BBO Discussion Forums: Tricky slam line - BBO Discussion Forums

Jump to content

Page 1 of 1
  • You cannot start a new topic
  • You cannot reply to this topic

Tricky slam line How evaluate percentage difference?

#1 User is offline   Valardent 

  • PipPipPip
  • Group: Full Members
  • Posts: 91
  • Joined: 2008-July-11
  • Gender:Male
  • Location:Belgium

Posted 2011-August-30, 15:32

Hi, you play in 6 in South after a transfer sequence.

Lead is K, imps scoring.

IMPS




Obviously, if are 3-3, slam will always win.


I think these are the best 2 lines (+ one variation) :

a) Unblock AQ, A, back to hand in , and K

b) Unblock AQ, back to hand in , K, and finesse if K didn't survive.

Variation :

Follow a) or b) depending whether J falls on AQ or not.


Which one is best?

How calculate this 1) with enough accuracy at the table to choose the best one?
2) in the post mortem analyse to establish the percentage difference between the best 2 lines?




What about this way (focusing on the distributions)?

- If are 2-2, both lines are equivalent (50% chance)

- If are 3-1 or 1-3 :

Line a) wins when a defender holds x and xx or when K is stiff behind Ace

Line b) wins when a defender holds Kxx(x) and xxin front of dummy

(If J falls, line a) must be superior cos one wins also with Kxx and Jxin front of dummy)

- If are 4-0 or 0-4 :

Line b) wins when are in front of dummy with xx

Thoughts?
1

#2 User is offline   aguahombre 

  • PipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 12,029
  • Joined: 2009-February-21
  • Gender:Male
  • Location:St. George, UT

Posted 2011-August-30, 18:55

See post on using a calculator at the table to figure out the odds. Without it, and having to do something, I try the Ace of trump first.
"Bidding Spades to show spades can work well." (Kenberg)
0

#3 User is offline   nige1 

  • 5-level belongs to me
  • PipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 9,128
  • Joined: 2004-August-30
  • Gender:Male
  • Location:Glasgow Scotland
  • Interests:Poems Computers

Posted 2011-August-30, 20:40


Valardent asked for our thoughts on the play of 6 by South at IMPS.


IMO Cash A, AQ, Q, advance K.
- If both follow, discard a and run 9.
- If LHO ruffs, then over-ruff. Now...
- - If J hasn't appeared, then K, run 9.
- - If LHO had J doubleton, then: cash A, K, advance 9.

0

#4 User is offline   JLOGIC 

  • 2011 Poster of The Year winner
  • PipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 6,002
  • Joined: 2010-July-08
  • Gender:Male

Posted 2011-August-31, 02:00

Your analysis seems almost spot on Valardent. At the table I would think:

There are 3 cases of a spade and 2 small diamonds on my left.
There is 1 case of 3 spades and 2 small diamonds on my left.
There are 3 cases of Kxx spades and 2 small diamonds on my right.

There are 3 cases of 2 small spades on my left and 2 small diamonds.
There is 1 case of Kxxx of spades on your left with 2 small diamonds
There are 3 cases of Kx of spades on my right with 2 small dimonds

*** You seemed to forget 4-0 spades on your left with a doubleton diamond as a win for not cashing the SA

Calculating how many breaks is easy, for instance there are 3 possible combos of a stiff on your left, because there are 3 small stiffs LHO can have. There are 3 combos of Kx because K? the ? can only be one of the 3 small ones.


So, it is 7 cases to 7. Normally the edge would be to cashing the ace becuse it loses to a 4-0 break but picks up a 3-1 break, and a 3-1 combo is more likely than a 4-0. However, here it must be realized that LHO having short spades AND short diamonds is unlikely. Therefore, him having King third of spades and 2 small diamonds is far more likely than a stiff spade and 2 small diamonds.

So, keeping the ace in dummy is right imo. This is how I would consider it at the table, no fancy math is required and it wouldn't take you that long.
0

#5 User is offline   Valardent 

  • PipPipPip
  • Group: Full Members
  • Posts: 91
  • Joined: 2008-July-11
  • Gender:Male
  • Location:Belgium

Posted 2011-August-31, 11:50

View PostJLOGIC, on 2011-August-31, 02:00, said:

Your analysis seems almost spot on Valardent. At the table I would think:

There are 3 cases of a spade and 2 small diamonds on my left.
There is 1 case of 3 spades and 2 small diamonds on my left.
There are 3 cases of Kxx spades and 2 small diamonds on my right.

