[quiddity]

---

Contract: 4♥, no opp bidding.
♦J led to dummy's ace, East encourages.

Can someone help me analyze this hand and find the percentage line at matchpoints? My robot opponent played off the AQ of trumps at this point and I'm trying to see if this is really the best line.

[rhm]

This type of hand is hard to analyze because there are too many approaches and probably as many one is prone to overlook.
If you can never be sure whether you are even aware of all possible lines of play how can you know which line has the highest probability of success?
I can only compare line of plays I am aware of.

The contract looks normal though some could be in 6♥ (not a bad contract) and others in notrumps.
If both black kings are with East, and clubs and hearts break you can make 13 tricks and also when East has two hearts and Kx in clubs.
Chances for all that is less than 10%.

My understanding how robots play is that they sample some layouts of the East - West hands and check which lines win most tricks most often under double dummy defense.
Their accuracy depends on the number of deals they can analyze in the allotted time frame times the allotted processing power.
But this hand shows another deficiency, namely assuming double dummy defense.

Quick entries to North are scarce, considering that you want to ruff your third diamond and take finesses in clubs and possibly in spades.
For example one line, which looks attractive to me and I would probably adopt at the table is as follows:

After winning the diamond take an immediate club finesse. (low from dummy)
If this wins go back in diamonds and repeat the club finesse. (again low from dummy)

If it wins again or the king pops up with East draw 2 rounds of trumps before continuing clubs, except when the club king popped up and East has 4 trumps, in which case I would draw a third round of trumps before continuing clubs for 11 tricks.
When somebody ruffs in my trick potential varies between 11 and 13, depending on whether clubs break , who ruffs, whether trumps break and whether the spade finesse works.
For example if trumps and clubs break and West ruffs the fourth club and the spade finesse is on I make 13 tricks.

If the club finesse loses in theory West could duck or win and return a club. Surely a big consideration for the robot.
But I would bet that even in a strong field more than 90% of all West players would win and continue diamonds.
(Encouraging diamonds instead of giving count is not so clever in my opinion, but that is what most defenders play)

On a diamond continuation I can ruff a diamond, draw 2 rounds of trumps go back to the ♣Q and draw trumps ending with 12 tricks (or 11 tricks if hearts are 4-1) without needing the spade finesse.
Since the ♣K is wrong this should give you a maximum score for being in 4♥.

This line does not look attractive to a robot, because it will not assume that West would continue diamonds.
Of course this line could also run into a 5-1 club break.
Against that West is very unlikely to lead a diamond with a singleton club though he could have five. Most of the time the hand short in cluby will be long in hearts.

Rainer Herrmann

[wanoff]

quiddity, on 2014-July-09, 19:26, said:

---

Contract: 4♥, no opp bidding.
♦J led to dummy's ace, East encourages.

Can someone help me analyze this hand and find the percentage line at matchpoints? My robot opponent played off the AQ of trumps at this point and I'm trying to see if this is really the best line.

Put simply, there are 2 club finesses required and there are only 2 entries to dummy.
Ergo the first finesse has to happen at trick 2.

[Winstonm]

It looks to me that to make 13 tricks a lot of good things have to happen - 3-2 hearts or 4-1 hearts with clubs behaving, onside, with no J10xx of hearts anywhere.

I don't claim this as right or wrong but at the table I would set my sights on making 12 tricks. To do this requires ruffing a diamond and a favorable heart break. My plan is to test the hearts with the AQ, then if they are 3/2 go ahead and ruff the 3rd diamond then lead the J of clubs from dummy.

My risk is the club K offside with the long diamond and an uppercut. But if the diamonds are 5/3 that way, the club king is more likely onside. This line also abandons the chance of all 13 tricks - which may be wrongheaded on my part.

That's why I never liked matchpoints as much as other scoring forms - you had to think too much.

