Bubble - to blind steal or not to blind steal?

[Scuba Chuck]

Party Poker No-Limit Hold'em Tourney, Big Blind is t400 (4 handed) converter (http://www.selachian.com/tools/bisonconverter/hhconverter.cgi)

BB <font color="#A500AF">(Villain)</font> (t1830)
UTG (t2596)
Button (t1820)
Hero (t1754)

Preflop: Hero is SB with 3/images/graemlins/heart.gif, 7/images/graemlins/heart.gif.
UTG folds, Button folds, <font color="#CC3333">Will you try to blind steal with this crap? If so, do you push, or miniraise?

[gorgeous]

What buy-in?

[Xhiggy]

very much player dependent (in terms of if you steal, and how). default move would definitely be fold.

[Scuba Chuck]

Buyin is $20+2.

Let's say the player isn't a maniac or loosey goosey.

[eastbay]

[ QUOTE ]

Party Poker No-Limit Hold'em Tourney, Big Blind is t400 (4 handed) converter (http://www.selachian.com/tools/bisonconverter/hhconverter.cgi)

BB <font color="#A500AF">(Villain)</font> (t1830)
UTG (t2596)
Button (t1820)
Hero (t1754)

Preflop: Hero is SB with 3/images/graemlins/heart.gif, 7/images/graemlins/heart.gif.
UTG folds, Button folds, Will you try to blind steal with this crap? If so, do you push, or miniraise?

[/ QUOTE ]

Do the math, Scuba. Give a man a fish...

eastbay

[ilya]

That's a powerful suited connector you got there, Scuba...

poo shhhh!

[Scuba Chuck]

Thanks, that's what I needed to hear.

[mosch]

This isn't a decision that should be made in isolation.

Blind steal frequencies are largely depeondant on how often your blind is being attacked. If it's never being attacked, there's no reason to expand your raising range very much. If it's under constant assault, then retaliatory measures are neccessary and it's time to play some hard cards.

[El Maximo]

My default play on the bubble vs an equal sized stack is to push any 2. I will admit to the occasional fold when Im running poorly. I lose my nerve sometimes. I also will stray from my default play if the blind is a real maniac.

[El Maximo]

Let me take a shot at the math on this one:

SHOVE EV = 1.5 - C + CX (2W-1)
X= stack size
C = % of time you are called
W = % of time you win when you are called

I think a reasonable calling range for BB would be the Top 20%.

73s wins 33% against this calling hand range

SHOVE EV = 1.5 - .20 + (.20 x 4.38) x ((2 x .33) -1 )
SHOVE EV = 1.3 + (.876 x -.44)
SHOVE EV = 1.3 - .385
SHOVE EV = .915

.915 x T400 = T366

[Scuba Chuck]

El Maximo, I like the math.

What does it mean?

[El Maximo]

It is my long term expectation in tournament chips. This is assuming my read on the BB is correct. It is also assuming my math is correct. I have trouble with simple addition sometimes.

[Scuba Chuck]

Scuba Chuck at the chalkboard in front of the classroom...

Due to the nature of this scenario, the bubble, this seems to me a prize structure question. So I will approach it that way. (versus a chip EV question)

There are two calculations:
1. Folding
2. Pushing (with 3 sub-calculations)
a. Push, I steal
b. Push, I lose
c. Push, I win

Push, I steal
Due to the fact that we all have similar stacksizes, let’s assign the probability that we all have a 75% chance of being ITM. Further, at this time, I will assign a probability that I only have a 1/3 chance at getting each of the ITM prizes.

75% (1/3 * ((40+60+100) – entry fee)) = 75% ($66.66 – entry fee) = (.75)($44.66) = $33.50

Push, I lose (Booby prize)
37 of hearts is such a dominant hand “hot and cold”, I will give it a 72% chance of NOT winning against any group of hands that might call it.

72% (-$22) = -$16

Push, I win
…but if I get some dumb luck, then I will be the chip stack leader with decent odds of taking first or second place. 28% probability

(I will assume here that Villain is OTM if he loses here, as 76 chips won’t carry him far)

3508 Total chips if I win, out of 8,000 total = 44% of total chips.
44% probability of taking 1st
32% probability of taking 2nd
24% probability of taking 3rd

28%((.44*100)(.32*60)(.24*40) – entry fee) = 28% ($72.80 – entry fee ) = 28% ($50.8) = $14.22

Okay, here’s where I have some trouble. What kind of probability do I assign to getting caught stealing? 50/50? 60/40? 70/30?