There are 3 cases of 2 small spades on my left and 2 small diamonds.
There is 1 case of Kxxx of spades on your left with 2 small diamonds
There are 3 cases of Kx of spades on my right with 2 small dimonds

*** You seemed to forget 4-0 spades on your left with a doubleton diamond as a win for not cashing the SA

Calculating how many breaks is easy, for instance there are 3 possible combos of a stiff on your left, because there are 3 small stiffs LHO can have. There are 3 combos of Kx because K? the ? can only be one of the 3 small ones.


So, it is 7 cases to 7. Normally the edge would be to cashing the ace becuse it loses to a 4-0 break but picks up a 3-1 break, and a 3-1 combo is more likely than a 4-0. However, here it must be realized that LHO having short spades AND short diamonds is unlikely. Therefore, him having King third of spades and 2 small diamonds is far more likely than a stiff spade and 2 small diamonds.

So, keeping the ace in dummy is right imo. This is how I would consider it at the table, no fancy math is required and it wouldn't take you that long.


To sum it up (not bothering when are 2-2):

Line "A" is better in those 4 cases : _____ Line "finesse" is better in those 4 cases :

x xx in West (3 cases) (1)_____________ Kxx xx in West (3 cases) (2)

xxx xx in West (1 case) (3)____________ Kxxx xx in West (1 case) (4)


So (4) is a bit less frequent than (3), but (2) is much more frequent than (1)

When J falls, playing "A" line wins also in (2), making "A" line best.

When J does not fall, "finesse" line is best.

Anyone out there who can explain how to calculate the odds difference between (1) & (2) and (3) & (4) ?
0

#6 User is offline   gnasher 

  • Andy Bowles
  • PipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 11,993
  • Joined: 2007-May-03
  • Gender:Male
  • Location:London, UK

Posted 2011-August-31, 14:24

View PostValardent, on 2011-August-31, 11:50, said:

Anyone out there who can explain how to calculate the odds difference between (1) & (2) and (3) & (4) ?


This is how you do it:

With diamonds 2=4, the vacant spaces are initially 11:9. Each time we deal a spade to one of the players, the vacant spaces for that player are reduced by one. Hence:

Chance of a specific 3=1 break: 11/20 * 10/19 * 9/18 * 9/17 = 0.077
Chance of a specific 1=3 break: 11/20 * 9/19 * 8/18 * 7/17 = 0.048
Chance of a 4=0 break: 11/20 * 10/19 * 9/18 * 8/17 = 0.068
Chance of a 0=4 break: 9/20 * 8/19 * 7/18 * 6/17 = 0.026

The divisors are only relevant if you're comparing chances in two different suits. In this case you could ignore them, but to do exact calculations at the table you'd still have to be very good at mental arithmetic.

The other way to do this type of calculation is to use Richard Pavlicek's calculator at http://www.rpbridge.net/xsb2.htm
... that would still not be conclusive proof, before someone wants to explain that to me as well as if I was a 5 year-old. - gwnn
0

#7 User is offline   gnasher 

  • Andy Bowles
  • PipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 11,993
  • Joined: 2007-May-03
  • Gender:Male
  • Location:London, UK

Posted 2011-August-31, 14:42

View PostValardent, on 2011-August-31, 11:50, said:

When J falls, playing "A" line wins also in (2), making "A" line best.


Unless J unexpectedly falls when LHO has Kx and Jxxx. (OK, I know I'm dreaming.)
... that would still not be conclusive proof, before someone wants to explain that to me as well as if I was a 5 year-old. - gwnn
0

#8 User is offline   inquiry 

  • PipPipPipPipPipPipPipPipPipPip
  • Group: Admin
  • Posts: 14,566
  • Joined: 2003-February-13
  • Gender:Male
  • Location:Amelia Island, FL
  • Interests:Bridge, what else?

