Two Plus Two Publishing LLC
Two Plus Two Publishing LLC
 

Go Back   Two Plus Two Poker Forums > >

Notices

Commercial Software Discussion of commercial gambling-related / poker software & commercial graphics modifications

Reply
 
Thread Tools Display Modes
Old 02-21-2019, 04:40 AM   #2176
NoSurprises
old hand
 
NoSurprises's Avatar
 
Join Date: Feb 2008
Posts: 1,743
Re: ICMIZER 2

Hi,

Just purchased pro version this week for the first time and have been on SNG Coach.

I've just done a few sets of questions and it is giving me incorrect results. For example, it tells me that KQcc, 55 & 44 are all pushes. I chose pushes but it says 0/4 . Is it always this buggy?
NoSurprises is offline   Reply With Quote
Old 02-21-2019, 01:12 PM   #2177
Q
Pooh-Bah
 
Q's Avatar
 
Join Date: Sep 2007
Posts: 4,525
Re: ICMIZER 2

Quote:
Originally Posted by NoSurprises View Post
Hi,

Just purchased pro version this week for the first time and have been on SNG Coach.

I've just done a few sets of questions and it is giving me incorrect results. For example, it tells me that KQcc, 55 & 44 are all pushes. I chose pushes but it says 0/4 . Is it always this buggy?
Hi.
No, it should be working just fine. Everyday SNG Coach handles thousands of questions and I am not receiving such bug reports at all.
Please make sure that your internet connection is fine. Sometimes this may happen due to connection issue, the questions can timeout on server and in that case all answers are treated as incorrect.
If this happens several times per day for you please contact me at support@icmpoker.com and we will investigate your account issues.
Q is offline   Reply With Quote
Old 02-23-2019, 08:43 PM   #2178
hainberg25
newbie
 
Join Date: Jul 2018
Posts: 16
Re: ICMIZER 2

Hi I have a few questions about how icmizer calculates nash raise ranges.

1)does icmizer create raise ranges(less then all in) assuming all villians will push fold.
2)if icmizer takes calling into account, how does it split villians ranges between calling and raising
3)Does icmizer take equity realization of calling into account

some minor questions.

i)Is there anyway to see the ev of hands in villians range.
ii)is there anyway to integrate the ev of our entire pushing range(instead of seeing the ev of individual hands)
hainberg25 is offline   Reply With Quote
Old 02-24-2019, 10:56 AM   #2179
Q
Pooh-Bah
 
Q's Avatar
 
Join Date: Sep 2007
Posts: 4,525
Re: ICMIZER 2

Quote:
Originally Posted by hainberg25 View Post
Hi I have a few questions about how icmizer calculates nash raise ranges.

1)does icmizer create raise ranges(less then all in) assuming all villians will push fold.
2)if icmizer takes calling into account, how does it split villians ranges between calling and raising
3)Does icmizer take equity realization of calling into account

some minor questions.

i)Is there anyway to see the ev of hands in villians range.
ii)is there anyway to integrate the ev of our entire pushing range(instead of seeing the ev of individual hands)
Hi hainberg25

1) Yes. If there is 1 raiser, then ICMIZER finds optimal opening range assuming other players will push/fold against this raiser. If there are 2 raisers (raise + reraise) ICMIZER finds optimal ranges for these 2 raisers assuming other players will only push or fold.
2) Generally, when finding ranges everyone except for players who already made a VPIP action either push or folds. If there are a raise and call of the raise, the calling is taken into account.
3) Yes, the strategy I believe is as follows, since we cannot predict postflop equity the postflop pot is split among callers according to their range strength.

i) Not sure what you mean here, the calculation can be performed for any player so "villain" is a relative term. I believe the answer is yes
ii) No. You can find the EV Flow feature useful but it doesn't do exactly that.
Q is offline   Reply With Quote
Old 02-24-2019, 03:29 PM   #2180
hainberg25
newbie
 
Join Date: Jul 2018
Posts: 16
Re: ICMIZER 2

I was wondering how ICMizer generates the initial raising ranges. Like say we're 20bb deep and button minraises and hero is in the sb and we click the calculate nash equilibrium. We get this result:



If we then click on the button's opening range, it looks like this.