For the moment, I will use 70/30, as that seems like a likely number. (My second choice is 80/20 BTW)

Final Equation:
(70%)Push, I steal + (30%)(72%*Push, I lose)(28%*Push, I win)

(.70 x $33.50) + (.30 x .72 x $-22) + (.30 x .28 x $50.8)
$23.45 + (-$4.75) + $4.26
= $22.96

Is this correct?

Folding
Post folding, hero will have 1554 chips, and the small stack. Due to the fact that I have the smallest stack, I have a less than 75% probability of being ITM. But I believe the range of probabilities is somewhere between 50-75%. Rather than coming up with a probability, I thought I’d back out a minimum probability using the PUSH results.

$22.96 = PI*((F1*100)+(F2*60)+(F3*40)-entry fee)

PI = probability of finishing ITM (Variable we’re solving for)
F1 = probability of finish 1st
F2 = probability of finish 2nd
F3 = probability of finish 3rd

To make calculations easy, I will assume F1=F2=F3. This I’m sure is debatable, but so are many of our other assumptions.

Fitting numbers into our calculation than

$22.96 = PI * ((($100+60+40)/3)-$22)

$22.96 = PI * ($44.66)

51% = PI

In otherwords, if I believe my probability of finishing ITM is 51% or more by folding, this is the best choice.

So, is this math correct?

Other thoughts:

Does Shove EV really apply here?
What is $EV? Is that Prize structure implications?

[assron]

you can definitely pull this off imo... he has to have a hand to call off his whole stack. you'll probably need a good hand the next time, but not this one.

[ilya]

Your results may accidentally be correct, but your methodology is not.

[KenProspero]

Chuck:

What do you think your table image is here? If you've been tight as a drum all game, you can certainly pull it off unless BB really has a big hand.

If you've been loosey goosey all game, or have already shifted gears so Villian may suspect that you're now playing loose, it's a bit more of a crap shoot.

If say you've stolen the blinds the last three orbits and Villian is observant, I think you're in trouble if you try it.

[El Maximo]

Scuba,

There are several examples of ICM calculations and how to do them in the archives. Use the search function. I think Deathgrind did a very good article on how to calculate prize pool using ICM. I use PUSH EV because it is a simple calculation. ICM Calculations should yield similiar results.

Fold = 21.93% of Prize Pool
Push = 26.5% of the Prize Pool

[Scuba Chuck]

To Ilya, and El Maximo.

I searched out the Dethgrind thread. I don't see how my methodology is that far off base. I do appreciate the advantage of having an ICM calculator, but I want to be able to do "long hand" calculations before I rely on any calculators.

Below, I attached the Dethgrind thread. I will try and update my math so it better matches, but I don't see how I am that far off.

BTW, according to El Maximo's ICM calc, folding is not the correct answer.

Please, I am not trying to be critical. I am truly trying to learn.

http://forumserver.twoplustwo.com/showflat.php?Cat=&amp;Number=1122239&amp;page=&amp;view=&amp;sb=5&amp; o=&amp;fpart=all&amp;vc=1

[microbet]

[ QUOTE ]
I think a reasonable calling range for BB would be the Top 20%.
[ QUOTE ]
T366

[/ QUOTE ]

[/ QUOTE ]

Maybe just picking a nit here, but the first statement has 1 significant figure. That's all the last one should have, if that many.

It's not just this post, but there are an awful lot of precise looking numbers in these math posts when they often depend on what range of hands you think your opponant has.

[El Maximo]

[ QUOTE ]
Maybe just picking a nit here, but the first statement has 1 significant figure. That's all the last one should have, if that many.

[/ QUOTE ]

I don't understand what you mean here.

[ QUOTE ]
It's not just this post, but there are an awful lot of precise looking numbers in these math posts when they often depend on what range of hands you think your opponant has.

[/ QUOTE ]

You could very easily come up with several ranges of hands to give you a EV range. This would give you a best case / worst case scenario. This is typically how I set up my Push EV charts.

[willie24]

Yes it is quite the nitpick.

we all know that he didn't mean that 366 is exactly right and 337 and 398 are wrong.


He's not saying that 366 is the answer. he's saying 366 is the answer if you assume that 73 will beat EXACTLY 33% of the hands BB will push with. why would you have a problem with that?

