Going Crazy Studying GTO

Been studying GTO for a couple of weeks now. Having a hard time coming to grips with this statement:

What do you guys think? I'm okay with everything GTO says to do except when it really, really, really seems like I'm way behind but the GTO Trainer wants me to keep calling anyway.

While we could just say, we're humans and we know best, then we have to ask ourselves why are we even trying to learn GTO in the first place? Which by the way, is kind of the question I've been asking myself a lot lately.

I'm sure GTO is not worthy of going crazy over.
I'm also sure that I probably never will play at high enough stakes where GTO would actually start to play any kind of role.

Flyer35 said:

Been studying GTO for a couple of weeks now. Having a hard time coming to grips with this statement:

GTO Reality: GTO assumes your opponent is an omniscient, perfectly balanced supercomputer. It will force you to call down with hands at your Minimum Defense Frequency (MDF) to prevent an optimal opponent from auto-profiting against you. Against real-world human opponents who don't bluff enough, calling down according to GTO will often look and feel like throwing money away.​

​

What do you guys think? I'm okay with everything GTO says to do except when it really, really, really seems like I'm way behind but the GTO Trainer wants me to keep calling anyway.

While we could just say, we're humans and we know best, then we have to ask ourselves why are we even trying to learn GTO in the first place? Which by the way, is kind of the question I've been asking myself a lot lately.

Click to expand...

I think this is an area where you should quickly deviate from GTO if you have any sense villain is under-bluffing or you have any understanding that the spot is under-bluffed (even in the slightest in both cases).

So why have you spent time learning GTO to deviate sharply in spots like this?
I'd say so you understand the unexploitable base line and can then make an informed decision to deviate from it if you choose.
This gives you the option of sticking to GTO if you do find yourself against a good aggressive villain who you think will be bluffing to the point of exploiting you.

When 'training' I'd make the GTO correct play AND mentally note how, when and why you'd deviate if the GTO line(s) feel odd to you.
May be write down "troubling" GTO spots as you train and explore more later. I don't know if your trainer allows you to mark hands, that would be a nice feature.

Tero said:

I'm sure GTO is not worthy of going crazy over.
I'm also sure that I probably never will play at high enough stakes where GTO would actually start to play any kind of role.

Click to expand...

Good point. I was watching some high stakes wsop play yesterday. While the players may have/may not have been thinking GTO, the vast majority of plays certainly were close to, if not exactly GTO.

So the next question becomes - if you have any aspiration to go further, when do you start learning/practising GTO, and when is it too late to start? And even if you never plan to play/win the WSOP, isn't it a normal next step regardless?

monkeytilter said:

I think this is an area where you should quickly deviate from GTO if you have any sense villain is under-bluffing or you have any understanding that the spot is under-bluffed (even in the slightest in both cases).

Click to expand...

How many times have I had a perfect read on a player only to find that all of a sudden the nit is throwing in a big bluff line, or the maniac was dealt Aces?

Which makes me think of another GTO'ism:

I win lots of pots, but also lose lots of tournaments. If poker is a high variance game, how do we combat that if not GTO?

Flyer35 said:

So the next question becomes - if you have any aspiration to go further, when do you start learning/practising GTO, and when is it too late to start? And even if you never plan to play/win the WSOP, isn't it a normal next step regardless?

Click to expand...

It is never too late.
I think I could beat myself up over and over by honing other poker skills without ever even touching GTO. Is it the next step? Not necessarily.
One has to ask oneself what good comes of it? There are so many other interesting things in life that one could self-improve in.

Flyer35 said:

If poker is a high variance game, how do we combat that if not GTO?

Click to expand...

With volume. I don't think anyone ever promised that GTO would help defeat variance.

Tero said:

It is never too late.
I think I could beat myself up over and over by honing other poker skills without ever even touching GTO. Is it the next step? Not necessarily.
One has to ask oneself what good comes of it? There are so many other interesting things in life that one could self-improve in.

With volume. I don't think anyone ever promised that GTO would help defeat variance.

Click to expand...

Way past you on your first point - I've decided that poker "is" the thing that I want to self-improve. I'm finding that you can't study GTO without studying everything else that makes a good poker player. The fact that AA is a great starting hand and should be treated as such is GTO and common sense and basic poker skill. equity, pot odds, implied odds & EV aren't just fundamental to GTO. They're fundamental poker concepts.

I think GTO may be the main curriculum thread off of which everything else hangs. It kind of gives me a direction, so when I go off track, I know where to go to get back on track.

You're right about variance. But maybe:

• No Tilting / Dynamic Adjustments: A solver doesn't get frustrated, change play styles out of emotion, or leak chips by over-folding or over-bluffing.

I definitely do those things left to my own devices, which led me to create this thread in the first place. I couldn't figure out why GTO wanted me to lose, but it's making more sense now. It's just the law of averages.

Solvers don't get emotional. If we mimc a solver when nothing else seems to be working, maybe we're ahead. Not sure though if that's not just discipline under a different name.