I couldn't figure out why you would open hands like 54s if all villians can do is jam or fold. Seems like you would just open hands with good blockers that you were gonna raise fold.


I think what icmizer does is pick the hands that have the most equity vs a jamming range as its opening range. So the EV of calling a sb jam for every hand in btn's opening range looks like this.




So 54s has more equity then q8o so its included in the opening range. And the btn opens 46% because given that sb and bb jam 59% of the time, the raise folds are break even.

Is that how it works?

Last edited by hainberg25; 02-24-2019 at 03:34 PM. Reason: added pictures
hainberg25 is offline   Reply With Quote
Old 02-28-2019, 05:51 AM   #2181
KiNgLeGa
journeyman
 
Join Date: Jul 2008
Posts: 395
Re: ICMIZER 2

Is it possible to calculate early game PKO MTT hands using your software?
KiNgLeGa is offline   Reply With Quote
Old 02-28-2019, 10:52 AM   #2182
Q
Pooh-Bah
 
Q's Avatar
 
Join Date: Sep 2007
Posts: 4,525
Re: ICMIZER 2

Quote:
Originally Posted by KiNgLeGa View Post
Is it possible to calculate early game PKO MTT hands using your software?
Hi KiNgLeGa
Yes, but only in chip EV mode.
ICM calculations are available for PKO FTs only.
Q is offline   Reply With Quote
Old 02-28-2019, 10:53 AM   #2183
Q
Pooh-Bah
 
Q's Avatar
 
Join Date: Sep 2007
Posts: 4,525
Re: ICMIZER 2

Quote:
Originally Posted by hainberg25 View Post
I was wondering how ICMizer generates the initial raising ranges. Like say we're 20bb deep and button minraises and hero is in the sb and we click the calculate nash equilibrium. We get this result:



If we then click on the button's opening range, it looks like this.



I couldn't figure out why you would open hands like 54s if all villians can do is jam or fold. Seems like you would just open hands with good blockers that you were gonna raise fold.


I think what icmizer does is pick the hands that have the most equity vs a jamming range as its opening range. So the EV of calling a sb jam for every hand in btn's opening range looks like this.




So 54s has more equity then q8o so its included in the opening range. And the btn opens 46% because given that sb and bb jam 59% of the time, the raise folds are break even.

Is that how it works?
Hi

I really love your question.
Please reach me at support@icmpoker.com I got a gift for you for it
Note that for situations with raisers ICMIZER 2 supports Beta Nash calculations.
It will be significantly improved in ICMIZER 3. The ranges will make more sense and the approximations will have less room for exploitability.
Q is offline   Reply With Quote
Old 02-28-2019, 12:57 PM   #2184
balls'n'all
journeyman
 
Join Date: Jan 2009
Posts: 249
Re: ICMIZER 2

Any updates as to when icmizer3 will be released ?
balls'n'all is offline   Reply With Quote
Old 03-01-2019, 07:21 PM   #2185
IGotMilkBaby
newbie
 
Join Date: Dec 2013
Posts: 47
Re: ICMIZER 2

Hello. My question nr.1 is how the RFI frequencies in ICMizer are established? Is it the same principle as in HRC where flatting behind (unless given the option) is not considered as well as 100% of equity realisation thus making the RFI % from positions higher than they realistically should be also raising more blocker type hands like off suit aces and kings due to the raw equity from EP when in reality the composition of our opening range should be different?

2) Is card removal considered in ICMizer and do you know if it is in HRC (I know you don't answer for HRC but I can't find their thread on 2p2 and maybe you have the information)?

