You (and your team) have a brilliant mind and excellent presentional skills. It is delighful to watch these videos even that I don't play nlh.

I love you Finding EQ!!

Once again, best video series on KZread! Unreal content bro! Please never leave us!

Nice presentation! For the JhTc example, the solver is suggesting mixing between all 5 actions, meaning by definition each action have the same EV, since solvers are always EV maximizing. The difference in EV is probably due to what percent exploitability the solve is going down to.

Excellent!!! It’s looking better with every video. Total game changer for GTO noobs like me.

"You can view EV regret as a sort of quantification of opportunity cost. There's an inherent tradeoff when learning poker between simplification and EV retention. And since the human race is not a monolith, GTO Check leaves it up to the individual user to determine his own appropriate balance between simplification and EV retention."

Nice, always look forward to your videosr

One thing I'm curious about. Take a spot where x combo shows ev regret in gtox of near 0%, is that factored in after solving for villains range where villain has calculated against us using a pure strategy for that combo. I.e. if we node lock a combo of ours to shift from mix bet and check to pure bet, the ev of that bet actually changes since the equity composition of our bet range changes so v's response range also changes thus altering the bet ev, and potentially making it much worse. Sorry for the poor wording, hope my question makes sense

@FindingEquilibrium

3 жыл бұрын

EV regret is based on the current, "GTO" strategies so it does not account for shifts away from the strategies set forth in the sim. This is why I mention that frequencies are still important because if your opponent adjusts he can exploit you when you shift your mixed strategies to pure. In other words, EV Regret calculates max EV loss assuming your opponent continues to play GTO. Is this assumption true? Of course not always because your opponent can shift at any time. But most human opponents cannot recognize and take advantage of frequency imbalances for individual or groups of combos as long as you keep your overall frequencies in check. Also, we know its not possible to memorize GTO frequencies so simplifications are needed, and as far as I know there is no better method to rationally and systematically simplify GTOs complex mixed frequencies currently.

Thank you!

Great piece here!

seriously how have not more people seen this video

Another awesome video. These videos have really opened up my perspective on solvers. Is it possible to do the type of studying you do on a chromebook?

is there a way I could book a zoom session with someone on the GTOx team to help me get used to the interface and its features?

What is the "GTO" or "Finding Equilibrium" theme song? I need it while grinding, coupled with changing my avatar to Terminator

How do you find the spots where you are the furthest from the GTO? Any good tool that could help? I've heard about Leakbuster, but not sure if it's any good.

Wow this is so Meta! A system for making solver-based learning more efficient... ive never bought a solver or a sim before, thinking I wouldn’t be able to learn efficiently from them... but I do recall spending a lot of time years ago trying to learn how to invest in micro loans offered on peer lending sites. I just spent hours looking over loan data. Then I got bored. But maybe this would apply well to poker solvers... The thinking here is *excellent*! “regret” is a label I think will pull people in, but I don’t think really a fitting term as this is basically a math approach and a way of comparing actions, whereas “regret” suggests illogical emotions... basically the system is Spock and the term regret is more McCoy🤣🤣. Why not just call it “EV deficit”-you are really looking at the degree to which a play is non-optimal, or the deficit between your play and the optimal choice.

if there is mixing going on all actions have equal EV (the difference in EV is there b/c the algo only converges to a nash eq. ("in infinite steps"), it doesn't calculate one exactly). there is no "highest EV action" since all actions have the same EV when mixed. Ofc you can't be certain if the mixing goes away or the EV difference becomes 0 in the nash equilibrium your solution should converge to without further investigation ;)

Pfff!! cannot stop now! jajaja.

What stakes do you play?

hI!! The best of GTO!! When we have GTOX Online?

Take my money 🤑🤑

Every time you've mentioned EV regret, I've thought "this sounds like opportunity cost" and today you confirmed. Makes sense. So for the next evolution, how about sensitivity of strategies to small changes in inputs? I find my play very solver approved because, you know, I change the input ranges or bet size options until it says so. Ahem. joking. Good vid.

@FindingEquilibrium

3 жыл бұрын

Yeah, this is actually one of the risks of node-locking IMO. Slight changes in the strategies can dramatically change outputs and you can justify a lot of things once you start manipulating the inputs. But this is one of the reasons why I think emphasizing EV over frequencies makes sense because EV tends to be more robust to slight changes in the inputs compared to frequencies which can shift 0-100% or vice versa very easily.

Is there a release date?

The EV regret as i understand is for given bet sizes i.e 1/2 pot,2/3pot 1pot ect. Is it possible to check if the EV regret is changing proportionally to between those sizes i.e if Ev regret b/n 1/2pot is 0 and 2/3 pot is 0.2 can you find if betting 0.60 pot will be lets say proportionally 0.1 Thx alwsome stuff

@FindingEquilibrium

3 жыл бұрын

Thanks. No EV regret can only calculate max EV loss based on the current GTO strategy so you would need to rerun the sim to check the EV regret for different sizings.

Great but how can I use this to help me get better?

"enhance"

FYI, GTO Wizard is copying most of what you are trying to do. If you didn't know.