t****t
发帖数: 95 | 1 On the flop, what do you do when your opponent is on a flush draw and
pushes
all-in first, while you have the top set or straight flopped? | f*****g
发帖数: 15860 | 2 call, 101% of times.
2:1 ahead, if we're afraid of that 1, then no poker, hehe.
【在 t****t 的大作中提到】
: On the flop, what do you do when your opponent is on a flush draw and
: pushes
: all-in first, while you have the top set or straight flopped?
| t****t
发帖数: 95 | 3 That's what I did. BUT, my biggest variance came from these lately.
【在 f*****g 的大作中提到】
: call, 101% of times.
: 2:1 ahead, if we're afraid of that 1, then no poker, hehe.
| p*******p
发帖数: 13670 | 4 i will thank god and kiss their stinky ass all day long!
oh, also to call
【在 t****t 的大作中提到】
: On the flop, what do you do when your opponent is on a flush draw and
: pushes
: all-in first, while you have the top set or straight flopped?
| p****r
发帖数: 9164 | 5 that is just part of game. nothing you can do. it is a game of imcomplete
informatin and uncertainty. Getting in with good sharp is pretty much the
best you can do lots of time.
【在 t****t 的大作中提到】
: That's what I did. BUT, my biggest variance came from these lately.
| p****r
发帖数: 9164 | 6 lol. I always like Li Shu's post style. too bad that we miss you in Vegas
party.
【在 p*******p 的大作中提到】
: i will thank god and kiss their stinky ass all day long!
: oh, also to call
| t****t
发帖数: 95 | 7 right! the rest of it is beyond our control.
Indeed, in terms of pot odds, I had some thought which is related to
this
topic.
1. Pot odds implies the percentage hands that hit or miss.
2. for a given hand, the sequence of cards is deterministic but not
random.
3. If we must apply pot odd to one hand, the following must be true:
a) the pot size is consistently same. Here, pot odds/EV is perfectly
justified with little variance.
b) the pot size is random enough in large-number sample pool. EV of
EV is justified, BUT with variance dependent on the variance of the
pot
size.
For example, in coin flip situation, the pots won are smaller than the
pot
lost in average.
The assumptions are not easy to meet and the variance is expected and
sometimes would be a large number.
Thus, talking about pot odds/number of outs for one given hand is
misleading
without considering the overall picture.
imcomplete
the
【在 p****r 的大作中提到】
: lol. I always like Li Shu's post style. too bad that we miss you in Vegas
: party.
| c******q
发帖数: 456 | 8 I feel you are talking about two things here, EV of a hand and EV of a
session. For EV of a hand, the stack size doesn't play in the math because
it is a known parameter. However in term of EV of a session, stack size is
indeed a important factor to consider (especial for a live session since you
only play limited amount of hands, a few big hands could define the result
at the end). And under certain situation, playing a hand in a -EV way is
more appropriate if it can increase your EV for the session. One of hands
shared by player who folded his AQ TPTK on the flop sort of illustrated this
.
but if we move to long term view, the stack size effect should disappear if
you play in a consistent way.
BTW, "In coin flip, the pot we won is smaller than the pot we lose on
average" --- I don't see any sound logic behind this.
right!
★ Sent from iPhone App: iReader Mitbbs Lite 7.36
【在 t****t 的大作中提到】
: right! the rest of it is beyond our control.
: Indeed, in terms of pot odds, I had some thought which is related to
: this
: topic.
: 1. Pot odds implies the percentage hands that hit or miss.
: 2. for a given hand, the sequence of cards is deterministic but not
: random.
: 3. If we must apply pot odd to one hand, the following must be true:
: a) the pot size is consistently same. Here, pot odds/EV is perfectly
: justified with little variance.
| t****t
发帖数: 95 | 9 I think we share the same thought.
To be clear of my opinion:
When you calculate EV of a given hand by number of outs and pot odds, you
already made the assumption that pot size in the long run is either
consistent or random enough. Without the assumption, it's meaningless.
you
result
this
if
【在 c******q 的大作中提到】
: I feel you are talking about two things here, EV of a hand and EV of a
: session. For EV of a hand, the stack size doesn't play in the math because
: it is a known parameter. However in term of EV of a session, stack size is
: indeed a important factor to consider (especial for a live session since you
: only play limited amount of hands, a few big hands could define the result
: at the end). And under certain situation, playing a hand in a -EV way is
: more appropriate if it can increase your EV for the session. One of hands
: shared by player who folded his AQ TPTK on the flop sort of illustrated this
: .
