i****e
发帖数: 642 | 1 You miss KT8x in trump:
Jxxx
vs
AQ9xx
To try to pick up all tricks, is it better to play J first, or small to Q
first?
Everyone seems to say the latter. But playing J fails immediately only with
RHO holding singleton K, while playing to Q will immediately fail if LHO has
singleton 8, x or void. It is 3 vs 1 to me.
Did I missing anything? | b***y
发帖数: 2804 | 2 It's not 3 to 1. Reason is simple: say you play J, RHO covers K, LHO plays 8
or x, are you going to finesse 9 next? | i****e
发帖数: 642 | 3 Yes, that is a guess, but you have a guess. If you play small to Q, and LHO
is single 8 or x, you even don't have the chance to guess.
8
【在 b***y 的大作中提到】
: It's not 3 to 1. Reason is simple: say you play J, RHO covers K, LHO plays 8
: or x, are you going to finesse 9 next?
| b***y
发帖数: 2804 | 4 This is a common fallacy. Just having a chance to guess right doesn't mean
it is a superior line. For example, if you have a two-way finesse, in theory
you can make 100% of time if you guess right, but that doesn't mean taking
the finesse is better than an alternative line which has 60% of success rate.
Look it in another way. We are actually breaking down to three lines, not
two:
1) Play small to Q, then cash A.
2) Play J, cover by K, A. LHO follows 8 or x. Play Q next.
3) Play J, cover by K, A. LHO follows 8 or x. Finesse with 9 next.
In theory, you can combine chances for line 2 and 3, if you always know how
the cards lie. But you don't, so you have to choose one of them. So, we are
really comparing 3 lines, see which one is best.
【在 i****e 的大作中提到】
: Yes, that is a guess, but you have a guess. If you play small to Q, and LHO
: is single 8 or x, you even don't have the chance to guess.
:
: 8
| i****e
发帖数: 642 | 5 That sounds a good analysis. So we are comparing 3 lines, not 2.
(1) Q to A fails if LHO is void.
(2) J then Q fails if RHO has stiff K.
(3) J then 9 fails if RHO has K8 or Kx.
Both of above is a single case, and stiff K is a little higher probability,
but is almost negligible. But with possible line (3), starting with J still
seems much better.
theory
taking
rate.
how
【在 b***y 的大作中提到】
: This is a common fallacy. Just having a chance to guess right doesn't mean
: it is a superior line. For example, if you have a two-way finesse, in theory
: you can make 100% of time if you guess right, but that doesn't mean taking
: the finesse is better than an alternative line which has 60% of success rate.
: Look it in another way. We are actually breaking down to three lines, not
: two:
: 1) Play small to Q, then cash A.
: 2) Play J, cover by K, A. LHO follows 8 or x. Play Q next.
: 3) Play J, cover by K, A. LHO follows 8 or x. Finesse with 9 next.
: In theory, you can combine chances for line 2 and 3, if you always know how
| b***y
发帖数: 2804 | 6 Line 3 fails if RHO has K8, Kx, or stiff K.
Actually line 3 is the worst line.
【在 i****e 的大作中提到】
: That sounds a good analysis. So we are comparing 3 lines, not 2.
: (1) Q to A fails if LHO is void.
: (2) J then Q fails if RHO has stiff K.
: (3) J then 9 fails if RHO has K8 or Kx.
: Both of above is a single case, and stiff K is a little higher probability,
: but is almost negligible. But with possible line (3), starting with J still
: seems much better.
:
: theory
: taking
| i****e
发帖数: 642 | 7 That is true. But it also starts with J. So doesn't make it better to start
with J?
【在 b***y 的大作中提到】
: Line 3 fails if RHO has K8, Kx, or stiff K.
: Actually line 3 is the worst line.
| b***y
发帖数: 2804 | 8 The more failure cases, the worse the line, right? Case 3 fails on: RHO
stiff K, RHO Kx, RHO K8. How does that make it better?
【在 i****e 的大作中提到】
: That is true. But it also starts with J. So doesn't make it better to start
: with J?
| i****e
发帖数: 642 | 9 I did not mean line 3 is better. Both line 2 and line 3 start with J. My
point is it is better to start with J.
【在 b***y 的大作中提到】
: The more failure cases, the worse the line, right? Case 3 fails on: RHO
: stiff K, RHO Kx, RHO K8. How does that make it better?
| b***y
发帖数: 2804 | 10 Why? Because combined chances for line 2 + line 3 is better than line 1?
As I mentioned, you cannot add those together. Starting with J may not be
immediately fatal, but you have to guess on next round, and you have to
calculate the percentage that you will guess wrong, then derive the success
rate of either line 2 or 3, but not both, then compare that with line 1.
To give you an example, which looks different but actually the same. Say you
are in a slam, you cannot lose another trick. Suit 1 is AJT9 vs Kxxx, suit
2 is xxx vs AQJT9. Entry is plenty. You just need 5 tricks between those 2
suits.