[quiddity]

Thanks all (and especially Rainer for the detailed analysis). My feelings about the hand were similar to Rainer's; take a black suit finesse and if it loses hopefully it will provide an extra entry to draw trump after ruffing a diamond. The robot is usually pretty good at this kind of problem though. I wonder if it picked this line (AQ of trumps) because it keeps more balls in the air. It "works" if either black king is off in that if the SK is off it can lead a spade to the Q later, and if the CK is off it can lead a club to the Q. The robot doesn't notice the decision point coming up because it is analyzing everything double-dummy.

[johnu]

quiddity, on 2014-July-10, 12:05, said:

Thanks all (and especially Rainer for the detailed analysis). My feelings about the hand were similar to Rainer's; take a black suit finesse and if it loses hopefully it will provide an extra entry to draw trump after ruffing a diamond. The robot is usually pretty good at this kind of problem though. I wonder if it picked this line (AQ of trumps) because it keeps more balls in the air. It "works" if either black king is off in that if the SK is off it can lead a spade to the Q later, and if the CK is off it can lead a club to the Q. The robot doesn't notice the decision point coming up because it is analyzing everything double-dummy.

Playing 2 rounds of hearts right away is just a bad play. It puts you in the wrong hand for any black suit finesses, and if you don't take an immediate diamond finesse ruff, allows an opponent to play a 3rd round of trumps if you let them in before you ruff a diamond, so now you have a diamond loser to worry about.

[PhantomSac]

johnu, on 2014-July-10, 13:18, said:

Playing 2 rounds of hearts right away is just a bad play. It puts you in the wrong hand for any black suit finesses, and if you don't take an immediate diamond finesse ruff, allows an opponent to play a 3rd round of trumps if you let them in before you ruff a diamond, so now you have a diamond loser to worry about.

I mean, there's no question that 2 rounds of trumps is better against optimal defense than a club to the jack right? It's not like a club to the jack solves your problems, it just creates some more. Ofc I would play a club to the jack in real life and agree with rainer/quiddity.

[johnu]

PhantomSac, on 2014-July-10, 16:39, said:

I mean, there's no question that 2 rounds of trumps is better against optimal defense than a club to the jack right? It's not like a club to the jack solves your problems, it just creates some more. Ofc I would play a club to the jack in real life and agree with rainer/quiddity.

What optimal defense are you talking about? 2 immediate rounds of trump is more of a misguided analysis that GIB can declare double dummy more than it will get optimal defense.

[PhantomSac]

johnu, on 2014-July-11, 00:40, said:

What optimal defense are you talking about?

How are you planning to play if a club to the jack wins (hint: the CK may still be offside!). How are you planning to play if they win the CK and return a club?

[johnu]

PhantomSac, on 2014-July-11, 09:33, said:

How are you planning to play if a club to the jack wins (hint: the CK may still be offside!). How are you planning to play if they win the CK and return a club?

How are you planning on playing once you have drawn 2 rounds of trump and are in your hand?

[PhantomSac]

johnu, on 2014-July-11, 09:52, said:

How are you planning on playing once you have drawn 2 rounds of trump and are in your hand?

Lol that is a nice troll. You keep saying how awful playing 2 trumps is compared to playing a club to the jack and that GIB doesn't understand bridge, yet you have no plan for what to do when you win a finesse (or lose a finesse and they return a club). You have not accomplished anything unless they win the CK and return a non club, or if you can assume that the club king is onside when the club holds. Meanwhile you have opened yourself up to club ruffs. That is the point: GIB is trying to take good lines assuming double dummy defense, human beings don't do that. If you want to argue that a club to the jack is even correct against double dummy defense then why don't you tell us what your plan is?

[johnu]

PhantomSac, on 2014-July-11, 10:06, said:

Lol that is a nice troll. You keep saying how awful playing 2 trumps is compared to playing a club to the jack and that GIB doesn't understand bridge, yet you have no plan for what to do when you win a finesse (or lose a finesse and they return a club). You have not accomplished anything unless they win the CK and return a non club, or if you can assume that the club king is onside when the club holds. Meanwhile you have opened yourself up to club ruffs. That is the point: GIB is trying to take good lines assuming double dummy defense, human beings don't do that. If you want to argue that a club to the jack is even correct against double dummy defense then why don't you tell us what your plan is?