Would really appreciate answers to these questions
IGotMilkBaby is offline   Reply With Quote
Old 03-04-2019, 07:48 PM   #2186
Q
Pooh-Bah
 
Q's Avatar
 
Join Date: Sep 2007
Posts: 4,525
Re: ICMIZER 2

Quote:
Originally Posted by balls'n'all View Post
Any updates as to when icmizer3 will be released ?
Hi
ICMIZER 3 is taking longer than expected.
Can't give a date right now, we got a few complex features to finish.
I plan to make bigger announcements later this March.
Stay tuned for the updates
Q is offline   Reply With Quote
Old 03-04-2019, 07:55 PM   #2187
Q
Pooh-Bah
 
Q's Avatar
 
Join Date: Sep 2007
Posts: 4,525
Re: ICMIZER 2

Quote:
Originally Posted by IGotMilkBaby View Post
Hello. My question nr.1 is how the RFI frequencies in ICMizer are established? Is it the same principle as in HRC where flatting behind (unless given the option) is not considered as well as 100% of equity realisation thus making the RFI % from positions higher than they realistically should be also raising more blocker type hands like off suit aces and kings due to the raw equity from EP when in reality the composition of our opening range should be different?

2) Is card removal considered in ICMizer and do you know if it is in HRC (I know you don't answer for HRC but I can't find their thread on 2p2 and maybe you have the information)?

Would really appreciate answers to these questions
Hi. I assume RFI stands for raise first in.
This is an ICMIZER support thread, and pretty much all questions can be asked without mentioning other software.
1) Flatting is not considered after a raise, other players either push or fold.
ICMIZER is solving a certain poker subgame, namely where one player: raises X first, then other players either push or fold. So the ranges are optimal for this kind of game, they aren't optimal to full poker game with flats and postflop, yes.
2) Sure, yes, card removal is considered by ICMIZER. It takes into account cards of active players (not folders). You can use "XX detailed result" button for any starting hand and check how it affects probabilities of other opponents ranges.
Q is offline   Reply With Quote
Old 03-11-2019, 04:45 AM   #2188
NoSurprises
old hand
 
NoSurprises's Avatar
 
Join Date: Feb 2008
Posts: 1,743
Re: ICMIZER 2

Hi,

I'm building push/fold charts for 6max hypers using ICMIZER2 but am getting some strange results depending on the process I use.

So firstly, I used the quiz function with the inbuilt Pokerstars 6max Sat quiz to build my charts. I then uploaded a hand history , calculated FGS3(4) and get totally different results.

So this is the order I have done it this morning.

First I upload the hand history and analyze the first hand. It has picked up that it is a pokerstars 6max hyper sat but I went in and double clicked it to make sure.

https://ibb.co/jfFbv5M

So it tells me to jam 100% from cut off.

However, the quiz answers when ICMIZED tell me this:

https://ibb.co/3kB0Vqm

So it is telling me to push 41%, a huge difference but what my chart was built around.

I then double click on hand 1 - 93o - to analyse again, and this time it gives me the same result as the quiz:

https://ibb.co/x1VwSX2


So what am I doing wrong and why is it giving me a 100% jam range the first time I analyse it?
NoSurprises is offline   Reply With Quote
Old 03-11-2019, 05:32 PM   #2189
robwa
grinder
 
Join Date: Nov 2010
Posts: 483
Re: ICMIZER 2

Hello all,

I am happy with ICMizer 2 and I've been using it quite frequently to analyse my push/fold decisions. Now the other day I heard from a fellow poker player that ICMizer can also be used to analyse flatcalling spots. I tried Googling for this, to see how this is possible but no luck so far.

Is there any documentation / How to on this available?
And if so, would that be flatcalling with the assumption of just seeing the flop, or assuming a 100% equity realization?

Happy if someone can guide me in the right direction, no luck thus far after ~ googling and pressing buttons in the software

Thanks in advance
robwa is offline   Reply With Quote
Old 03-14-2019, 02:15 PM   #2190
JustForReads
newbie
 
Join Date: Feb 2018
Posts: 22
Re: ICMIZER 2

ICMIZER can not really be used for what that person suggested.
JustForReads is offline   Reply With Quote
Old 03-15-2019, 02:51 PM   #2191
Q
Pooh-Bah
 
Q's Avatar
 
Join Date: Sep 2007
Posts: 4,525
Re: ICMIZER 2

Quote:
Originally Posted by NoSurprises View Post
Hi,

I'm building push/fold charts for 6max hypers using ICMIZER2 but am getting some strange results depending on the process I use.