Posted 2011-August-31, 15:11

View Postgnasher, on 2011-August-31, 14:24, said:

This is how you do it:

With diamonds 2=4, the vacant spaces are initially 11:9. Each time we deal a spade to one of the players, the vacant spaces for that player are reduced by one. Hence:

Chance of a specific 3=1 break: 11/20 * 10/19 * 9/18 * 9/17 = 0.077
Chance of a specific 1=3 break: 11/20 * 9/19 * 8/18 * 7/17 = 0.048
Chance of a 4=0 break: 11/20 * 10/19 * 9/18 * 8/17 = 0.068
Chance of a 0=4 break: 9/20 * 8/19 * 7/18 * 6/17 = 0.026

The divisors are only relevant if you're comparing chances in two different suits. In this case you could ignore them, but to do exact calculations at the table you'd still have to be very good at mental arithmetic.

The other way to do this type of calculation is to use Richard Pavlicek's calculator at http://www.rpbridge.net/xsb2.htm


Actually, doesn't WEST's heart king (assumption he has KQ) make vacant space after 2=4 diamond split is learned, 9:9?
--Ben--

#9 User is offline   inquiry 

  • PipPipPipPipPipPipPipPipPipPip
  • Group: Admin
  • Posts: 14,566
  • Joined: 2003-February-13
  • Gender:Male
  • Location:Amelia Island, FL
  • Interests:Bridge, what else?

Posted 2011-August-31, 15:12

View Postgnasher, on 2011-August-31, 14:24, said:

This is how you do it:

With diamonds 2=4, the vacant spaces are initially 11:9. Each time we deal a spade to one of the players, the vacant spaces for that player are reduced by one. Hence:

Chance of a specific 3=1 break: 11/20 * 10/19 * 9/18 * 9/17 = 0.077
Chance of a specific 1=3 break: 11/20 * 9/19 * 8/18 * 7/17 = 0.048
Chance of a 4=0 break: 11/20 * 10/19 * 9/18 * 8/17 = 0.068
Chance of a 0=4 break: 9/20 * 8/19 * 7/18 * 6/17 = 0.026

The divisors are only relevant if you're comparing chances in two different suits. In this case you could ignore them, but to do exact calculations at the table you'd still have to be very good at mental arithmetic.

The other way to do this type of calculation is to use Richard Pavlicek's calculator at http://www.rpbridge.net/xsb2.htm


Actually, doesn't WEST's heart king (assumption he has KQ) make vacant space after 2=4 diamond split is learned, 9:9?
--Ben--

#10 User is offline   gnasher 

  • Andy Bowles
  • PipPipPipPipPipPipPipPipPipPip
  • Group: Advanced Members
  • Posts: 11,993
  • Joined: 2007-May-03
  • Gender:Male
  • Location:London, UK

Posted 2011-August-31, 15:49

View Postinquiry, on 2011-August-31, 15:12, said:

Actually, doesn't WEST's heart king (assumption he has KQ) make vacant space after 2=4 diamond split is learned, 9:9?


I don't know. We could probably expect a heart lead anyway. If he'd led the 10, would you make the same argument about the 109?
... that would still not be conclusive proof, before someone wants to explain that to me as well as if I was a 5 year-old. - gwnn
0

#11 User is offline   semeai 

  • PipPipPipPipPip
  • Group: Full Members
  • Posts: 582
  • Joined: 2010-June-10
  • Gender:Male
  • Location:USA
  • Interests:Having eleven-syllable interests
    Counting modulo five

Posted 2011-August-31, 16:28

View Postinquiry, on 2011-August-31, 15:11, said:

Actually, doesn't WEST's heart king (assumption he has KQ) make vacant space after 2=4 diamond split is learned, 9:9?

View Postgnasher, on 2011-August-31, 15:49, said:

I don't know. We could probably expect a heart lead anyway. If he'd led the 10, would you make the same argument about the 109?


I think inquiry is basically correct. If West leads K whenever he has KQ and never otherwise, then the lead gives us precisely the inference that he has KQ (and no other info). This means we should take it into account for vacant spaces. I think it's not far from the truth that West leads K whenever he has KQ and never otherwise given our hands and the contract.

The 109 situation is much less clear. A lead from some sequence is not unlikely for this contract, so probably there's little to no adjustment to a vacant spaces calculation.
0

Page 1 of 1
  • You cannot start a new topic
  • You cannot reply to this topic

2 User(s) are reading this topic
0 members, 2 guests, 0 anonymous users