LOL, calling my post a troll is raising the standard of trolling. If you very carefully read my previous posts, you'll see that I didn't say what I thought about an immediate club finesse. I also didn't say that GIB doesn't understand bridge, I said that 2 rounds of trump was a bad line of play. Everybody takes a bad line of play sometimes and that doesn't mean they don't understand bridge.

If you don't want to be a troll, answer my question. What is your line of play after you play 2 rounds of trump first? GIB has taken a line of play where it assumes it knows what the opponents have. The only problem is that once it gets to a further decision point, it has to play single dummy. If you think that is a good line of reasoning and you know what the opponents have, more power to you.

If I knew I was playing against double dummy defense, I would probably lead a spade to the jack at trick 2.

[PhantomSac]

Johnu, you seem to have forgotten how this interaction went. I said:

Quote

I mean, there's no question that 2 rounds of trumps is better against optimal defense than a club to the jack right? It's not like a club to the jack solves your problems, it just creates some more. Ofc I would play a club to the jack in real life and agree with rainer/quiddity.

You replied:

Quote

What optimal defense are you talking about? 2 immediate rounds of trump is more of a misguided analysis that GIB can declare double dummy more than it will get optimal defense.

I replied:

Quote

How are you planning to play if a club to the jack wins (hint: the CK may still be offside!). How are you planning to play if they win the CK and return a club?

So, I claimed that 2 rounds of trumps is better against optimal defense than a club to the jack. You asked what optimal defense. I illustrated that to you by asking you how you will play the hand after a club to the jack wins, or loses and a club comes back. You still have not answered how you are planning to play against optimal defense. You are now simply saying you would not play a club to the jack. But let's remember the first statement I made that you responded to that started this interaction, in case you forgot already it was:

Quote

there's no question that 2 rounds of trumps is better against optimal defense than a club to the jack right? It's not like a club to the jack solves your problems, it just creates some more.

You can keep trolling me and calling me a troll and avoiding what we were actually talking about, I don't really mind. I stand by my statement that 2 rounds of trumps is better against optimal defense than a club to the jack, and that GIB assumes optimal defense (to its detriment!). You have failed to state any line that shows superiority over 2 rounds of trumps after a club to the jack, in fact you have just continually avoided it so I am not sure on what basis you are disputing that a club to the jack is worse than 2 rounds of trumps on optimal defense other than OMG TWO ROUNDS OF TRUMPS IS BAD.

[Fluffy]

I was missing some action on BBF. I miss some comments now by Hannie or jdown

[Winstonm]

Now I'm curious. My line of play comes to 12 tricks unless the clubs are 5/1 or 6/0. What kind of board is +680?

[rhm]

Winstonm, on 2014-July-11, 15:44, said:

Now I'm curious. My line of play comes to 12 tricks unless the clubs are 5/1 or 6/0. What kind of board is +680?

That is not true.
12 tricks require much more, not least that trumps break.
You also risk a simple trump promotion in diamonds, say when West has the ♣K and 5 diamonds and East the last trump.

But better still when the club jack is played the defense could duck.
It is fairly easy to see for the defense that declarer is stuck in hand and wants to get to dummy to draw the last trump.
This is good defense, but much easier than ducking an immediate club finesse or returning a club instead of a diamond on an immediate club finesse.
How do you continue?
Presumably you now switch to spades.
East wins and a club is returned putting you back in hand. (You won't risk the club finesse now, would you?)
Getting to dummy, no matter whether you try to ruff the third club or the fourth spade (after discarding your club loser on the third spade), risks a trump promotion if East has the third trump not to mention that spades could be 5-2.

I am not claiming this line is terrible, only that 12 tricks are not assured against reasonable defense and it looks to me that 12 tricks are much more likely single dummy on the line I suggested irrespective of where the club king is.