So firstly, I used the quiz function with the inbuilt Pokerstars 6max Sat quiz to build my charts. I then uploaded a hand history , calculated FGS3(4) and get totally different results.

So this is the order I have done it this morning.

First I upload the hand history and analyze the first hand. It has picked up that it is a pokerstars 6max hyper sat but I went in and double clicked it to make sure.

https://ibb.co/jfFbv5M

So it tells me to jam 100% from cut off.

However, the quiz answers when ICMIZED tell me this:

https://ibb.co/3kB0Vqm

So it is telling me to push 41%, a huge difference but what my chart was built around.

I then double click on hand 1 - 93o - to analyse again, and this time it gives me the same result as the quiz:

https://ibb.co/x1VwSX2


So what am I doing wrong and why is it giving me a 100% jam range the first time I analyse it?
Hi NoSurprises.
You need to be using calculate Nash equilibrium button.
This screenshot demonstrates the problem:
https://i.ibb.co/g9CmM7T/Co-push-Hand-history.jpg

The first column has callers ranges at default 8%, this causes the pushing range to change. If you Calculate Nash equilibrium, 8% ranges will get replaced with optimal ranges, and the result will become consistent with what you expect.
Q is offline   Reply With Quote
Old 03-15-2019, 02:56 PM   #2192
Q
Pooh-Bah
 
Q's Avatar
 
Join Date: Sep 2007
Posts: 4,525
Re: ICMIZER 2

Quote:
Originally Posted by robwa View Post
Hello all,

I am happy with ICMizer 2 and I've been using it quite frequently to analyse my push/fold decisions. Now the other day I heard from a fellow poker player that ICMizer can also be used to analyse flatcalling spots. I tried Googling for this, to see how this is possible but no luck so far.

Is there any documentation / How to on this available?
And if so, would that be flatcalling with the assumption of just seeing the flop, or assuming a 100% equity realization?

Happy if someone can guide me in the right direction, no luck thus far after ~ googling and pressing buttons in the software

Thanks in advance
Hi, robwa.
Well you can review squeeze spots like this one: https://www.icmpoker.com/screenshot/pFkuRP/
However, Nash ranges in situations like this are calculated in Beta mode (can be far from optimal) and postflop equity realization follows a simple approach.
If you want to use ICMIZER and find optimal flat calling ranges, it is not going to be a good idea because ICMIZER can compare between 2 decisions (push/fold or bet/fold), and flat call is rarely the best alternative to fold if we choose to limit our actions to just 2 options.
Q is offline   Reply With Quote
Old 03-25-2019, 05:08 PM   #2193
Q
Pooh-Bah
 
Q's Avatar
 
Join Date: Sep 2007
Posts: 4,525
Re: ICMIZER 2

ICMIZER 2 Update 2.16.0

● Significantly improved performance of ICMIZER with a long list of loaded hands both in Analysis and Histories views
(and the switching speed between the two for Nitro users).
● Improved the speed of switching between loaded hand histories
● Fixed some problems with the Load hands dialog.
Q is offline   Reply With Quote
Old 03-26-2019, 04:59 PM   #2194
SageLee
journeyman
 
SageLee's Avatar
 
Join Date: May 2005
Posts: 200
Re: ICMIZER 2

HI, had a quick question about a certain situation (see image).



So I was checking out the EV flow in this hand, and saw something that I found peculiar. It says that the CO, who will be in the BB next hand with less than 2.5 BBs, gains EV the next hand, while the SB, who will be on the button next hand, will lose about that same amount of EV. I'm trusting that this isn't a bug or anything, but I'm uncertain as to how this could be the case. Could you explain how the extremely short stack in a 4-handed situation is going to gain EV next hand by going into the BB (and similarly for the SB, who next hand will be in the most favorable position farthest from the blinds, and yet loses EV)? Is there something favorable about being forced to (likely) commit such a tiny stack to a random hand such that you gain EV in this spot? Or does it have something to do with how likely some stacks might be to fold very bad hands to you, that they might otherwise push, because they know you have little to lose and prob have to call? Etc?

