If you bet 20% of the time for an EV of 30 and you check 80% of the time for an EV of 10, then the EV for the entire decision will be 20%*30+80%*10=14. This is the weighed average of the two decisions.

If I understand you correctly then these are two totally separate lines, so this would not be correct. If you want a more detailed answer, then please send a savefile to support with a specific question about how a certain number is calculated.

really nice tool man and great support!
couple of questions/suggestions:

1) say i use the equilibrium solver to solve a flop situation. i build the tree and make the whole thing start at the flop. i give both players ranges. i enter a flop. i run the equilibrium solver. lets say its btn vs bb and i want to examine specifically a turn situation after btn checked back flop. i enter a turn card. if i understand it correctly i have to now run the EV tool to get an overview of the percentages shown in the tree. whenever i want to look at a different turn card, i have to use the EV tool again to get the tree to represent the right percentages for each action taken. since the EV tool is instant, would it be possible to add an option that automatically runs the tool when i change the turn card?

2) say i build a pretty large tree for btn vs bb (in this case: oop [sb] checks flop 100%, after that every player on every street has the option to bet 2/3 pot or check and to raise call or fold). is there any way to use that tree as kind of an template for a different scenario? for example co vs btn? i mean i guess i could just enter different hand ranges and act as if SB was CO. but it would be easier if i could just change the label of SB to CO. is this possible already or could it be added? i know there is a tree building wizard but from the limited time i spend playing around with it, it would still take me quite some time to rebuild a large tree like i described above from scratch.

thanks!

Thank you. Tho i do have a problem...

I can't get to the accept changes button on crev it won't let me scroll down and none of it will save.

I'll see if I can add something along these lines in the future.

The wizard should build a tree for you with a single action. So rebuilding a tree for different player names with the wizard should not be particularly difficult, unless you want to make all kinds of custimizations. That being said, I can consider an option to change the names, but if you just change the ranges to whatever you want then that should in itself be sufficient.

Do you by any chance use a very low screen resolution?
Or can you possibly just increase the window size?

Okay speak of the devil as soon as you said this it started working, but SOMETIMES/MOST OF THE TIME the scroll bar won't let me scroll down and i am already @ max window size.

Maybe it's screen resolution but i just have an issue with scroll bar or the whole thing table matrix isn't showing fully like i can't seem the bottom part where the accept changes button is think it's a really weird glitch.

Think i just got lucky this time tho

---

I'll post a image here of what i'm talking about gimme a sec

http://imgur.com/a/DhKik

Like thats as far down as my scroll bar will go, it won't go past the equity distribution graph... :l

it is a problem on laptops with a screen resolution of 1366 * 768

Ok, a x768 vertical resolution is indeed smaller than what was assumed to be the smallest resolution used on any device nowadays. I can see if in these kinds of cases I can move some buttons around or create a shortcut for accepting the input.

Thank you for the feedback,

Scylla

Please move some buttons, i can't affrod another $500 dollar laptop haha

Any plans to introduce aggregate reports, similar to PIO, and scripting functionality for the Equilibrium solver?

Hi Scylla,

Will the equilibrium solver ever be scriptable similar to PIO Pro?

I can look into it, however, in order to help me get an understanding of how such a feature should be designed, can you please give me an idea of what sort of things you'd like to accomplish with it?

It would be nice to be able to aggregate a lot of different board textures into one equilibrium %. Be able to run the solver on X different Ahi flops for a given line to see the different %s, see where the differences are and what they converge towards. Then repeat with either new board texture or line.

If the solver was scriptable right now you could input a line you want to study, solve, do new random board, solve & repeat until you have a significant amount of different board textures to study the output from.

Being able to get the solver output condensed/summarized from a range of board textures would be quite nice instead of having to manually filter through each board situation one by one.

Thanks for considering!

Set flop
Solve to X accuracy
Change bet size
Store variable Y
Graph

Set flop
Solve to X accuracy
Change top % of hands
Store variable Y
Graph

Those are a couple example scripts of what you can do. All you really need to do is add the solver as an option in the script menu just like the max exploit button. I think.... You can create a game tree change a variable, then find a nash solution and store that variable then graph or whatever. To learn how changing a game trees variable impacts the solution. If you wanted to graph how the EV changes as you increase your flop c-bet size. Or another example will be you set up a game tree and change the top X % of hands someone plays and then solve and then see how your defense changes or something. Its endless.

