Useful bridge/math tidbit
#1
Posted 2010-September-28, 12:01
A432
This is your trump suit. Your LHO gets a ruff early in the play. When you gain the lead, you cash the Ace of trump and lead a second round, but you see nothing except small cards. Finesse or drop?
This situation is not uncommon in real life and (I think!) I have found a simple rule for knowing what to do (though, as usual, I might have made a math error):
If your RHO has 3+ more known cards than your LHO, you should finesse.
If your RHO has exactly 2 more known cards than your LHO, you are faced with a complete guess (Qxxx on your left and Qx on your right are equally likely).
If your RHO has 1 or 0 more known cards than your LHO, you should play for the drop.
This is somewhat obvious if you think about it, but by remembering this rule you don't have to think about it
Fred Gitelman
Bridge Base Inc.
www.bridgebase.com
#2
Posted 2010-September-28, 14:27
#3
Posted 2010-September-28, 16:14
fred, on Sep 28 2010, 06:01 PM, said:
A432
This is your trump suit. Your LHO gets a ruff early in the play. When you gain the lead, you cash the Ace of trump and lead a second round, but you see nothing except small cards. Finesse or drop?
This situation is not uncommon in real life and (I think!) I have found a simple rule for knowing what to do (though, as usual, I might have made a math error):
If your RHO has 3+ more known cards than your LHO, you should finesse.
If your RHO has exactly 2 more known cards than your LHO, you are faced with a complete guess (Qxxx on your left and Qx on your right are equally likely).
If your RHO has 1 or 0 more known cards than your LHO, you should play for the drop.
This is somewhat obvious if you think about it, but by remembering this rule you don't have to think about it
Fred Gitelman
Bridge Base Inc.
www.bridgebase.com
Any chance of an explanation for the somewhat idiots like myself who don't find this somewhat (or the least bit) obvious?
#4
Posted 2010-September-28, 16:35
jonottawa, on Sep 28 2010, 10:14 PM, said:
fred, on Sep 28 2010, 06:01 PM, said:
A432
This is your trump suit. Your LHO gets a ruff early in the play. When you gain the lead, you cash the Ace of trump and lead a second round, but you see nothing except small cards. Finesse or drop?
This situation is not uncommon in real life and (I think!) I have found a simple rule for knowing what to do (though, as usual, I might have made a math error):
If your RHO has 3+ more known cards than your LHO, you should finesse.
If your RHO has exactly 2 more known cards than your LHO, you are faced with a complete guess (Qxxx on your left and Qx on your right are equally likely).
If your RHO has 1 or 0 more known cards than your LHO, you should play for the drop.
This is somewhat obvious if you think about it, but by remembering this rule you don't have to think about it
Fred Gitelman
Bridge Base Inc.
www.bridgebase.com
Any chance of an explanation for the somewhat idiots like myself who don't find this somewhat (or the least bit) obvious?
The LHO shows two more trumps than RHO.
Therefore, if the difference of side suit shown distribution of LHO and RHO < 2,
RHO is more likely to hold the Q. (because the probability of holding Q is proportional to the unknown cards in one's hand)
For example, LHO leads a doubleton to partner's AK and obtains a ruff and your side holds 8 card in that side suit, then LHO shows 2 cards and RHO shows 3 cards in that suit. Therefore, you should play to drop.
If it is greater than 2, LHO is more likely to hold Q.
For example, LHO leads his singleton to partner's ace and gets a ruff, his partner holds 5 card in that suit. You should take the finesse.
If it equals 2, it's just a guess.
For example, LHO leads a singleton, his partner holds Axx and gives him a ruff.
#5
Posted 2010-September-28, 17:10
junyi_zhu, on Sep 28 2010, 10:35 PM, said:
I do not think Fred said anything about trump length having been shown. Fred, would your rule also apply if we had 9 trumps and RHO had, say, opened a weak 2 and marked the distribution in a suit that way instead? (ie 6 opposite 4 = guess)
#6
Posted 2010-September-28, 17:32
Zelandakh, on Sep 28 2010, 06:10 PM, said:
junyi_zhu, on Sep 28 2010, 10:35 PM, said:
I do not think Fred said anything about trump length having been shown. Fred, would your rule also apply if we had 9 trumps and RHO had, say, opened a weak 2 and marked the distribution in a suit that way instead? (ie 6 opposite 4 = guess)
I think what he meant was, at the time you played A and a second trump (and LHO followed low), LHO has shown up 3 trumps and RHO 1. At that point (RHO is yet to play), LHO showed 2 more trumps than RHO. That of course doesn't mean LHO actually holds 2 more trumps originally.
#7
Posted 2010-September-28, 17:44
Suppose that a side-suit broke 2=4. RHO has been dealt two more cards than LHO. Then we deal LHO three specific trumps, and RHO one specific trump. Now each of them has been dealt five cards. Therefore it is evens whether the remaining trump goes to LHO or to RHO. Hence a specific 4=1 break is exactly as likely as a specific 3=2 break.
