[free software] new version excel worksheet for buble simulation

[crunchh]

Hi there following to my recents post I upgraded a little bit the excel file so that it now stores your history for analysis pushing the new game button is now optional

The first post can be found HERE for reference

the new version can be found
download it HERE

As in the previous post i repost the installation guidelines


What it does :

ICM calculator is an ICM calculator i wrtote to try to evaluate my different play

Bubble trainer is a small excell file which allows me to train with different hands wether to shove or not while 4 handed

installation
what i would recomand

i) Download from HERE

2) unzip in a folder


3) first i would suggest opening the file called bubble trainer
- excel should ask you to allow the macro . it is important to allow macro for program to work

(note : if you know a little bit VBA what youcan do is go in the macro directory there is only 3 display macro lest than 10 line of code , that you can check manually before allowing any macro to run)

4) now you should be positioned on a worksheet called move in trainer, if it is not the case this is the page ou want to move to

5) on this pasge there a many information you should be able to find detais on each command on the screenshot provided in the zip file

6) you press either push or fold

7) the result show

8) press ok


all your moves are now stored in the history sheet for you to be able to analyse it

hope you will like it, any comments more than welcome


thanks

have a nice day

[darom03]

Tried it a few times and this scenario appeared: Seems wrong to me that a push here is better than a fold when we have a ultra short on the btn.

[AMT]

[ QUOTE ]
Tried it a few times and this scenario appeared: Seems wrong to me that a push here is better than a fold when we have a ultra short on the btn.



[/ QUOTE ]


if big stack folded its a pretty easy shove having other stack covered. otherwise youre right

[Mark1978]

Those numbers make no sense. This is -1.1%.

[crunchh]

I agree that this scenario doesn t make sense,

for the moment what I use is a random generation of staks and calling propabilities, it doesn t take into account strategical things like very short stack taking the blinds etc

I am sorry about that but it seemed quite complicated to do otherwise

remember this is for push or fold only here i would just call for a better chance to end in the money

[crunchh]

the numbers are actually calculated in $
as the ICM model is calculates chances to end up in 1rst 2nd or 3rd place I thought it would be influenced by prize distribution so I made the choice to calculate all in dollar in case someone wants to change prize distribution

maybe it is the wrong aproach

if you have any idea on that plz help
keep in mind that i am relly a semi beginner and that the only thing i can do with some degree of sucess is Excell and certainly not poker even if i am trying hard to do poker well

If you have suggestions or you can give me more detail on how you calculate -1.1% please tell me

by the way it is a situation where the value of a push is still negative just a little bit less than letting the blinds go making it a bit marginal push influenced mostly by very low calling % of all 3 players remaining to act

If someone has the courage to modelise something I would think it could be worthwile to develop a model where you don t use flat probabilities but stochastic processes byt his i mean probabilities influenced by past events

my math are a bit rusty on that side but I can refresh them a bit if someone a bit more abl is willing to work with me

[Mark1978]

How do you get -1.63 for folding? What prize pool size are the $s calculated on?
You have 32.55% equity at the start of the hand. I don't know how you have distributed the blinds in the case of a fold, but if everyone folds to the BB you have 32.63%.

[crunchh]

prize pool size is 90$
I use a 50% 30% 20% distribution
and I don t affect blinds to anyone I just expand the total number of chips out of which ICM is calculated
preflop ICM do not take into account the pot fold ICM total number of chips

maybe this is wrong

[Mark1978]

Here's how I work it out. I'm assuming that if you fold then it is folded to the BB who picks up the blinds (there may be better ways, but this is the 'standard' approach that has been adopted).
I'm not taking into account the possibility of a 3-way pot either (which makes these calculations pretty worthless in this situation with the Button so small).

Preflop stacks and ICM$EV (before posting blinds)

BB 7301 39.5%
SB 2328 27.6%
BN 43 0.6%
CO 3829 32.3% (Hero)

Hero is dealt A[img]/images/graemlins/heart.gif[/img]8[img]/images/graemlins/heart.gif[/img]

Hero folds:

BB 7401 39.7%
SB 2228 27.3%
BN 43 0.6%
CO 3829 32.4% (Hero)

(Hero actually gains a tiny bit of EV at the cost of the SB.)

Hero Pushes, no callers

BB 7101 39.1%
SB 2228 27.2%
BN 43 0.6%
CO 4129 33.1% (Hero)

Hero Pushes, BN calls

Hero wins

BB 7101 39.2%
SB 2228 27.5%
CO 4172 33.3% (Hero)

Hero loses
BB 7101 39.0%
SB 2228 26.3%
BN 172 2.4%
CO 4000 32.4% (Hero)

A8s has 34% pot equity against 55+,A8s+,ATo+ so you expect to win 34%:
34% * 33.3% = 11.3%
So you lose 66% of the time:
66% * 32.4% = 21.4%
The total EV of being called by the button is:
11.3% + 21.4% = 32.7%

Hero Pushes, SB calls

Hero wins

BB 7101 40.5%
BN 43 20.2%
CO 6357 39.4% (Hero)

Hero loses
BB 7101 39.4%
SB 4856 35.2%
BN 43 0.8%
CO 1501 24.7% (Hero)

34% * 39.4% = 13.4%
66% * 24.7% = 16.3%
11.3% + 21.4% = 29.7%

Hero Pushes, BB calls

Hero wins

BB 3472 31.6%
SB 2228 27.4%
BN 43 0.6%
CO 7758 40.4% (Hero)

Hero loses
BB 11130 46.6%
SB 2228 33.1%
BN 43 20.3%
CO 0 0% (Hero)

34% * 40.4% = 13.7%
66% * 0% = 0%
11.3% + 21.4% = 13.7%


So now we need to assign probabilities to each outcome.

Hero Pushes, BN calls

Button is calling with top 10% of hands (55+,A8s+,ATo+ by KS rank).
With you holding some of the cards that make those hands though, he will only hold one of those hands 8.8% of the time.

So Button calls 8.8% of the time.

Hero Pushes, SB calls

Since we are only calculating for one caller, for SB to call we need button to fold (100% - 8.8% = 91.2% chance).
91.2% * 8.8% = 8%

Hero Pushes, BB calls

91.2% ^ 2 * 8.8% = 7.3%

Hero Pushes, no callers

91.2% ^ 3 = 75.8%


Multiply probability by the value of each outcome

Hero Pushes, BN calls

8.8% * 32.7% = 2.9%

Hero Pushes, SB calls

8% * 29.7% = 2.4%

Hero Pushes, BB calls

7.3% * 13.7% = 1%

Hero Pushes, no callers

75.8% * 33.1% = 25.1%


Add up the values of all the possible outcomes

2.9 + 2.4 + 1 + 25.1 = 31.4

The EV of pushing is 31.4%
The EV of folding is 32.4%

Pushing is -1% compared to folding.

[crunchh]

ok now I see why we have slightly different numbers

there is 2 things wrong in my sheet thanks for pointing them out

first when one stak is less than BB I have the impression it is not propely taken into account (minor mistake since when someone is all in if only taking the blinds i am less likely to push

Second mistake more trouble some and that i have to fix is that when calling range is 10% which is 10% of hands I am assuming he gets those hands 10% of the time

and the 10% range i sused is the one of pokerstove which slightly differs from yours :
88+,A9s+,KTs+,QTs+,AJo+,KQo
which gives 39% chances of win (poker stove)

other wise i have very close numbers

thanks for the comment I will look into that a bit deeper now to see how to adapt the worksheet