Hi Scylla, I'm very curious about this theory, that bet sizing doesn't matter in GTO solution, because I've read before in GTORangeBuilder that there *is* an optimal bet sizing that maximizes EV, here's the link: http://blog.gtorangebuilder.com/2014...barreling.html

I purchased GTO+ and did indeed try to verify the betsizing-doesn't-matter-much theory, and it turns out that it appears to be that between 1/3-2x pot, EV difference *is* as small as neglectible.

Well, I'm conflicted here by this delimma. Is this one of those theory vs practice thing? if so, why is it that theory says sizing matters while in practice it doesn't (btw I'm not concerned about different sizing exposing opponents different weakness - I know this is true already)

We currently don't offer that. We only display the aggregate frequencies for the entry phase (which will usually be the flop).
Beyond that point you're looking at an average (of all flops) of an average (of all turns).

If OOP has multiple 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 each data point will have been measured for 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.

There's usually a dozen or so different lines that will lead from the flop to the turn. And each line will be reached with different ranges and stack-to-pot ratios for both players. Any turn report for flop trees would need to be created separately for each line. And if in real play the players deviate only slightly from their GTO strategies, this data is already no longer valid.

I don't believe that theory currently says this. In the past it was (quite reasonably) believed that there would be such a thing as an optimal bet size for a certain situation. However, there was no way to calculate it. Now that it can be calculated, as it turns out, there's virtually no difference. There's even no real advantage to using multiple bet sizes vs using single sizes.

Now, of course, in practice it may be that a certain player pool reacts poorly to certain sizes.
However, this is situational; if you check back a year later, that player pool may be playing differently.

Hi Scylla, great software !

After building a tree, running it against 163 flop subsets and saving it, I'm trying to just change the ranges to solve the same configuration (saving it under a different name), but when hitting "process database", it changes the ranges back to the initial values.

What am I doing wrong ??

You should clear the database and build it again.
Just changing the ranges is not enough, given that the trees in the database are not affected by this.

Ty, it works now :-)

Hi Scylla, is there something similar to rounded strategy feature of PIO?

Suggestion: in addition to display methods 1 and 2 and "all at 100%", it might be nice to have an option to have a "normalized" view where the most common combo is displayed as 100% and then the other weights are scaled off that. In some narrow range spots I have trouble seeing all the combos when using my preferred setting of display method 2, as all combos appear as just a sliver of the box (similarly for display method 1 they would all appear as a very faded color). But if I use "all at 100%" to see everything then I don't get a visual representation of which combos are more frequently in this very narrow range.

Come on now - Your rationale behind not including bet sizes when PIO does is that PIO's playerbase specifically wanted that. I think it's obvious now that your playerbase wants it also, no?

Saying that bet sizing has nothing to do with EV is ridiculous and several users have expressed that sentiment now. GTO+ is fantastic and there's no reason it should be lacking when it comes to this feature.

I believe that we already had this discussion in an earlier part of the thread, starting at post #8488. Around the last post it was pointed out by several users that pio does not offer such a feature. Currently I'm also still not familiar with such a feature, nor was anyone else at that part of the thread. That being said, in our interface it's actually possible look into the effect of bet sizing. This can be done by building a database of the same tree, but for different bet sizes. So if you want to look into the effect of bet sizing on EV, then this is already possible. I think you'll find though that the influence of bet sizing is marginal at best. As far as I've ever seen or heard of, in GTO play, all bet sizes will perform roughly the same.

Not at the moment no. What rounding would mostly be useful for would be to make solutions easier to memorize. The problem here is that no matter how much the solutions are simplified, memorizing thousands of hands will still be a nearly impossible task. Even moreso, memorization will still not tell you how to adapt if the ranges or board are different. In my opinion a more productive approach is to look at the hand values (top pair, middle pair, set, flushdraw, etc) and see if you can find patterns in how certain spots are approached. Understanding why certain strategies are chosen is far easier to apply in practical spots.

Try turning ON "Settings->Scale small stats".
This will stretch small stats a bit, making them easier to discern.

hi i want buy the cardrunnersEV and there is a free GTO+ solver right?
i looked some youtube videos from GTO+
i can choose between Cashgame and SNG,
but my question is:
can i also choose for MTTs (Tournaments) ?
because i am a Tournament Player and want GTO Solution for Tournament Spots, for example with 50BB deep and i got a range opening BTN and BB defends with 35BB stack.
is this possible?

There will be no ICM considered if u solve it with GTO+...I think CREV will be useful but it is more complicated for use.

Correct.Optimal bet size can be applied only on river if u have perfect balanced range.And that's is always ALL IN.

At the moment we do offer a SNG feature, which allows for up to 10 players and up to 6 prizes. If your hand falls within these restrictions, then it’s possible. Please do note though that GTO does not really work that well with non-zero-sum situations, such as tournaments. The problem is that multiple GTO solutions exists, and if villain decides to deviate from his optimal strategy, you would need to adapt your own strategy as well. This contrary to zero-sum spots such as cash games, where you can always play your GTO strategy, even if villain deviates. It’s for this reason that we are currently not too invested in tournament adaptations of GTO+. That being said, if you're far away from the money, then MTT's are usually comparable to cash games with 10% rake. So for MTT's this is a workaround that could be used.

10% rake means playing super tight pre flop.Since GTO+ is postflop solver is it correct to input 10% rake when solving?PokerSnowie compares tournament style with it's ante games but because it starts from pre flop.What will be the output if we put 10% rake before solving?I think no rake is correct to put it in before solving when solving tournament spots far from the money.

Hi, How can I use play vs database feature?

What I was mainly saying was that we are currently not heavily invested in tournament play for GTO+, given that these situations are non-zero-sum. A workaround to still use it for tournament purposes would be to add some rake to simulate the fact that some of the money is "lost" (whenever two players get into a pot, the non-active players profit, due to the chance of someone being knocked out of the tournament; so whatever is lost by player 1 will not all go to player 2; some of it goes to the other players). When being very far away from the money I assume you could just use no rake.

For that, calculate a database.
In the "Play against the solution" feature select "Play versus database".

hi!! Is there any "force IP bet" option like in Pio? I want to force the program to bet whole range at 33% size in flop and see what happens. i have more than 30 boards so going one by one is not an option. Ty

Hello

I am not sure about rake cap: is it expressed in chips or in bb?

example: NL1000, rake structure is: 5% cap 4bb

On the cap, should I write 4, or 10*4= 40?

Thanks

You can just use the editor to remove the "Check" action. After that, clear the database and rebuild it from the new tree. If you want to use the same flops in the new database, then you can export the flops in your current database with "Export flops to file". When rebuilding, enter the same filename for "Import flops from file" to build the new database. In the example below the file that is used is called myflops.txt.

Typically cap will be expressed in $. However, if you're expressing everything (stacks, pot) in bb then you should express cap in bb as well. So basically, the same unit should consistently be used, whether it's bb, $, or anything else.