Hello my friend.

1. Go to edition tree
2. Click edit action colors

Can you view EVs for suboptimal actions in GTO+, say if I want to quantify how costly calling a pure fold was by looking at EV loss?

For this, switch to the EV tab (F3).
For each hand, not only the EV is given, but also the EV for every action.
So, in the screenshot below, AcJd has an EV of 3.85 for betting, 3.93 for checking, and 0 for folding.

]

Scy, its possible when "Open a load selection on fgiles in the same directory" (ctrl+w), start defauilt unselect all nodes?

Now, the default is all nodes are select, so when you filter 1 flop, you need to start to clicking buttons.

https://gyazo.com/0e46ccd82147ef62cdb57990bf006586

So.. if all nodes start unselect, you will use one click and load, i hope yoiu understand me.

Ok, I think that we'll add some customization, so that the user can set the desired default.

Scy how are you?

This is an agg report for "draws" on the flop on all runouts for this node(also there is the possibility for filter boards for go more in detail).

https://gyazo.com/207aa99a92cdfcb9774eb72d137b603b
By Hand

By Combo
https://gyazo.com/5a3c3c2d76bbd955be214b4732f0137e

By Draw
https://gyazo.com/235cc760164cf617dd06203a7daebce7

its possible to have this type of report not only excel but like on this image?

You can also select by action, hands, combos, draws.

GTO+ only have the feature for action only. Can you check this feature please? i think its a good one!

Ok, I will consider it for future releases.
However, please do keep in mind that these sorts of aggregate reports are not particularly useful.
For example, if the board is 9h9c9d, then you have trips.
If the board is 9c9s2h, and your hand is T9, then you also have trips.
And having two pair with T9 on a T96 board is not the same as having two pair with T9 on an AAT board.
Different hand strengths have entirely different meanings on different boards.
We can add them together into an aggregate reports, but this report will be adding data together that can realistically not actually be treated as being similar.

But here, for example, you have the option for filter flops and go for more details

https://gyazo.com/5e84c2164714f0736c1e7d0b2a64358b


Here on 999, they doesnt interpret as trips

https://gyazo.com/8acf7fec8314bac8be37c97d2cd67806



Example of T9 on AAT nad T96

https://gyazo.com/13a24b79382992aa662ccc2e60db9acb

https://gyazo.com/8acf7fec8314bac8be37c97d2cd67806



Look how easy was for me filter flops and looking the strat!

Ok, I will consider it for future releases.

Should I use the rounding function when playing against the solution? Im not sure how it works and sometimes it completely changes the solution.

Also, whenever I get dealt another hand without playing the previous one the frequencies of the actions are not updated and are even recognized as mistakes

Ok, thank you for pointing that out.
It looks like we're not properly resetting certain parameters when the hand is artificially ended.
We'll take care of that for v154.

In your last picure, the action "Raise to 5.17" is part of the mix, but only by 0.04%.
This means that it's technically correct, but only barely part of the solution.
If you want these smallish actions to be considered as "incorrect", then you can apply rounding.
For this, as a default, I would recommend setting "Round for 1" and "Multiply with nr of actions".
This means that, if there's 3 actions, you're rounding for 3; and if there's 4 actions, you're rounding for 4.

Would it be possible to just practice a decision over and over too? Without having to play the entire hand

You can drill a decision with "Drill the current decision" (use the navigator on top of the interface to navigate to the target decision).
To restart the hand, at any point in the hand, press F1.

Hi, scylla! I found strange locking behavior. In this example it is impossible to set different strategy for some combos. They are always locked in pairs. Likely, this happens only on fd or monotone flops.

After a deeper thought, this may be not a bug at all. Program just forcibly groups strategically equivalent hands together. But it was not clear at first glance.

Scy hello.

I'm using the new filter system a lot, so thanks for the addon.

Regarding the image I have uploaded:
OTRIVER

Can you add to the Nomade hand category filters of for example SD blocker and FD Blocker on the river?
Since we can't do this ourselves and I think it's good to have it separately.

Then another thing, the suit blocker system I think it's better to have it in Pio's style.

It's like this:
It has 8 button panels, (2 per suit)
Corresponds to high card and low card.
In addition to this, you can also not only select, but block so that a certain suit is not displayed.


Hope all going well, thanks!

We expect to add more customization for future releases.
In the current release we're testing if everything with the new system is working as intended.

Actually, you can do all of this with our approach as well.
Just select whichever suits you want to see, and leave out the ones you want to block.

Yes, that's correct.
JsJh and JsJd are the same hand on a Qs7c4s board.
By bucketing them together, we can achieve a speedup of roughly 40%.
However, when editing, they also need to be treated as being identical; otherwise the speedup is lost.

Yes, but here if you want to look how mucho 1suit blocker bets, you need to do 7 clicks. compared to 1

Testing yesterday i see if you click and drag you can select something fast the suiteness.

Ok, I can see if we can add some option for quickly entering the data here.

Hello.
Can you add the opportunity to calculate MTT ICM mode including bounties? Also can you add the opportunity to quickly lock nodes at turn and river, as it was in Crev. (the ability to set average frequencies for all runouts of the turn and river, select hands for actions (not just for one runout, as it is now)).

Ok, we will take it into consideration for future releases.

A problem there is that if you set a stat like "top pair", then this will mean something different on different boards.
For example, if the board is T96, then you'll probably want to play differently, depending on whether the turn is an A, K, Q, J, 8, 7 or 2.
It's next to humanly impossible to create general strategies that cover all possible turn cards.

In the solver, we want to build a max exploit strategy. We have Х2Н and research, in which we can see that this type of opponents plays the T96 structure with such and such reactions, and then plays with such and such frequencies on the turn with the cards selected in the research.

At the moment, in order to build an acceptable strategy, you have to manually lock each turn, adjusting it to the frequencies from the research and assuming humanized strategies.

So the situation you describe is not a problem. It is required that it be possible to create buckets from turn runouts and perform locks on them by "hand categories", also to be able to be possible to cut off hands by equity. It would be even more convenient if it would be possible to set the frequency of actions, and the solver by itself would distribute the hands into branches (for example, relative to their EV for the already calculated balanced strategy).

Also, the creation of turn buckets would be convenient for analysis the calculated strategy (the ability to combine the turn runouts and see the average values that are in GTO +, such as the average equity of ranges, average frequencies at nodes, etc.)