: but if we move to long term view, the stack size effect should disappear if
| t****t
发帖数: 95 | 10 BTW, "In coin flip, the pot we won is smaller than the pot we lose on
average" --- I don't see any sound logic behind this.
This is an example of variance. In coin flip, if the pot we won is
smaller than the pot we lose on average, PL is not zero.
because
size is
since
you
result
is
hands
illustrated
this
disappear
if
【在 c******q 的大作中提到】
: I feel you are talking about two things here, EV of a hand and EV of a
: session. For EV of a hand, the stack size doesn't play in the math because
: it is a known parameter. However in term of EV of a session, stack size is
: indeed a important factor to consider (especial for a live session since you
: only play limited amount of hands, a few big hands could define the result
: at the end). And under certain situation, playing a hand in a -EV way is
: more appropriate if it can increase your EV for the session. One of hands
: shared by player who folded his AQ TPTK on the flop sort of illustrated this
: .
: but if we move to long term view, the stack size effect should disappear if
| | | W********m
发帖数: 7793 | 11 Pot size has no effect on whether your ev is + or - for one hand or over
long term. Variance also has no impact on true EV. Your actual money gain
or loss is only an estimate of the true EV. Variance has an impact on how
often your estimate can deviate from your true EV. It can deviate up or down
. The larger the sample size is, the less variance it has for the estimate
and it should always converge to your true EV. (We say variance, it is
really standard error which should be used to describe the accuracy of the
estimate of EV--we call mean. It is normally calculated by sqrt(sample
variance/sample size). Very naturally, the larger sample size, the smaller
the standard error is (more accurate of our mean estimate). That is also
why people keep talking about sample size when talking about winrate.
bottomline is here: if you always go all in with better odds. Over long run,
you should always converge to a positive ev no matter what different pot
sizes they are.
One thing pot size can affect is your probability of ruin (bankrupt)
depending on your total bankroll. Basicly, we can compute the probability of
ruin for each N (N is the number of times you go all in) if given your
bankroll, pot size and pot odds. You would be surprised that how often we
can go bankrupt even if we go all in everytime with better odds if bankroll/
potsize is small. Hence the importance of bankroll management. But that is a
totally different issue from what was dicussed here. | T*********k
发帖数: 1621 | 12 I can't agree with you more. Excellent points.
When ppl talk about long term, we have to put players in the context.
For poker pros or ppl can play everyday, the long term could mean a year, 2
years. But for recreational players who only play 30-50 sessions a year, the
long term could mean really long. Thus stack size should be a huge factor
in our decision making process. You lose 2-3 big pots in 60% vs 40% or 55%
vs 45% situation in one session, sometimes it takes you 1-2 month or even
longer to dig out of the hole, for recreational players, of course.
Sometimes, protecting your captials in marginal situations is more important
than playing +EV poker. That is my view.
Of course, flop nut str and being put all-in by flush draw and got drawn out
is just unfortunate. Nothing you can do. I had it happened to me
conservtive weeks in live game last year, cost me 100bb and 90+ bb, still
painful.
you
result
this
if
【在 c******q 的大作中提到】
: I feel you are talking about two things here, EV of a hand and EV of a
: session. For EV of a hand, the stack size doesn't play in the math because
: it is a known parameter. However in term of EV of a session, stack size is
: indeed a important factor to consider (especial for a live session since you
: only play limited amount of hands, a few big hands could define the result
: at the end). And under certain situation, playing a hand in a -EV way is
: more appropriate if it can increase your EV for the session. One of hands
: shared by player who folded his AQ TPTK on the flop sort of illustrated this
: .
: but if we move to long term view, the stack size effect should disappear if
| s*******o
发帖数: 4896 | 13 laf I saw this arguements way many times.
One mathmatical side always claim bankroll management, EV, and long run.
On the other side, people always claim you cannot find a realist way to
get large sample size, thus reduce variance/sample size ratio.Thus, they
always consider protecting their stack as most important thing, even it
is -EV play(e.g., buy insurance).