Would love to understand the math better on these EV shifts here. Thanks for any input you (or anyone else) might have.
SageLee is online now   Reply With Quote
Old 03-26-2019, 08:48 PM   #2195
Q
Pooh-Bah
 
Q's Avatar
 
Join Date: Sep 2007
Posts: 4,525
Re: ICMIZER 2

Quote:
Originally Posted by SageLee View Post
HI, had a quick question about a certain situation (see image).



So I was checking out the EV flow in this hand, and saw something that I found peculiar. It says that the CO, who will be in the BB next hand with less than 2.5 BBs, gains EV the next hand, while the SB, who will be on the button next hand, will lose about that same amount of EV. I'm trusting that this isn't a bug or anything, but I'm uncertain as to how this could be the case. Could you explain how the extremely short stack in a 4-handed situation is going to gain EV next hand by going into the BB (and similarly for the SB, who next hand will be in the most favorable position farthest from the blinds, and yet loses EV)? Is there something favorable about being forced to (likely) commit such a tiny stack to a random hand such that you gain EV in this spot? Or does it have something to do with how likely some stacks might be to fold very bad hands to you, that they might otherwise push, because they know you have little to lose and prob have to call? Etc?

Would love to understand the math better on these EV shifts here. Thanks for any input you (or anyone else) might have.
Hi SageLee.
Thanks for the question, can I ask first what kind of browser or setting you are using, which causes everything to be so ugly? All fonts are screwed up, its not supposed to be like that..

Regarding your question - first of all EV Flow is a rather advanced feature so it is naturally hard to understand.
One thing to notice is that the difference between hands (current and next) really doesn't mean too much, however, it can indicate that the currently selected ICM model is pretty flawed for this particular situation. A perfect model & Nash ranges shouldn't change the value of stacks by much in just 1 hand.

However, getting closer to actual reasons. You are seeing how EV of CO changes according to the selected model after this hand. The + side is explained by the non zero probability of big stack busting some other stack, not the CO. If there is a smaller probability of bubble burst this hand, our +EV estimation according to FGS3 model, which you're using in this case, goes down.

https://www.icmpoker.com/screenshot/NTdypz/

With FGS3 we're looking at 4 hands and in them, CO goes BB next hand, then he can get to SB position and hand after that he can get to BTN. All these additional hands improve his forecast for this tournament.

Check what FGS1 sees and forecasts for CO. FGS1 looks at current and just 1 hand ahead, where CO becomes BB. Looks pretty bad for him, right?

https://www.icmpoker.com/screenshot/Oceisn/

FGS2 model sees how he becomes SB and improves his equity by a lot:

https://www.icmpoker.com/screenshot/OleSvb/

Hope this makes some sense
Q is offline   Reply With Quote
Old 03-28-2019, 02:08 AM   #2196
Eudaimonia
journeyman
 
Eudaimonia's Avatar
 
Join Date: Jul 2017
Location: Spirit Rei
Posts: 234
Re: ICMIZER 2

Hey is this software able to tell me push/rif/fold strategies preflop like in the middle stages of a tourney and for spots say someone is opening light UTG how wide we can raise/jam like nodelocking for UTG villain's range
Eudaimonia is offline   Reply With Quote
Old 03-29-2019, 07:56 PM   #2197
Kanji
stranger
 
Join Date: Jul 2016
Posts: 4
Re: ICMIZER 2

Hi!

I just bought IcMizer Pro Version with add Nitro but when I Load my hands I don't see the Analisys... :/



Can You help me please? : ) Thanks ! : )
Kanji is offline   Reply With Quote
Old 04-03-2019, 10:11 AM   #2198
Q
Pooh-Bah
 
Q's Avatar
 
Join Date: Sep 2007
Posts: 4,525
Re: ICMIZER 2

Quote:
Originally Posted by Eudaimonia View Post
Hey is this software able to tell me push/rif/fold strategies preflop like in the middle stages of a tourney and for spots say someone is opening light UTG how wide we can raise/jam like nodelocking for UTG villain's range
Hi.
Generally, I believe the answer is yes.
How deep are you in these middle stages?
Regarding "node locking" - in ICMIZER 2 you can manually lock all opponent strategies and then calculate an optimal hero push/fold table against them.
In ICMIZER 3 we will allow Nash calculations with node locking.
Q is offline   Reply With Quote
Old 04-03-2019, 10:15 AM   #2199
Q
Pooh-Bah
 
Q's Avatar
 
Join Date: Sep 2007
Posts: 4,525
Re: ICMIZER 2

Quote:
Originally Posted by Kanji View Post
Hi!

