[CSGibson]

...

[Vampyr]

 CSGibson, on 2012-February-17, 02:05, said:

It depends. If you think that you would be right 9/10 times, going for the overtricks has positive effect of .5 imps on this hand over the long run

Yes, it is a common misconception among inexperienced players that you should "never" risk your contract in a teams game. Disillusionment sets in when you have lost a few matches by an IMP...

[aguahombre]

 Vampyr, on 2012-February-17, 03:52, said:

Yes, it is a common misconception among inexperienced players that you should "never" risk your contract in a teams game. Disillusionment sets in when you have lost a few matches by an IMP...

It is common among inexperienced participants or observers of one-IMP matches to focus on a board where a small IMP difference occurred. While those boards might represent the margin, they don't mean risking one's contract for an unsure overtrick is winning strategy.

We might look to sloppy extra undertricks on other hands, or sloppy failure to go for an extra overtrick when a contract was assured; or failure to achieve minus 100 vs minus 140 on a conpetetive deal; or playing in a minor slam when we could have counted 12 sure tricks in Notrump with slightly better bidding.

The "long run" percentage plays are much more appropriate to matchpoints than to team matches...regardless of the length of the match.

I have even seen uncorrected false claims on Vugraph resulting in minus 12 IMPS when it should have been minus 13.

[HighLow21]

 CSGibson, on 2012-February-17, 02:05, said:

<snip>
The point of the article wasn't really to make that determination - it was to open people's eyes that criticizing opponents "mistakes" might really just show how much better your opponent is than you (not you, tate, but you the ubiquitous critic), because they have seen something that makes their line stand out.

Very good point and definitely taken. I tried to take this into account in my analysis and only credited others with a mistake when it was obvious a mistake had been committed --> for example, I *might* give them an error of 1 IMP had they made 3N with no overs, since the line I described earlier is foolproof for 10; I might not though. I would definitely credit an error if the contract failed.

And absolutely the contract could go down on her line of play. If Lefty had K♣ ATx♦; or K♣ J♥ Ax♦; or even Kx♣ J♥ x♦, remaining at the start of trick 10. Declarer had Q♦ AJT♣ remaining and dummy was worthless. The defense takes K♣, and either A♦ and 2 more diamonds, or K♣, J♥, and Ax♦. Down 1.

[HighLow21]

 Vampyr, on 2012-February-17, 03:52, said:

Yes, it is a common misconception among inexperienced players that you should "never" risk your contract in a teams game. Disillusionment sets in when you have lost a few matches by an IMP...

Please read my prior post, above. The odds of losing the contract are definitely higher, when weighting for the cost of failure relative to the value of gain, than the value of taking the K♣ finesse. I can even calculate the exact odds if you want me to.

[HighLow21]

 aguahombre, on 2012-February-17, 06:54, said:

It is common among inexperienced participants or observers of one-IMP matches to focus on a board where a small IMP difference occurred. While those boards might represent the margin, they don't mean risking one's contract for an unsure overtrick is winning strategy.

We might look to sloppy extra undertricks on other hands, or sloppy failure to go for an extra overtrick when a contract was assured; or failure to achieve minus 100 vs minus 140 on a conpetetive deal; or playing in a minor slam when we could have counted 12 sure tricks in Notrump with slightly better bidding.

The "long run" percentage plays are much more appropriate to matchpoints than to team matches...regardless of the length of the match.

I have even seen uncorrected false claims on Vugraph resulting in minus 12 IMPS when it should have been minus 13.

This is another point I was going to make just above but didn't; thank you for making it for me.

If your team loses by 1 IMP because you took only 10 tricks in this 3N contract, and you focus blame on the results of this hand, you are looking in the wrong place.

MUCH more likely is that you will lose a match you should have won because you tried for an overtrick on this hand and actually blew the contract.

[CSGibson]

 HighLow21, on 2012-February-17, 11:44, said:

Very good point and definitely taken. I tried to take this into account in my analysis and only credited others with a mistake when it was obvious a mistake had been committed --> for example, I *might* give them an error of 1 IMP had they made 3N with no overs, since the line I described earlier is foolproof for 10; I might not though. I would definitely credit an error if the contract failed.

