Hey, I have a lot of profiles in GTO and would find it useful to be able to save and organize those in folders.

Hi Scylla,

I am wondering why I can only see player one's bet and check but cannot see player two's bet and check frequencies in the database distributions. I turned off force check for OOP.

Thanks

Thx for your detailed answer!

If i unterstand it correctly, then is the betsize not a crucial factor for the overall EV at most of the boards - is that right?

When the answer is yes, is there another way to check the "best" betsize? "best" means not in the sense for the overall ev, rather for Villains play. Perhaps some Boards with different betsizes have the same overall EV, but one betsize is harder do defend as one other betsize.

When i use multiple betsizes in one tree and GTO+ prefers a certain size, can i assume that the preferred size is a good way to use this size only at this board?

Thx in advance and sry for my bad english!

Ok, I will see if I can do anything here for later releases.

Thank you for the feedback,

Scylla

If player 1 has multiple actions in his first decision, then there is no single decision for player 2 where these frequencies can be measured; there will be a separate decision for every action by player 1. So for that reason it's not very straightforward to display overall data for player 2; a separate table would be needed for each decision. Secondly, and more importantly, the problem is that for every tree, player 1's range will be different for every action. And as a result, all data for every decision by player 2 will be measured under different circumstances. This means that this data is not really suitable to be displayed in a table. We can consider displaying it anyhow for later releases, but right at this moment, it has been left out for this reason.

That is correct.
The quality of play is far more important than the sizing of the bets.
Just about any bet size seems to perform similarly to any other bet size, as long as you play GTO.


The main thing to consider, is in practical play, versus which bet sizes do your opponents seem to make the biggest mistakes? Given the fact that your opponents don't play GTO, and are inherent to making mistakes, the best strategy will most likely be to try and put your opponents in a spot where the weak spots in their play are the most pronounced. This will however need to be derived from practical experience, as it can not be calculated mathematically.


When comparing the bet sizes separately, you will probably not find a significant difference in performance. Different bet sizes typically have a very comparable EV for any hand. Most likely the best approach will be to focus on using bet sizes that provoke the biggest mistakes by your opponents in actual play.

Scylla,

I'm wondering if you can explain to me why the EV of both players on the flop doesn't add up to the pot size exactly.

It should, unless the situation is not zero sum. This will happen if rake is involved, or if it's a tournament. In CREV another exception might be in the case of using delete actions. In all other cases you should be able to calculate the overall EV for IP by using the weighed average of his EVs. The sum with OOPs EV should be the pot.

So for example, below, in 35.9% of the cases IP will have an EV of 12.67 and in 64.1% of the cases he will have an EV of 17.89. That makes his overall EV 35.9%*12.67+64.1%*17.89=16.02. The EV of OOP is 13.99. And their combined EV is therefore 16.02+13.99=30.01, which is the same as the pot. Admittedly there's a 1 cent difference, but this is due to rounding in the EV values and frequencies.

Should you indeed have a spot where you feel the EV of both players does not add up to the pot, then please send a savefile to support (or possible a screenshot).

Hy scylla,

in the database mode is it possible to put a weight on the different flop?

otherwise in the columns near the aggregate graphic it would be possible to write EV of each combo for every single flop of the database?

OK, Just bought the software and really plunged in never used solvers before so I have a real noob question:

Range 1 as oop in the inputs I did usually has the tree weighted heavily towards a check, where can I choose to make Range 1 to reflect a strategy if he is the preflop aggressor?

From a GTO standpoint it doesn't matter who was the preflop aggressor. When making decisions in poker, it's never relevant which past actions have led to a certain spot; the cards have no memory in that respect. That being said, in most cases you will find that the preflop aggressor will bet out pretty frequently in a GTO solution, in particular on flops with high cards. For this however, the GTO algorithm doesn't need to know which actions have taken place preflop. If everything is as it should, the preflop aggressor will typically have a stronger range as compared to the player who was passive preflop (for this look at the preflop equity for both player's ranges). On most flops this will mean that OOP will be able to continue his aggression. If in your tree OOP checks frequently, then I would expect that most likely IP's range has hit the flop. For example, on low connected flops, such as 8h9hTs the player who was passive preflop has often hit with most of his range, and the preflop aggressor will need to cease his aggression. In the situation that you're referring to, did OOP perhaps have a high equity preflop, which dropped once the flop came? Or, alternatively, did IP perhaps have a strong preflop range as well? If you want me to take a look at the hand, then please send a savefile to support.

Such a feature is currently not available, but we can consider it for later releases.

Under "Specific hand" you can select a hand.
The table will then display the data for that hand for every tree.

Thank You so much Scylla, that makes perfect sense.

For future reference, how is a savefile created and what is the email for support ?

Thanks for the answer!

I saw that, I was looking for the combo AQs or AQo, instead of having to search for every combination, but now I'm realizing that probably doesn't make that much sense

For support, there's a contact form on the website.
And to create a savefile, go to "File->Save as …" in the menu.
Or press on Ctrl+S on your keyboard.

When it comes to scripting abilities, is there anything GTO+ lacks that piosolver pro has?

Not that I can think of. We offer the ability to create databases of solved trees, as well as a feature where GTO+ will solve all files in a given directory. With "basic" saves the solutions can be stored in very small savefiles; as a result, even entire databases will rarely exceed a disc size of more than a few dozen MB. To the best of my knowledge, our approach is far more flexible and easier to use than pio.

Scylla,

Thanks for making some awesome software.

I recently started with GTO+ and was wondering if there was any possibility of making a night/dark mode? I would imagine it's not a top priority but it would be a nice addition sometime down the road.

Either way thanks again and keep up the good work.

Hey scyla, (example situation) I solved a database with a lot of flops, but I would like to know which frequency the solver bets with overpair on Flop 1, 4, 6, 7, 8 for example, there is some way to filter these flops and then get an agregated report for them? Or I have to check one by one and use excel?

Enviado de meu XT1033 usando o Tapatalk

Hey Scylla, one very small bug. Sometimes when i press the stop solver button, it keeps restarting calculation. This is when im parsing a folder full of gto files. Thanks.

The best way that I can think of at the moment is to make a copy of the savefile and then delete all trees that you're not interested in. To delete trees, in the database under "Run solver", right-click them. Or right-click their icon in order to have them ignored in the report. I can look into further management functions for later releases.

Ok, I'll see what we can do here.

Ok, thank you for pointing that out.
I'll see if I can reproduce this issue.

Any hints about future features coming down the pipeline in the next (beta) releases, Scylla?

I prefer to not discuss development in this thread, but there will be a small update within a few days with some minor bugfixes and a much higher framerate. The higher framerate should allow GTO+'s interface work better on systems with slower processors, as well as provide fluid functioning in case of extensively drawing very large trees.