I'm sure all the experts can figure out the correct line, so do give my fellow I/As a crack at it please.
Play 7♦ better than I did
#1
Posted 2014-October-05, 10:23
I'm sure all the experts can figure out the correct line, so do give my fellow I/As a crack at it please.
-- Bertrand Russell
#6
Posted 2014-October-07, 09:06
#7
Posted 2014-October-07, 10:23
-- Bertrand Russell
#9
Posted 2014-October-07, 11:56
Disclaimer: I am notoriously bad at these calculations. Take the word of an expert over mine.
West was forced to show 5 cards: the spade at trick 1 and his 4 trumps. East was forced to show 1 card: his spade at trick 1. The heart discard was a free choice.
That means that at that point West has only 8 vacant spaces, whereas East has 12. This means that the probability for a successful finesse against the ♥K has increased from 50% to 60%. At the same time, the probabilities for the club split are affected. Originally, the probability for a 3-2 split was 67.83%, making playing for the club split superior. But when the vacant spaces split 8-12, this decreases.
When you ruff 2 spades (assuming no one shows out) the probabilities change even more. Now West has 6 vacant spaces and East 10. The finesse is now 62.5%. And, the probability for a 3-2 split will be even lower.
I would guess that playing for the 3-2 split is technically still superior, but not by much. This means that against robots I would play for the dummy reversal, but against people I might take the finesse based on my table feel (which, unfortunately, is also notoriously bad ).
Rik
The most exciting phrase to hear in science, the one that heralds the new discoveries, is not Eureka! (I found it!), but Thats funny Isaac Asimov
The only reason God did not put "Thou shalt mind thine own business" in the Ten Commandments was that He thought that it was too obvious to need stating. - Kenberg
#10
Posted 2014-October-07, 13:03
#11
Posted 2014-October-07, 15:00
Trinidad, on 2014-October-07, 11:56, said:
Playing fast and loose with vacant spaces unfortunately does not give the correct answer in this case.
Given that diamonds split 4-0, the a priori chances of various spade splits are:
3-6 28.98%
4-5 32.60%
5-4 18.11%
6-3 4.83%
If we know that it is one of these splits, we can normalise these to 100% and get
3-6 34.29%
4-5 38.57%
5-4 21.43%
6-3 5.71%
Multiplying these chances by the respective portion of heart cards East holds in these scenarios gives only a 61.43% chance for the finesse to work.
On the other hand, here are some W-E open spaces and the chance of a 3-2 or 2-3 club split in each case:
6-7 73.43%
5-8 65.27%
4-9 50.35%
3-10 31.47%
Multiplying with the previous table, there is a 62.94% chance that clubs break.
So you are correct that it is actually very close.
Edit: I was originally off by one on the number of hearts held by East giving worse odds for the finesse.
This post has been edited by mgoetze: 2014-October-07, 15:10
-- Bertrand Russell
#12
Posted 2014-October-07, 15:19
#13
Posted 2014-October-07, 17:23
#14
Posted 2014-October-07, 17:55
johnu, on 2014-October-07, 17:23, said:
So basically what you're saying is, you're not even going to bother trying to make the contract?
-- Bertrand Russell
#16
Posted 2014-October-07, 19:40
mgoetze, on 2014-October-07, 17:55, said:
I'm saying that all those detailed heart/club calculations are basically a waste of time because if clubs come in you have the 2 additional tricks you need without the parlay of heart finesse and squeeze where you probably East to make a mistake.
#17
Posted 2014-October-08, 10:22
johnu, on 2014-October-07, 17:23, said:
The situation in heart is particolar: whatever East do doesn't escape. Let's suppose that he had King and play it then we have a re-entry with Jack and we ruff 5 (second ruff the first in spade). Now we run diamonds on the last East is squeezed spade/club (dummy reversal+simple squeeze)To avoid this East if don't play King Queen run, 5 to Ace and now we have spade J against Q, heart J against K, xxx/AKQxx in club for an automatic triple squeeze to gain two tricks (let's note that we don't consider clubs covering all shapes in suit). If East had (but is not probable here) hearts 6th 5 can consider menace (rare).
#18
Posted 2014-October-08, 18:04
Lovera, on 2014-October-08, 10:22, said:
♠A, ♥A, 2 ♠ ruffs, 4 top ♦s, and 3 top ♣s = 11 tricks, so you need to find 2 more.
If ♣s are 3-2, then the long ♣s suffice.
Otherwise..
Lovera can't' execute a ♣/♠ squeeze because all North's ♠s have been ruffed. Hence, as JohnU points out, if you have to finesse ♥ before finding out if ♣s break, then the ♥ finesse is likely to be a so-called "practice" finesse -- useful for rehearsing your finessing technique but unlikely to help make the contract. The ♥ finesse helps only if you can develop a ♣ / ♥ squeeze i.e. East fails to cover or he has 6 ♥s.
What might slightly increase your chances is (after ♠A, ♦Q, finding that west has 4 ♦s, ♠ ruff), at trick 4, try to cash ♣AK (see earlier reply).
#19
Posted 2014-October-08, 18:29
mgoetze, on 2014-October-07, 15:00, said:
Given that diamonds split 4-0, the a priori chances of various spade splits are:
3-6 28.98%
4-5 32.60%
5-4 18.11%
6-3 4.83%
If we know that it is one of these splits, we can normalise these to 100% and get
3-6 34.29%
4-5 38.57%
5-4 21.43%
6-3 5.71%
Multiplying these chances by the respective portion of heart cards East holds in these scenarios gives only a 61.43% chance for the finesse to work.
On the other hand, here are some W-E open spaces and the chance of a 3-2 or 2-3 club split in each case:
6-7 73.43%
5-8 65.27%
4-9 50.35%
3-10 31.47%
Multiplying with the previous table, there is a 62.94% chance that clubs break.
So you are correct that it is actually very close.
If East covers the North's first ♥ honour, then you need him ot have started with 6 ♥s for the squeeze to work.
Finally, the squeeze-line works only if West has fewer than 4 ♣s.
#20
Posted 2014-October-08, 21:17
Welp, I just got 54.81% on a Robot Rebate 55%, mainly thanks to blowing this hand: I'm sure all the experts can figure out the correct line, so do give my fellow I/As a crack at it please