And absolutely the contract could go down on her line of play. If Lefty had K♣ ATx♦; or K♣ J♥ Ax♦; or even Kx♣ J♥ x♦, remaining at the start of trick 10. Declarer had Q♦ AJT♣ remaining and dummy was worthless. The defense takes K♣, and either A♦ and 2 more diamonds, or K♣, J♥, and Ax♦. Down 1.

Tate, we're talking about different things (I hope). I'm saying that if declarer can always choose to play it safe for 9 tricks even after the spade at trick 3, LHO showing out, and that she has never put her contract in danger by doing so. The defense doesn't have the tempo to do anything about it. Declarer simply plays the 9 of spades, giving up a trick, and if opponents return a diamond, she has the tempo to drive out the K of clubs, and if they return a club, she has the tempo to drive out the ace of diamonds. - entries may make 10 tricks impossible to guarentee at that point, but 9 is always on the table with a good chance for 10 or 11.

Later in the play, I think its possible for declarer to deduce my hand from my chosen defense, and that she might have felt that deduction to be a high enough probability that it was worth risking the contract to go for the 2 imps.

Now, as said before, I'm not 100% sure that she did actually make those deductions instead of making a late-night mistake, but I do think it possible.

[HighLow21]

 CSGibson, on 2012-February-17, 12:11, said:

Tate, we're talking about different things (I hope).

<snip>

Yes, sorry I thought that was clear (it was in my mind, but sometimes I don't communicate well, and in any hand analysis there can be multiple points so it can definitely get confusing). We are indeed talking about 3 different things actually.

1. Her line of play from the beginning was slightly inferior. Yes, even after 2 rounds of spades the contract cannot be defeated (even if spades are 1-4 with RHO holding QTxx♠) IF the declarer does the right things after the 2 round of spades. Win the non-heart return, knock out the minor suit stoppers, and take 5♣/♠ + 3♥ + 2♦. 10 tricks. However, I think it's slightly inferior to play on spades first because the chances of making 4 spades are only 34%, and the chances of making only 2 are about 11%. And in that 11% scenario, you are exposing losers quickly. By contrast, clubs offer a guaranteed minimum of 3 tricks and about a 30% chance of taking 4. (Entries factor into this. Spades are her only entry for the hearts, which should come later, and the club finesse, which requires re-entry to maximize odds for 4 tricks. This tilts the best play towards spades first somewhat, so it's not THAT inferior, but technically, minors first is a better line in my view.)

2. Again, the contract cannot be defeated on best defense as you said. RHO cannot lead a major, and LHO cannot lead the 3rd round of diamonds before South's hand has been squeezed by the heart winners in dummy. I agree fully on this point. I agree also that there is a slight inference that you might hold the 2 relevant remaining cards: A♦ and K♣. This makes the play for 11 assured if she is right about the K♣: win the passive spade return in dummy, and finesse clubs by leading Q♣. When the K♣ is played she is back in hand, and that is the time to establish 2 diamond tricks. Because spades were 3-2, the long spade is an entry for 2 hearts. Making 11 whenever you have K♣: 3♠, 3♥, 2♦, 3♣. 11 guaranteed whenever K♣ is on her right.

3. Her ACTUAL continuation after the passive spade return did not do this. It did make 11 tricks. What it did not do is guarantee the contract, and what it did not do is produce any more tricks than the line I just described to you in (2).

So really, her LOP after the spade return gained nothing and risked everything. This is assuming my analysis is correct, and everyone knows I might have missed something. But I cannot think of a layout that fails to produce 11 on the line I just described, on the condition that RHO holds K♣ with any possible length.

By switching to ♦ after the passive spade return, then cashing ♠ and ♥, and finally finessing in ♣, she could have gone down in hand worth a dead minimum of 10 tricks.

Again, Chris, I think you did a very good job in analyzing this hand, and you drew a conclusion I would have never thought of: she might have made the inference that you held the 2 important remaining cards. Thus, you can miscredit 'mistakes' or 'good plays' if you're not certain that you can understand the line of reasoning behind the line of play. That is an excellent point and I absolutely appreciate being exposed to it!

My only thought in the post-mortem would have been this: why did she take that line of play when she will be set more than 0% of the time, and when it gains nothing over the guaranteed line of play I just described to you?