There are four possible Qxxx-x breaks, and four possible xxx-Qx breaks. Hence the two breaks are equally likely.
#8
Posted 2010-September-28, 17:49
#9
Posted 2010-September-28, 17:59
Zelandakh, on Sep 28 2010, 11:10 PM, said:
junyi_zhu, on Sep 28 2010, 10:35 PM, said:
I do not think Fred said anything about trump length having been shown. Fred, would your rule also apply if we had 9 trumps and RHO had, say, opened a weak 2 and marked the distribution in a suit that way instead? (ie 6 opposite 4 = guess)
This is the correct way to count. His assumption is that 4 trumps are shown and trump Q is missing.
The magic number of 2 is just due to the shown trump number difference, which is also 2.
The other number 2 rule case is the 10 trumps case,
suppose your LHO gets a ruff and plays low under your Q, should you drop or finesse K?
This is also determined by the side suit shown cards difference.
#10
Posted 2010-September-28, 18:06
#11
Posted 2010-September-28, 18:31
#12
Posted 2010-September-28, 18:52
gnasher, on Sep 29 2010, 12:31 AM, said:
Dude, don't self-edit like that, gnasher gibberish is way more interesting/insightful than 95% of the stuff I read on here anyway.
#13
Posted 2010-September-28, 21:20
i certainly agree with fred about the arithmetical side of his demonstration
however, the fact that lho has got an early ruff basically means that he led to get one..do we all do that with a potential trump trick?
#14
Posted 2010-September-28, 22:28
ritong, on Sep 28 2010, 08:20 PM, said:
i certainly agree with fred about the arithmetical side of his demonstration
however, the fact that lho has got an early ruff basically means that he led to get one..do we all do that with a potential trump trick?
Yes when it's Qxxx.
#15
Posted 2010-September-28, 23:04
#16
Posted 2010-September-28, 23:37
junyi_zhu, on Sep 28 2010, 06:59 PM, said:
Zelandakh, on Sep 28 2010, 11:10 PM, said:
junyi_zhu, on Sep 28 2010, 10:35 PM, said:
I do not think Fred said anything about trump length having been shown. Fred, would your rule also apply if we had 9 trumps and RHO had, say, opened a weak 2 and marked the distribution in a suit that way instead? (ie 6 opposite 4 = guess)
This is the correct way to count. His assumption is that 4 trumps are shown and trump Q is missing.
The magic number of 2 is just due to the shown trump number difference, which is also 2.
The other number 2 rule case is the 10 trumps case,
suppose your LHO gets a ruff and plays low under your Q, should you drop or finesse K?
This is also determined by the side suit shown cards difference.
I'm a little confused. For deciding "odds" shouldn't it also matter how many cards remain to be played? Or does that simply determine how much of an advantage these observations give you?
Never tell the same lie twice. - Elim Garek on the real moral of "The boy who cried wolf"
#17
Posted 2010-September-29, 00:32
Little Kid, on Sep 28 2010, 05:06 PM, said:
You can't count cards for vacant spaces that have been played in a suit when only some of the cards of a suit have been played. However, you can subdivide a suit into "interesting cards" and "spots" and count them for vacant spaces as if they were two different suits and count one of the suits once it has been played (all the spot cards in this example). Essentially there are "two" opponents trump suits in this example: Q and 5678. In Fred's example we've already seen the full set of spot cards (one trump, one card each on the A, and a small card on the 2nd round of trump). Basically an opponent can play the spot cards in any order, but can't play the Q until they play the spot cards (without losing a trick), so you can count the spot cards in the vacant spaces calculation. So if some side suit was 1=3 (the one that the trump was scored in maybe) and you've got a complete count on it and don't have a complete count on any other suit then it is 50-50. If the side suit were 1=4 you'd want to finesse.
#18
Posted 2010-September-29, 00:43
rogerclee, on Sep 28 2010, 11:28 PM, said:
ritong, on Sep 28 2010, 08:20 PM, said:
i certainly agree with fred about the arithmetical side of his demonstration
however, the fact that lho has got an early ruff basically means that he led to get one..do we all do that with a potential trump trick?
Yes when it's Qxxx.
lol
#19
Posted 2010-September-29, 08:48
ritong, on Sep 28 2010, 10:20 PM, said:
i certainly agree with fred about the arithmetical side of his demonstration
however, the fact that lho has got an early ruff basically means that he led to get one..do we all do that with a potential trump trick?
generally a good question but given our spots I can see it is easy going for a ruff in this situation
the Freman, Chani from the move "Dune"
"I learned long ago, never to wrestle with a pig. You get dirty, and besides, the pig likes it."
George Bernard Shaw
#20
Posted 2010-September-29, 09:38
Little Kid, on Sep 29 2010, 12:06 AM, said:
yeah, just a simple matter of counting vacancies....