3

In an uncontested auction, N/S reach 6♠.

West leads ♦K.

Assuming you play one round of trumps, West plays ♠J and East plays ♠3.

Now, how do you handle the ♥ suit?

D.

AK and finesse seems to pick every doubleton queen and doubleton 9 on left, so looks better. You can pick only half of the 4-3 anyway.

Fluffy, on 2015-August-21, 16:25, said:

This is my thought as well. Of course as well all see we need to be able to pitch 3♣ and have only 1♥ to take the finesse the other way. Therefore, I also try the ruffing finesse.

I guess 2S(!)-6S auction? Well, whatever, now you need to get rid of 3 C losers so 3 discards to find.

Option 1 : finesse J to discard 2 on the AK plus 1 on the 5th H if the Q didn't fall (and W ) - I dont have the maths but probably a little short of 50%

Option 2 : A, K and ruf, wins when the Q is 3rd at max and I cant help thinking it is less likely.

At the table after surviving trick 2, I would just take a deep breath, mutter a silent prayer to the HQ and finesse.

Or did I miss sth?

apollo1201, on 2015-August-21, 16:34, said:

Yes, you missed playing the AK and advancing the Jack. This wins, compared to low to the Jack on the first round, when RHO holds the Queen, unless he holds Q9xxx(x), when it breaks even. Essentially AK wins just a few more times than does low to the Jack. I am not going to try to estimate the percentage. It will be low. This is the perfect swing hand if playing a match where one is either stuck a lot or outgunned. In that case, if you assume the opps would bid this horrible slam (which is a real stretch) then if you need a swing, play low to the Jack. If you want the technically correct play, cash the AK and advance the J. Actually, since this is such a terrible contract, I would assume the other team will miss it or get a club lead, and i would make the technical play.

First find out of they play multi, they might be playing 6♠ from north on a club lead.

mikeh, on 2015-August-22, 00:11, said:

Haha I missed that answering expert questions late at night and jetlagged is an idea at best as good as this slam

Thanks for the replies.

I see it like this;

Possible ♥ distributions are...

Q xxxxxx
Qx xxxxx
Qxx xxxx
Qxxx xxx
Qxxxx xx
Qxxxxx x
Qxxxxxx

with either East or West holding the above.

So, the finesse works 7/14.

A,K, followed by a ruff works 8/14 (Q, Qx, Qxx or Qxxx held by either East or West) though I'm open to correction on this. If you force out the A Spades and defence puts you in dummy with a diamond continuation, you still have A clubs and one trump entry in case of Qxxx by either player. Right?

But....

A,K, and running the J works 9/14, because of possibility of singleton or doubleton ♥Q with West, as well as the seven possibilities of queen with East.

Therefore, I think that A,K and running the J is the correct play, by the tiniest of margins. The finesse is, I think, the worst play.

Lastly, in order to assure your A and K of hearts as two winners, you must remove trump A first. If either is ruffed due to void or singleton, then there is nothing to talk about. Right?

D.

Ps.

For the record, this was a top class teams events. Both rooms were in ♠6.

In one room, South finessed and claimed. In the other, South, a mathematical wizard, played ♥A,K and ran the ♥J for minus one.

Your analysys has several flaws:

The most obvious is that you never make against most 5-2* or worse distribution since you either can't capture the queen, or the 8 ends up captured by the 9.

Your method of counting cases is wrong. Every card is different, cases with 'x' on each side are multiplied.

Q-976432 = 1 case
Q9-76432
Q7-96432
Q6-97432
Q4-97632
Q3-97642
Q2-97643 = 6 cases

There are 6 different ways a queen can be doubleton on west, but only 1 were it can be singleton on west. Saying that queen doubleton is 6 times more likely as queen singleton is a good aproximation, for cases where queen is short you could say there are 7 total cases, but for a full suit analysis there are going to be 128 total cases.

I personally don't ever go beyond counting total cases, but purists, and suitplay, and other programs will also factor that each 4-3 case is a tiny bit more likely than 5-2 cases which are more likely than 6-1. Wich makes even distribution even more likely.



*: you don't make against any queen on the left with 5+ cards, there is no way to trap it. Playing AK+ run the Jack you capture any queen doubleton, But If east has ♥Q9xxx you are down, You can only make 3 heart tricks and a club will be lost in the end. When west has ♥9x then you can make as ♥8 will beat the last pip from East.

Dinarius, on 2015-August-22, 06:04, said:

The first step in conducting a campaign is to figure out your objectives. In this particular hand we need to be able to pitch 3 clubs on the hearts so any hand where west holds
5 or more hearts to the Q is a loser and any hand where east holds 5 or more hearts to the Q9 is a loser. That means those distributions should be eliminated entirely from consideration before calculating how to proceed. Our target therefore is to aim for hands where hearts break 3/4 4/3 and a limited supply of 5/2 2/5 (6/0 0/6 forgetaboutit).

Next is how to count distributions. I note that you counted Q xxxxxx as 1 (which it is) and also counted Qx xxxxx as one which it is not. Let's replace al those x's with 234569.
Q2 34569
Q3 24569
Q4 23569
Q5 23469
Q6 23459
Q9 23456

Note that Qx is actually 6 different possible card combinations vs the 1 you attribute to it in your analysis. Qx occurs 6 times more often than Q solo. Now for Qxx
Q23
Q24
Q25
Q26
Q29
Q34
Q35
Q36
Q39
Q45
Q46
Q49
Q56
Q59
Q69

Note how there are 15 different Qxx combinations vs the 1 you attributed it in you original idea in fact you would have given Q or Qx a 2/1 advantage over Qxx yet with 15 combinations, vs only 7 total for Q or Qx, Qxx is more than a 2/1 favorite to occur.

Now for xxx

234
235
236
239
245
246
249
256
259
269
345
346
349
356
359
369
456
459
469
569

Note there are 20 combinations of xxx opposite Qxxx

By now I hope you get the idea that instead of simply counting the different suit lengths as equals you must try and envision all of the subtypes for each length to see which carries more weight before deciding how to proceed. This does not exhaust the analysis (you can take it from here) but the numbers get sort of large and we primarily look for shortcuts since we have only limited time available for calculation.

Don't be in a silly 6♠ contract.

problem solved.

Thanks for all the clarifications.

As I see it now, there are two requirements;

1. We must make the best attempt to find the queen - which I think is A,K and then run the J through East.

...and....

2. The 9 must be no worse than 9xxx

Correct?

D.

9 can be 5th as long as queen falls. You can simplify suit combinations to a single card.

The bidding is important. Who dealt? Did W pass at a low level? W presumably holds SAJ. The lead marks him with DKQ so he can't hold HQ and have passed as dealer, in which case you play E for the HQ.

fourdad, on 2015-August-22, 12:47, said:

Silly comment. Are you the only player on the planet who has never been in a bad contract on a hopeless auction? Assuming you do get into bad contracts that have a chance to make, do you just lay your cards on the table and tell the opponents to take all the tricks they are entitled to? For your peace of mind, assume that the contract is 4♠ and you are trying to make 2 overtricks.

GrahamJson, on 2015-August-23, 02:44, said:

Agree that the bidding may be important, as well as vulnerability and type of scoring. All should be included, even if the OP doesn't think they matter because they may make a difference to an expert.

West doesn't have ♠A. After winning diamond ace, presumably a low spade was led from dummy and declarer put up the king. If west had ♠AJ, he would have won the trick and exited instead of holding the ace so he could possibly be endplayed at a later trick.