I can't see either these two side can convince the other side. hahahaha
down
estimate
【在 W********m 的大作中提到】
: Pot size has no effect on whether your ev is + or - for one hand or over
: long term. Variance also has no impact on true EV. Your actual money gain
: or loss is only an estimate of the true EV. Variance has an impact on how
: often your estimate can deviate from your true EV. It can deviate up or down
: . The larger the sample size is, the less variance it has for the estimate
: and it should always converge to your true EV. (We say variance, it is
: really standard error which should be used to describe the accuracy of the
: estimate of EV--we call mean. It is normally calculated by sqrt(sample
: variance/sample size). Very naturally, the larger sample size, the smaller
: the standard error is (more accurate of our mean estimate). That is also
| W********m
发帖数: 7793 | 14 I am not trying to convince anyone how they play. People can choose to play
with maximum ev and high variance or less ev and low variance or anyway they
want.
What I said here is just stat 101. I just thought there were some mis-
conception in the early post.
【在 s*******o 的大作中提到】
: laf I saw this arguements way many times.
: One mathmatical side always claim bankroll management, EV, and long run.
: On the other side, people always claim you cannot find a realist way to
: get large sample size, thus reduce variance/sample size ratio.Thus, they
: always consider protecting their stack as most important thing, even it
: is -EV play(e.g., buy insurance).
: I can't see either these two side can convince the other side. hahahaha
:
: down
: estimate
| p****r
发帖数: 9164 | 15 we can assume most ppl are gambling with limited BR. We should always think
about risk averse factor when making decision, even at poker table ie, EV
is NOT the only factor that we should consider. This is also fun part of
poker,IMO.
In Bill chen's book, it gave a simple example. If we are given a
chance to choose two following options. We use coin flip as gambling tool.
Chance are strictly 50/50
1. We keep 1 million$ without any risk/gambling.
2. We gamble 1 time coin flip, if we win , we got 4 million, if we lose
, we got 0$. And we can only gamble once.
Option 2 is very postive EV since we make 1 million more long term.
Guys like Bill Gates would prolly choose option 2. But for most ppl, option
1 is way better since 1 million$ make difference for their life and 3
million more would not make that much difference.
But if we choose the amount to 100$ instead of 1 million. most ppl
would choose option 2 since it is very +EV and they do not care losing it.
It is related to utility theory in economics.
What if we choose the amount to 10,000$ and we do a gambling game
with closer odds? Say if we have QQ and we know we are against AKs. Do we
choose to keep the 10k$ without taking any risk or choose to gamble with QQ
vs AKs race and have 54% chance to take home 20k$?
I guess the answer is really up to your BR and your willingness
to take risk. We can assume 10k$ is a lot of money for most ppl but not life
changing money. You will lose money/EV long term if you choose option 1 ,
but it is a lot safer as well. We can find chance later on like getting in
preflop with AA VS AK and getting much better long term return with much
lower risk.
所以我同意饼干兄和AK兄，做一个决定的时候，不能只想一个hand 的EV, 也
要“量力而为”。
【在 s*******o 的大作中提到】
: laf I saw this arguements way many times.
: One mathmatical side always claim bankroll management, EV, and long run.
: On the other side, people always claim you cannot find a realist way to
: get large sample size, thus reduce variance/sample size ratio.Thus, they
: always consider protecting their stack as most important thing, even it
: is -EV play(e.g., buy insurance).
: I can't see either these two side can convince the other side. hahahaha
:
: down
: estimate
| p*******p
发帖数: 13670 | 16 没bank roll的么就随便玩玩了，for fun，for fun呀
think
lose
【在 p****r 的大作中提到】
: we can assume most ppl are gambling with limited BR. We should always think
: about risk averse factor when making decision, even at poker table ie, EV
: is NOT the only factor that we should consider. This is also fun part of
: poker,IMO.
:
: In Bill chen's book, it gave a simple example. If we are given a
: chance to choose two following options. We use coin flip as gambling tool.
: Chance are strictly 50/50
: 1. We keep 1 million$ without any risk/gambling.
: 2. We gamble 1 time coin flip, if we win , we got 4 million, if we lose
| b*******s
发帖数: 1175 | 17 excellent analysis. I think this applies to most recreational players.学习了
。
think
lose
【在 p****r 的大作中提到】
: we can assume most ppl are gambling with limited BR. We should always think
: about risk averse factor when making decision, even at poker table ie, EV
: is NOT the only factor that we should consider. This is also fun part of
: poker,IMO.
:
: In Bill chen's book, it gave a simple example. If we are given a
: chance to choose two following options. We use coin flip as gambling tool.
: Chance are strictly 50/50
: 1. We keep 1 million$ without any risk/gambling.
: 2. We gamble 1 time coin flip, if we win , we got 4 million, if we lose
|
|