I just bought IcMizer Pro Version with add Nitro but when I Load my hands I don't see the Analisys... :/



Can You help me please? : ) Thanks ! : )
Hi.
For hands to be "icmizeable" the situations have to be in the certain phase of the tournament - closer to end game and push/fold.
In the screenshot, you are quite deep, and these hands do not get analyzed in pure preflop push/fold terms. Post-flop is very likely to happen when stacks are deeper than 25 blinds.

Please refer to this article to learn more:
https://icmizer.zendesk.com/hc/en-us...lysis-feature-
Q is offline   Reply With Quote
Old 04-04-2019, 10:00 PM   #2200
SageLee
journeyman
 
SageLee's Avatar
 
Join Date: May 2005
Posts: 200
Re: ICMIZER 2

Quote:
Originally Posted by Q View Post
Hi SageLee.
Thanks for the question, can I ask first what kind of browser or setting you are using, which causes everything to be so ugly? All fonts are screwed up, its not supposed to be like that..
I'm using Firefox with Linux Mint 18.3 with KDE. As far as I know I have not altered any fonts or settings.

Quote:
Originally Posted by Q View Post
Regarding your question - first of all EV Flow is a rather advanced feature so it is naturally hard to understand.
One thing to notice is that the difference between hands (current and next) really doesn't mean too much, however, it can indicate that the currently selected ICM model is pretty flawed for this particular situation. A perfect model & Nash ranges shouldn't change the value of stacks by much in just 1 hand.

However, getting closer to actual reasons. You are seeing how EV of CO changes according to the selected model after this hand. The + side is explained by the non zero probability of big stack busting some other stack, not the CO. If there is a smaller probability of bubble burst this hand, our +EV estimation according to FGS3 model, which you're using in this case, goes down.

https://www.icmpoker.com/screenshot/NTdypz/

With FGS3 we're looking at 4 hands and in them, CO goes BB next hand, then he can get to SB position and hand after that he can get to BTN. All these additional hands improve his forecast for this tournament.

Check what FGS1 sees and forecasts for CO. FGS1 looks at current and just 1 hand ahead, where CO becomes BB. Looks pretty bad for him, right?

https://www.icmpoker.com/screenshot/Oceisn/

FGS2 model sees how he becomes SB and improves his equity by a lot:

https://www.icmpoker.com/screenshot/OleSvb/

Hope this makes some sense
Just to make sure I follow, you were detailing reasons the model might do this, but the bottom line is that the model simply may not be accurate in a more extreme case like this?

It just seems unlikely that the EV lost from losing more than half stack in the BB/SB would be made up by better future (and very temporary, considering it's 4-handed) position such that you would actually gain even more EV back over those 2-3 hands. I just want to make sure I correctly understand this to be true and am not being swamboozled by some non-obvious math in a highly-nuanced situation. I wasn't sure how much the model takes into account; for instance is it assuming you fold in those hands, or is it taking into account the chances you are dealt a big hand/terrible hand + taking into account equity % when it's good enough to move in, or the chances that other players in the intervening hands might get dealt big enough hands to probably clash and raise your EV by someone getting knocked out, etc.

Last edited by SageLee; 04-04-2019 at 10:11 PM.
SageLee is online now   Reply With Quote

Reply
      

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off


Forum Jump


All times are GMT -4. The time now is 12:09 AM.


Powered by vBulletin®
Copyright ©2000 - 2019, Jelsoft Enterprises Ltd.
Copyright © 2008-2017, Two Plus Two Interactive
 
 
Poker Players - Streaming Live Online