The aggregated reports thing will be tricky to implement. If you want to study a weighted subset, you have to store the weight of each flop along with the flop, along with the variable you want to study. I am not sure about that.

List of flops with weight w
Set flop 1 on list
Solve to accuracy A
Store a bunch of variables that go with flop 1 on list and weight 1
(Optional: save solution) (possible small save flop only)
Set flop 2 on list
Solve to accuracy A
Store a bunch of variables that go with flop 2 on list and weight 2
(Optional: save solution)
...
Create report find weighted averages

copied from another user:
"sub-optimal ranges e.g. UTG range contains AA, 73o and 72o, after using the gto tool it says, bet w AA 100% of the time and 73o 50% and 72o 50%. I would like to have an answer saying AA 100%, 73o 100% and 72o 0%, so its more practical in the real world."


Would like to choose of something like 25%,50%,75%,100% of a combo.
So I can compare the EV of the range I would play, the sub-optimal range and the optimal solver solution.

Ok, I can look into it for later versions.

Thank you for the feedback,

Scylla

Ok, thank you for the feedback.
I think I should have enough for the moment.

+1 for scripts!

Ideally extended scripting capabilities should allow us to solve for optimal preflop ranges.
After solving particular flop with particular weight a Hand Matrix with EVs of all hands is stored somewhere, then another flop, another Hand Matrix and in the end we get a Hand Matrix with average EVs of all hands.

I tried doing this stuff manually in a spreadsheet but it takes huge amount of time.

That would answer an important question: "In a given preflop spot what hands can we profitably call?"

Hi

Does this software incorporate the range building tools from Flopzilla?

Also, can you import hand histories single/multiple?

Thanks

LET

Ok, thank you for the feedback.
I'll take it into account.

Yes, in fact CardRunnersEV's preflop range editor contains an exact duplicate of Flopzilla's editor. Also, it stores its ranges to the same newdefs2.txt file as Flopzilla. To exchange ranges between the two programs, all you would need to do is copy this file from one program to the other.

As for importing hand histories, such an option is offered in the startup screen. This will only import the single line from the hand history in question though, so you will still need to set ranges (instead of exact hands) and turn it into a decision tree in order to perform analysis. An easy way to do this would be with the equilibrium solver. For this, just create a subtree with the tree building wizard at the decision that you're interested in. After that, run the solver to see how the hand should be played from that point on. For more on the wizard and the solver, please see the videos here: http://www.cardrunnersev.com/download.html

Thanks for the information.

If CEV incorporates FZ then why would I need both? Am I missing something, does purchasing CEV negate the need for FZ?

Also, in your explanation of how I would import a full tournament history, could I use a script/macro/hotkey to sequence through and import said 'single line' as required. Ideally I want to import and review a full tourne quite quickly and then focus on specific spots as opposed to importing singe hands from known specific spots. Does that make sense?

Thanks

LET

Oh, and one more thing: I noticed somewhere that it uses Monte Carlo sims, how long does it take per tree calculation? Is it RAM or CPU intensive?

Thanks

LET

CardRunnersEV can do most things Flopzilla can, however, Flopzilla's interface is optimized for the purposes of range analysis. Generally, if you want to look at ranges, then Flopzilla would be the program to go for.

As for tournaments, what you can do is import single hands and then, for example, use the solver to see how a hand/range should have been played.

It will mostly depend on the number of simulations. More simulations will mean longer calculation times.

The monte carlo engine is toggled to when it's not possible to enumerate everything due to the number of matchups. This will happen if you have a preflop hand with play on an unknown flop. Or if there's a multiway hand. A disadvantage of the monte carlo engine is that all of its results are estimates (because they're based on running a large number of simulations). The main advantage is that it will always work. So it's not as accurate as the math engine, which will give mathematically accurate values, but it's the backup engine in case there's too many matchups.

Ideally you should try to focus on spots where you can use the math engine. These will be heads-up preflop spots or postflop heads-up spots on known flops. In this case it will be possible to enumerate all results and you will have mathematically accurate data. On top of that, in those spots you will also be able to use the GTO solver, which will let the software figure out for you how a hand/range should be played.

No, it's not particularly demanding in those areas.