Line 1: cash AK in suit 1, if Q doesn't drop, then take finesse in suit 2
Line 2: play J at suit 1, let it ride.
Line 3: Play J at suit 1, take K, then low toward 9.
Both line 2 and 3 cover the missing Q, combining them together you get 100%
chance. Does this mean starting J in suit 1 is better than line 1?
【在 i****e 的大作中提到】
: I did not mean line 3 is better. Both line 2 and line 3 start with J. My
: point is it is better to start with J.
| i****e
发帖数: 642 | 11 Your example is very different.
Line 2 (J first then Q to drop T) is almost as good as Line 1 (low to Q
first). The minor difference is the probability of KT8x is slightly lower
than stiff K with RHO (0-4 vs 1-3).
But starting with J will reduce the chance immediate failure, so give you
the chance to guess next round. The immediate fatal chance is 1 vs 3 (which
doesn't mean the successful chance is 3 vs 1, as I now understand). So it is
better in my opinion.
success
you
suit
【在 b***y 的大作中提到】
: Why? Because combined chances for line 2 + line 3 is better than line 1?
: As I mentioned, you cannot add those together. Starting with J may not be
: immediately fatal, but you have to guess on next round, and you have to
: calculate the percentage that you will guess wrong, then derive the success
: rate of either line 2 or 3, but not both, then compare that with line 1.
: To give you an example, which looks different but actually the same. Say you
: are in a slam, you cannot lose another trick. Suit 1 is AJT9 vs Kxxx, suit
: 2 is xxx vs AQJT9. Entry is plenty. You just need 5 tricks between those 2
: suits.
: Line 1: cash AK in suit 1, if Q doesn't drop, then take finesse in suit 2
| a****s
发帖数: 524 | 12 相关的情行一共8种：
KT82
KT8
KT2
K82
KT
K8
K2
K
1. 出小到Q：任意K双张，单K，还有K82，共5种情况成功。
2. 出J，再出Q：任意K双张，KT82, K82, 共5种情况成功。
3. 出J, 再飞9：任意K三张, KT82, KT, 共5种情况成功。
所以理论上，在你有理由认为上家这门花色较长时，路线3更好。否则，路线1最佳。
实战中，打J也许有一点附带的好处，就是也许能从对方的反应，猜对下面的打法。（
所谓的2和3的组合）。 这个就是艺术了。
with
has
【在 i****e 的大作中提到】
: You miss KT8x in trump:
: Jxxx
: vs
: AQ9xx
: To try to pick up all tricks, is it better to play J first, or small to Q
: first?
: Everyone seems to say the latter. But playing J fails immediately only with
: RHO holding singleton K, while playing to Q will immediately fail if LHO has
: singleton 8, x or void. It is 3 vs 1 to me.
: Did I missing anything?
| b***y
发帖数: 2804 | 13 Examples are the same. But I don't know how to make you see it.
Again, "give you a chance to guess better" doesn't mean it is better than "
immediate fail". For argument's sake, let's say line 2 and 3 are exactly the
same chance (they are actually not, but I am not going to elaborate it). So
you play SJ, covered by SK, SA.
Now, next round, you play low, RHO follows but not with 10. In theory, you
can make it 100% of time now, if you can guess right whether to drop 10 or
finesse. It's same as, in my example, you have a 2-way finesse, you can make
it 100% of time, but in practice you will fail 50% of time.
In other words, in those layouts, even though SJ doesn't fail immediately,
it still has 50% of chance to fail later. You are only delaying the problem.
What you care is "eventual success", not "I didn't fail initially but I
failed later". Down 1 is down 1, doesn't matter you immediately go down or
you go down at last trick.
【在 i****e 的大作中提到】
: Your example is very different.
: Line 2 (J first then Q to drop T) is almost as good as Line 1 (low to Q
: first). The minor difference is the probability of KT8x is slightly lower
: than stiff K with RHO (0-4 vs 1-3).
: But starting with J will reduce the chance immediate failure, so give you
: the chance to guess next round. The immediate fatal chance is 1 vs 3 (which
: doesn't mean the successful chance is 3 vs 1, as I now understand). So it is
: better in my opinion.
:
: success
| b***y
发帖数: 2804 | 14 如果AQ9xx在暗手的话，实战中打J还有一个好处，就是右手方不一定扑K。他要考虑同
伴单张Q的可能性。
所以具体到一副牌，根据叫牌打牌线索，以及临场反应，出J未必是错。但如果主题是
纯概率计算，那么仅仅因为2和3的组合概率更大，就说出J更优，这个是不准确的。要
把右手单K（immediate failure），以及后一轮猜错（later failure）一起算进去。
【在 a****s 的大作中提到】
: 相关的情行一共8种：
: KT82
: KT8
: KT2
: K82
: KT
: K8
: K2
: K
: 1. 出小到Q：任意K双张，单K，还有K82，共5种情况成功。
|
|