But it confirms that there are numerous ways of playing the hand.
For example one could start with the spade finesse and simply rely on that one of two black finesses will work, in which case you are likely to come to 12 tricks if trumps break.
Starting with spades has the advantages of establishing a way back to dummy.
Starting with the club finesse has the advantage that on many variations you will not need the spade finesse, in particular if the club finesse loses and the defense is not brilliant, which I never assume. (Catering to good defense is something different and prudent.)

Rainer Herrmann

[Winstonm]

rhm, on 2014-July-12, 03:54, said:

That is not true.
12 tricks require much more, not least that trumps break.
You also risk a simple trump promotion in diamonds, say when West has the ♣K and 5 diamonds and East the last trump.

But better still when the club jack is played the defense could duck.
It is fairly easy to see for the defense that declarer is stuck in hand and wants to get to dummy to draw the last trump.
This is good defense, but much easier than ducking an immediate club finesse or returning a club instead of a diamond on an immediate club finesse.
How do you continue?
Presumably you now switch to spades.
East wins and a club is returned putting you back in hand. (You won't risk the club finesse now, would you?)
Getting to dummy, no matter whether you try to ruff the third club or the fourth spade (after discarding your club loser on the third spade), risks a trump promotion if East has the third trump not to mention that spades could be 5-2.

I am not claiming this line is terrible, only that 12 tricks are not assured against reasonable defense and it looks to me that 12 tricks are much more likely single dummy on the line I suggested irrespective of where the club king is.

But it confirms that there are numerous ways of playing the hand.
For example one could start with the spade finesse and simply rely on that one of two black finesses will work, in which case you are likely to come to 12 tricks if trumps break.
Starting with spades has the advantages of establishing a way back to dummy.
Starting with the club finesse has the advantage that on many variations you will not need the spade finesse, in particular if the club finesse loses and the defense is not brilliant, which I never assume. (Catering to good defense is something different and prudent.)

Rainer Herrmann

I know that - no line is foolproof. And if you read the post, you will see I noted the uppercut risk myself. But thatnks for pointing it out again.

[awm]

Hmm I think I would play the spade queen at trick two. I figure one of two things will happen:

1. The spade queen loses to LHO's king. Opponents probably continue diamonds. But now I have a spade established to discard dummy's diamond loser, so I can continue with the top two trumps, then cash two spades. Of course if spades are 5-2 and trumps are 3-2 and the long trump is with the short spades I may lose a trick to a ruff this way.. but it still seems better than the potentially risky club finesse line (which really accomplishes little if LHO is good enough to duck with the king) or GIB's line.

2. The spade queen is covered by RHO's king and I win in hand. Then I've got a spade ruff for a potential dummy re-entry after cashing two hearts, diamond king and ruff a diamond.

I don't think GIB is good at finding lines like this, because leading a low spade towards the queen later in the hand is potentially a good play if spade king is with LHO. Of course, there is no obvious way to figure out that spade king is with LHO, but GIB tends to assume double-dummy play after the current trick, and thus wants to delay this decision as long as possible.

[rhm]

awm, on 2014-July-12, 16:26, said:

.. but it still seems better than the potentially risky club finesse line (which really accomplishes little if LHO is good enough to duck with the king)

That's the whole point.
How often do you expect even in a good field that LHO will duck or return a club? (By the way this was apparently BBO)

I never assume my right or left hand opponent to be Belladonna or Garozzo. If they were they would probably not play with me. And I get rarely enlightened or should I say disappointed.
To be good in this game does not mean you should play double dummy, but it does mean you should look at the problem from your partner's and opponents perspective.
That means as declarer you have to play 3 dummies not one.
At times I will reject a technically superior line, if the inferior line will give my opponent a chance to go wrong.
I admit I also find it satisfying when my opponents do what I want them to do.
There is not so much pleasure playing with robots.

Rainer Herrmann