Can you perhaps send a screenshot to support of the spot that you're referring to? As far as I know, the solver should always stop once reaching the stated dEV.


EV converges much faster than frequencies.
So the EV should remain basically unchanged between the two situations.

If OOP has mulitple actions, then there will also be multiple spots where IP gets to act. So for example, if OOP either bets or checks then IP's actions would need to be displayed separately for each scenario by OOP and for each possible turn. And each data point will have been measured, not only for a different turn, but also a different range and frequency for OOP. This means that this data for IP is essentially not suitable for being displayed in a table/graph. We can decide to ignore this, and plot it anyhow for later releases, but right at this moment it has been left out for this reason.


No, we offer many different display methods, but not this particular one.
This output type is however already available for aggregate reports.


The colors for different actions can already be edited. For this, create a tree and go to the editor. There, click on "Edit action colors" and set whichever colors you like. You may need to click on "IMPORT TREE" first in order to get the "Edit action colors" button to appear.

Ok, I will see what I can do for later releases.

Hello!
Is it possible to
1) have variable #1 - that varies within the defined range
2) have variable #2 that is formula like = (S-#1) / (p + #1)
3) Plot the graph that shows EV(variable#2) by varying variable#1?

I tried comparing ev changes while running the solver for different lengths and found it seems no matter what I do as far as ranges, boards, or time solved, the in position player with a range advantage always has .016 to 0.18 ev. Does this sound reasonable?

The ev seems about the same regardless of time spent solving but also regardless of range or board as long as its an advantage ip.

We can consider this for later releases, but right at this moment it's not there.

.016 or .16?

I can't really say much without knowing more details, but it's entirely possible for a player's EV to be in a very narrow range, as well as a very broad range. I would expect a broad range when the ranges are very asymmetrical. For example, see below a graph for the EVs for all 1755 possible boards for OOP having the top 20% versus IP having the bottom 20%. With such ranges, low connected flops will hit IP's range, while hands with many high cards will hit OOP's range:



On the other hand, there's a graph for "all hands" vs "all hands" for all 1755 boards. In this case each range will hit the board exactly the same, and the EV will be distributed across a very narrow range:

He Scylla, any solution for this?
Cheers

When creating ranges in the GTO+ program for future use is there a way to do so without having a range for every single situation? Ie using the upswing ranges - vs a RFI from UTG - im in the CO, I have 3 options within my range (3!, call, %50 3!/%50 call)

Do I have to create the range for all these different options or is there a way to represent them within one tree?

I'm also finding it difficult to see the following bet sizes within my tree

Starting pot: $60
Eff stact: $1200
1st bet 35
2nd bet 150
3rd bet 310

actions:
P1 bets $35
P2 raise $150
P1 calls

Turn:x
P1 check
P2 check

River: x
P1 check
P2 (only option is to bet $150; how can i show bet $310)

You could consider either using groups or, indeed, defining ranges separately.

I assume that you mean that the only option on the river is 160? With the given properties, after having put in a bet of 150, the pot becomes 60+2*150=360 and the effective stacks become 1200-150=1050. At the flop the 35/150/310 setup meant "Bet 35, Raise to 150, Raise to 310", meaning that the third bet is betting an additional 160 into a pot of 360. Or, in other words, a raise of 44% of the pot. The situation that you describe on the river will be the same. The tree builder will maintain betting 160 into the pot of 360, just like it would have been on the flop.

Hey, Skylla! Can you suggest powerful, easy to use and not expensive server service for simulations?
I want to make 5 sizings on each street, 200 flops for several ranges - to see the prefered sizings and lines in different situations. It seems it will work forever on my notebook.

You could try Amazon Web Services: https://aws.amazon.com/

That being said, using 5 sizings on every street will result in a huge tree. Even moreso, contrary to popular believe, this will not teach you anything about preferred sizes. In fact, there isn't even such a thing as a preferred size. In a GTO solution, any size will work roughly as well as any other size. And even using multiple sizes will not result in much better performance.

You can try this for yourself by creating two trees for different sizes (same board, same ranges, etc) and then checking the overall EV performance. The overall EV performance can be found by looking at OOP's EV below the table in his very first decision. See the screenshot below. As you will notice, this overall EV will be almost identical, regardless of which sizes are used.

Scylla?

Update on the weighted flop subsets that were mentioned itt a couple weeks ago?

How is there no such thing as a preferred size? For example:

I run a SIM SBvsBB on an 8d7c3d board:

I use 33% cbet sizing
I use 150% cbet sizing

As you can see here it cbets the 150% 25.1 to 2.4 or over 10x as often, would that not lead you to believe it is preferred?



The EV may be close to identical in theory but if our opponent knows how to react to 1/3 sizings and doesn't know how to react to 6/4 sizings then they will make more mistakes vs the overbets

The overall EV in the pic you showed is 2.85.

Please try creating two separate trees.
One for a bet size of 33% and one for a bet size of 150%.
In the tree with 33% leave out the bet of 150%.
And vice versa, in the tree with 150% leave out the bet with 33%.

Now compare the overall EV for both trees.
The tree with the highest overall EV will be the "preferred" one.
The difference, as it will most likely turn out, will be very small.
It's possible to construct a credible GTO strategy for any bet size that will perform similarly to any other bet size.

Now, although it may be argued that the tiny difference would still mean that a certain size is preferred, the problem is that a lot of detail is lost when translating the GTO solution to something that can be used in practical play. When deciding on sizes, it's better to just pick a size that you feel comfortable with in your own play, or where you feel your opponent will make the biggest mistakes.


Exactly, so instead of trying to rely on a tiny theoretical GTO advantage, it's better to just use a size that you feel works best in the actual practical situation.

Oh, sorry. It appears I overlooked your post. The problem here is that this data is stored in the language file. When updates are made, this file needs to be updated as well in order to provide the required text for whichever new content has been added. So not overwriting parts of the language file might turn out to be rather tricky.

The computers are working on it.

Hi Scylla,

Could you build in a function that shows accumulated percentages for the entire range like in Flopzilla? This way you wouldn't have to count them manually. That would be great! Thanks

That's a shame.
Would be really great to implement a function do some permanent configurations, maybe within the gto interface while it is running (like the possibility with the colors).

Thanks anyway

Ok, I will consider it for later releases.

Ok, I will see what I can do here.

Hello
I'd like to replace my old computer setup and want to find a CPU for GTO analyses.
Im not too competent in tech things...
What is important? number of cores? number of threads? clockspeed?
What do u recommend for poker+analyzes?
Are there may concrete type of CPU what you can recommend?

+1
I find some used workstations mostly with xeon cpu. Are they can be a good deal?