[Scuba Chuck]

Party Poker No-Limit Hold'em Tourney, Big Blind is t400 (4 handed) converter

BB (Villain) (t1830)
UTG (t2596)
Button (t1820)
Hero (t1754)

Preflop: Hero is SB with 3 /images/graemlins/heart.gif , 7 /images/graemlins/heart.gif .
UTG folds, Button folds, Will you try to blind steal with this crap? If so, do you push, or miniraise?


Okay, let’s try this again:

37 of hearts is such a dominant hand “hot and cold”, I will give it a 72% chance of NOT winning against any group of hands that might call it.

And I will assume, I will be able to steal this blind 8 out of 10 times.


ICM – I fold
21% of Prize Pool

ICM – I push, steal blinds
25% of Prize Pool

ICM – I push, I lose
0% of Prize Pool

ICM – I push, I win
36% of Prize Pool

Pushing is worth:
(80%*.25) + (20%*72%*0.00) + (20%*28%*.36)
= .20 + 0 + .02
= .22, or 22%

So from this analysis, pushing is a slight favorite. Now, am I doing this correct?

[Scuba Chuck]

Party Poker No-Limit Hold'em Tourney, Big Blind is t400 (4 handed) converter

BB (Villain) (t1830)
UTG (t2596)
Button (t1820)
Hero (t1754)

Preflop: Hero is SB with 3 /images/graemlins/heart.gif , 7 /images/graemlins/heart.gif .
UTG folds, Button folds, Will you try to blind steal with this crap? If so, do you push, or miniraise?


Okay, let’s try this again:

37 of hearts is such a dominant hand “hot and cold”, I will give it a 72% chance of NOT winning against any group of hands that might call it.

And I will assume, I will be able to steal this blind 8 out of 10 times.


ICM – I fold
21% of Prize Pool

ICM – I push, steal blinds
25% of Prize Pool

ICM – I push, I lose
0% of Prize Pool

ICM – I push, I win
36% of Prize Pool

Pushing is worth:
(80%*.25) + (20%*72%*0.00) + (20%*28%*.36)
= .20 + 0 + .02
= .22, or 22%

So from this analysis, pushing is a slight favorite. Now, am I doing this correct?

[microbet]

It is a nitpick for sure, but it is not whether it is 360 or 380 or 420, and it was not just about this example.

The answer is only as accurate as the assumptions, and they aren't very accurate. If you put in some range of calling ranges and come up with chip EVs of 250-500 then it is a bit misleading to say 366.

As applied to this particular problem, you might simply say the 20% has one significant figure so you only report the answer with one significant figure, thus chipEV = +400, but it would be better to say range of ranges is something like top 15% to top 25% and then calculate the range of chipEVs.

No offense meant to El Maximo. I've just seen many posts where the calculations look more accurate than they really are.

[binions]

I think you will find that pushing with any 2 cards is +EV when folded to in the SB and you or the BB both have stacks of 10-13xBB or less. An exception can be made with complete rags against someone who will call your all-in well over half the time.

http://teamfu.freeshell.org/tournament/theorem_blind_stealing.html

[El Maximo]

[ QUOTE ]
The answer is only as accurate as the assumptions, and they aren't very accurate.

[/ QUOTE ]

Actually, the assumptions of hand ranges can be fairly accurate by doing some research and monitoring the players at the table.

[microbet]

Ok, well, just pointing out that while some calcs like pot-odd s or draw odds are very presice, some where estimating range of hands or implied odds are estimations.

I didn't do the calcs, but in this particular case I don't think it matters much because 73s performs pretty much the same against a big range as a small range of hands.

Peace Out

[ddubois]

I get (http://ddubois.bounceme.net/poker/OddsBets.xls) +244 EV if he will call with 40% of hands (http://home.earthlink.net/~craighowald/data/matchup2.html), +360 EV if he will call with 25% of hands, and +440 EV if he will call with 15% of hands.

[El Maximo]

I use this method alot. I made a push spreadsheet of every hand from the SB and Button vs several calling hand ranges. This gives me a starting point for making push decisions. I actually use stack size and a combination of VP$IP and PFR% to get a pretty good approximation of what hands they will call with. Now I look at my spreadsheet and I know the EV range of my push. I can weight this with the other factors and make a decision.

[willie24]

think of 20.0/366 as a point on the graph of chip EV vs % of hands he will call with.

obviously we don't know what the variable % is

[Scuba Chuck]

Can you elaborate more about your push spreadsheet? You have said "Push Spreadsheet" enough times that I think you're cool talking about it.

[El Maximo]

I just ran push calcs for each hand from 5bb-10bb vs several ranges of calling hands. Simple to do with poker stove and a calculator.