Studying using a solver is very complex requiring us to study several layers of data.
Layer 1 GTOE solver vs solver output charts following those exactly is only 70% effective in ROI terms vs average humans.
Layer 2 adapting solver outputs by slightly changing the inputs. Putting a tight Villain in an earlier seat ie BTN actual seat but we move them in solver to CO range. That increases effectiveness but only slightly
Layer 3 using node locking to adjust Villain range preflop flop and river. This is very complex and requires us to use our hand tracking player pool data.
This is of course the most effective ROI solver study method, but a single hand can take hours to refine. The good news is we begin to see patterns and do not have to study every variation of similar hands.

AI assistance is good - single size simplification is good in AI solves but AI does not apply outliers to your player pool so ROI is not high vs the players making the biggest mistakes.

GTO study will not however help you with the variance you described. Low frequency occurrences like a NIT finding a bluff every 100 hands cannot be studied, nor can we study to avoid maniac suck outs.

This hand helps reassure me. BN was going for the bluff of the year by opening with 64o on the button and double barreling the flop and the turn. If I'm sitting in the BB with a totally missed board I'm thinking I'm busted, yet ....

GTO isn't as bad as they say, I play MDA, which was always the best thing in the game.

Flyer35 said:

Been studying GTO for a couple of weeks now. Having a hard time coming to grips with this statement:

GTO Reality: GTO assumes your opponent is an omniscient, perfectly balanced supercomputer. It will force you to call down with hands at your Minimum Defense Frequency (MDF) to prevent an optimal opponent from auto-profiting against you. Against real-world human opponents who don't bluff enough, calling down according to GTO will often look and feel like throwing money away.​

​

What do you guys think? I'm okay with everything GTO says to do except when it really, really, really seems like I'm way behind but the GTO Trainer wants me to keep calling anyway.

While we could just say, we're humans and we know best, then we have to ask ourselves why are we even trying to learn GTO in the first place? Which by the way, is kind of the question I've been asking myself a lot lately.

Click to expand...

Question: Have you tried testing what it suggests you to do at least once? If yes, what's the result? If no, would you want to try testing the waters?

Flyer35 said:

This hand helps reassure me. BN was going for the bluff of the year by opening with 64o on the button and double barreling the flop and the turn. If I'm sitting in the BB with a totally missed board I'm thinking I'm busted, yet ....

View attachment 414659

Click to expand...

That's such a hard core bluff, not easy to find for most? I hope I'm not getting bluffed like this in my games too often

I guess preflop raiser has all the better Kings and the 10 on the turn is a scare card to continue pressure on all pairs that just became 3rd pair and weak flush draws that missed.

Curious now, how often does GTO go for the 3rd barrel now that BB range is now much stronger after calling 2 bets?
Deeper effective stack does it fire the river more often?

monkeytilter said:

That's such a hard core bluff, not easy to find for most? I hope I'm not getting bluffed like this in my games too often

I guess preflop raiser has all the better Kings and the 10 on the turn is a scare card to continue pressure on all pairs that just became 3rd pair and weak flush draws that missed.

Curious now, how often does GTO go for the 3rd barrel now that BB range is now much stronger after calling 2 bets?
Deeper effective stack does it fire the river more often?

Click to expand...

I posted a couple of GTO Bot versus GTO Bot hands in the Optimus Poker thread. GTO Bot #1 decided to try a 3-bet pre-flop bluff from the BB into the Btn. GTO Bot #2 apparently didn't believe GTO Bot #1 had a hand, which it didn't. Both GTO bots went to an all-in river showdown with air.

GTO Bot #1 never had a hand beyond chasing a low backdoor gutshot that never completed. GTO Bot #2 had an OESD that also never completed, yet GTO Bot #1 shoved on the river with air and GTO Bot #2 called the shove with air.

https://www.casino.us/cardschat/pok...-web-based-gto-solver-ai-565206/#post-7359844

These GTO Bots can be very mistrusting!

Flyer35 said:

Been studying GTO for a couple of weeks now. Having a hard time coming to grips with this statement:

GTO Reality: GTO assumes your opponent is an omniscient, perfectly balanced supercomputer. It will force you to call down with hands at your Minimum Defense Frequency (MDF) to prevent an optimal opponent from auto-profiting against you. Against real-world human opponents who don't bluff enough, calling down according to GTO will often look and feel like throwing money away.​

​

What do you guys think? I'm okay with everything GTO says to do except when it really, really, really seems like I'm way behind but the GTO Trainer wants me to keep calling anyway.

While we could just say, we're humans and we know best, then we have to ask ourselves why are we even trying to learn GTO in the first place? Which by the way, is kind of the question I've been asking myself a lot lately.

Click to expand...

Kudos for your commitment to this journey.

I don't (presently at least) have the aspiration to play at that level, so I study concepts that I think are more relevant to the games that I do play. I'm only personally interested in tournament games and find most of my opponents don't even have a good understanding of ICM (or even more basic concepts like pot odds!), so practising GTO would be a wasted over-layer. That said, understanding GTO ideas like MDF and Nash at a conceptual level are extremely useful.

As for the hand you post in the last message , it's completely wild in the "normal" poker world but does make sense to me. GTO 1 shoves at the river for the reason you already quote (no chance of winning) , and GTO #2 on the button calls because even though they are likely beat they only need ~30% which they probably have if villain is going to bet a decent % of their air (which is going to outnumber value hands if middling hands want to instead showdown).

Interesting discussion.

Enjoy your poker experience and just put GTO down.