Jump to content

Great Diamond Hook


dustinst22

Recommended Posts

What's so great about it? Looks like just the normal line to try and make the contract to me.

 

"great" was my attempt at sarcasm, apparently it didn't come across that way.

 

Isnt it better to just ruff out a diamond? After E shows up with 2 spades, isn't it more likely he has 2-3 diamonds? I must be missing something, bc clearly you're better than me.

Link to comment
Share on other sites

"great" was my attempt at sarcasm, apparently it didn't come across that way.

 

Isnt it better to just ruff out a diamond? After E shows up with 2 spades, isn't it more likely he has 2-3 diamonds? I must be missing something, bc clearly you're better than me.

Ah, sorry, I miscounted the losers. Indeed this line:

[hv=sn=dustinst22&s=SAKQ7H85DA3C97542&wn=Robot&w=SJ43HKT94D42CAQJT&nn=Robot&n=ST982H76DKJ9865C8&en=Robot&e=S65HAQJ32DQT7CK63&d=n&v=e&b=9&a=P1H(Major%20suit%20opening%20--%205+%20H%3B%2011-21%20HCP%3B%2012-22%20total%20points)D(Takeout%20double%20--%203-5%20C%3B%203-5%20D%3B%202-%20H%3B%203-)2N!(Truscott%20%5BJordan%5D%20--%203+%20H%3B%2011+%20total%20points)3D(Free%20bid%3B%20new%20suit%20--%204+%20D%3B%206-10%20total%20points)PP3H(3+%20H%3B%2011-12%20total%20points)3S(4+%20D%3B%20biddable%20S%3B%209-10%20total%20points)P4S(3-5%20C%3B%203-5%20D%3B%202-%20H%3B%204%20S%3B%2012-18%20total%20poi)D(3+%20H%3B%2011+%20HCP%3B%2012%20total%20points)PPP&p=S6SAS4S2SKS3S9S5SQSJSTC3DAD4D5D7D3D2DKDTD6DQS7H4]480|360[/hv]

leads to a position with just 3 losers left, so you will make the contract. The problem is that GIB doesn't know that just making this contract will be a good score, it's trying to maximise its tricks. And at this point you are stuck in dummy with no way to reach your good diamonds, so you will lose those 3 tricks. The finesse gives you a chance to discard a heart before the defense cashes it. We can tell that this is not a worthwhile risk, but GIB can't (and it would be incredibly difficult to impart such understanding to it).

Link to comment
Share on other sites

I think it would be right to teach GIB that when he's doubled in a game contract, he should concentrate on making his contract.

 

Was going to say, seems pretty insane to be going for overtricks in a game that should be going down and doubled.

 

Shouldn't the algorithm allow the GIB to calculate whether the contract would have gone down on a different lead and make decisions based on this information?

Link to comment
Share on other sites

So, in the hand as originally posted, GIB must decide which diamond honor to play on Trick 5. If I'm using Pavlicek's Suit Break Calculator correctly:

  • There are two unplayed diamonds, Q and T.
  • West has 5 empty spaces, since we already know about 3 spades and 2 diamonds (played on the first 5 tricks) plus 3 hearts (bid).
  • East has 4 free spots, since we already know about 2 spades, 1 diamond and 1 club (played on the first four tricks) plus 5 hearts (bid).

There are equal (27.78%) chances that:

  • West has the Q and East has the T. Playing K would make contract; playing J makes overtrick.
  • West has the T and East has Q. Playing K makes overtrick; playing J goes down.
  • West has both Q and T. Playing K goes down; playing J makes contract.

Plus one possibility at 16.67%

  • East has both Q and T. Playing K makes contract; playing J goes down.

 

So, rearranging the possible results:

If GIB plays K: Goes down 27.78%, makes contract 44.44%, makes overtrick 27.78%

If GIB plays J: Goes down 44.44%, makes contract 27.78%, makes overtrick 27.78%

 

Gotta go up!

Link to comment
Share on other sites

So, in the hand as originally posted, GIB must decide which diamond honor to play on Trick 5. If I'm using Pavlicek's Suit Break Calculator correctly:

  • There are two unplayed diamonds, Q and T.
  • West has 5 empty spaces, since we already know about 3 spades and 2 diamonds (played on the first 5 tricks) plus 3 hearts (bid).
  • East has 4 free spots, since we already know about 2 spades, 1 diamond and 1 club (played on the first four tricks) plus 5 hearts (bid).

There are equal (27.78%) chances that:

  • West has the Q and East has the T. Playing K would make contract; playing J makes overtrick.
  • West has the T and East has Q. Playing K makes overtrick; playing J goes down.
  • West has both Q and T. Playing K goes down; playing J makes contract.

Plus one possibility at 16.67%

  • East has both Q and T. Playing K makes contract; playing J goes down.

 

So, rearranging the possible results:

If GIB plays K: Goes down 27.78%, makes contract 44.44%, makes overtrick 27.78%

If GIB plays J: Goes down 44.44%, makes contract 27.78%, makes overtrick 27.78%

 

Gotta go up!

It is good that you tried to factor in some of the information that is known about the East-West hands, but there is more to it than that:

 

1) The fact that each player's total points are known changes the odds. This makes the math quite involved.

2) If you are playing IMPs, you need to consider the IMP odds. For example, if you go up with the King and there was Queen fourth onside, you don't just "go down" - you go down 3 doubled tricks.

3) If you are playing Matchpoints, you need to consider things like what rates to happen at other tables.

 

Please note that I am not suggesting that GIB made the "right play". My point is that proper analysis of this situation is much more complicated than people seem to realize. In particular, the strategy of "always focus on making a doubled contract" is overly simplistic (and just plain wrong) both at IMPs and at matchpoints.

 

Assuming GIB made the "wrong play" (my sense is that it is fairly clear that it did), this hand is a good example of one of the flaws of the "generate some deals that fit what is known about the defenders' hands and then analyze double dummy" method of making such decisions. If the sample is small and/or biased, it is relatively easy for GIB to get "unlucky" in these situations.

 

The "basic" version of GIB is especially prone to making such errors because they are not given much time to "think" (and hence they frequently don't have enough time to generate and analyze enough deals to make for an effective sample).

 

Fred Gitelman

Bridge Base Inc.

www.bridgebase.com

Link to comment
Share on other sites

So, in the hand as originally posted, GIB must decide which diamond honor to play on Trick 5. If I'm using Pavlicek's Suit Break Calculator correctly:

  • There are two unplayed diamonds, Q and T.
  • West has 5 empty spaces, since we already know about 3 spades and 2 diamonds (played on the first 5 tricks) plus 3 hearts (bid).
  • East has 4 free spots, since we already know about 2 spades, 1 diamond and 1 club (played on the first four tricks) plus 5 hearts (bid).

There are equal (27.78%) chances that:

  • West has the Q and East has the T. Playing K would make contract; playing J makes overtrick.
  • West has the T and East has Q. Playing K makes overtrick; playing J goes down.
  • West has both Q and T. Playing K goes down; playing J makes contract.

Plus one possibility at 16.67%

  • East has both Q and T. Playing K makes contract; playing J goes down.

 

So, rearranging the possible results:

If GIB plays K: Goes down 27.78%, makes contract 44.44%, makes overtrick 27.78%

If GIB plays J: Goes down 44.44%, makes contract 27.78%, makes overtrick 27.78%

 

Gotta go up!

 

More importantly, need to consider if a different lead (i.e. cash hearts then tap in clubs) would have set the contract.....

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...