Thanks for the article but they are talking about exploitative vs optimal when they make this statement.

I was just talking about theoretical.

I've written some articles about this, but it's possible to summarize in a couple of sentences.

Solvers are NOT taking the highest EV line. They are taking a line that prevents them from being exploited by someone else who is applying a perfect GTO strategy, the net result being zero for both players if they are taking the best GTO lines.

The point of solvers is to understand when your OPPONENT is taking non-optimal frequencies, so you can exploit them. You only make money in poker when you exploit someone else's errors.

Case solved.

Thanks FD!

That is literally exactly what I wanted to know.

What is the deal with mixed frequencies? It seems like no one knows why a solver does what it does or checks X % and bets another X %.

The mixed frequencies is the balance part of GTO play. It's literally the whole concept behind pot size, hand strength, and EV balance. Quite honestly, it's fairly complex math to get to nash equilibrium solutions in a reasonable amount of time. That's why a good deal of solvers are within an error rate of nash 0 or are on AWS style servers.

Can you explain how those two things are different when GTO is an acronym for "game theory optimal"?

I thought it prevented exploitation from EVERY strategy; net result being 0 vs another GTO line, net result being positive (not necessarily the MOST positive) vs non-GTO line.
Am I wrong?

see doodoo, i TOLD you.
some of the small %s you see are just the strat not being converged fully, but running it to .5% pot is generally pretty close to converged.
when you see like 33/77 mixes its clearly for a reason, generally its for balance. we dont ALWAYS check the nuts in a spot, we dont ALWAYS bet them, it uses nash equilibrium to balance it. just like rock paper sciscors doing a 33 33 33 split.

when you factor in ranges and bet sizes , thats where the frequencies start to diverge from what you would assume is a normal ratio like 50 50 or 33 33 33. like i said, a lot of the frequencies are purely to not be exploitable with your lines, not because its the most +ev play for that one hand in that one spot. its to protect your range in the future, essentially.

for some of the frequency spots, what you can do is heat map the EV of bet vs EV of check, and see if the EV difference is negligible. this is a good way to simplify your strat, because often there is no ev difference in one , so you can always just do one and be fine in a vacuum as far as immediate profit is concerned. however, this does mess up your frequencies and you can technically be exploited by over or under doing something.

i was just trying to help but u attacked me.. i hope what i said makes sense, and i think now would be a good time for you to apologize to me so i can forgive you and we can move on as friends

As far as I can tell (and I hope FD corrects me), GTO should result in a strategy that nets 0 because each strategy is completely perfect. If we looked at a single hand then this isn't true but if we take a line with our range against another GTO range, it's going to be net 0 because we're playing the exact same strategy...one that doesn't let us be exploited which means neither of us have an edge, at all.

So in the article they say exploitative can be more profitable than GTO - which is absolutely true and I don't think anyone disagrees with.

I was strictly talk about theoretical (optimal) vs optimal. So when both players are playing GTO.

I am getting a little confused though - i talked to Scylla and he wrote this to my question:

Sorry buddy I was just getting frustrated because I wasn't understanding it! No hard feelings.

Took you long enough.. but at least you're on the right tracks now.
Wrong, in poker you should always take the highest ev line, regardless of playing exploitatively or not. If you believe someone isn't attacking your calls frequently enough that raising is higher ev than calling, then you should never call strong hands. If your assumption is correct then BY DEFINITION villain won't be exploiting your capped range.
Such a mystery I wonder why we want to bet/check more often than not in poker

gto will net 0 vs gto.
vs anything else, gto will net positive.

So Ojune you take the highest EV line with your range but some hands in your range may not be max EV individually - correct?

then youre playin each hand in a vacuum and anyone with a HUD and any basic understanding of exploitative play (or even a gto strat) will walk all over u. because youre gonna always be betting for value when you open, get called oop and the board is 724 because "it's most ev" even tho this board sucks for you, and you should be checking full range. thus now your calling range is going to be completely weak.

this is where the solver (n smart players like me) check sometimes, because even tho im losing immediate value on my 77, in the future other smart players (like me) won't be able to take advantage of my checks. it is immediately less ev but nets the most ev as a comprehensive strat. hopefully you can get on board with this

that IS correct but thats not what hes saying

Wow I really thought you understood that one but I guess this misunderstanding of theory is very well spread and tenacious, caps lock is needed.

YOU TAKE THE HIGHEST EV LINE WITH EACH INDIVIDUAL HAND, ALWAYS, REGARDLESS OF WHAT THE BEST STRATEGY IS.

THAT IS COMPLETELY WRONG

I gave you the explanation but it seems I'm talking to a wall.

lol okay back to being confused again.

I used to think this was correct but then everyone said it wasn't.

ojune, buddy, im giving you actual examples of boards and even posted some screenshots from solver lines. youre resorting to adhominem attacks (the last resort of a desperate man clinging to his fallacies).
based on this its more likely that youre the one who's mistaken.

i am describing the comprehensive unexploitable (and profitable) gto strat that a solver spits out, which is focused around overall balance and protection, sometimes at the sacrifice of immediate EV, because it NETS the most ev.

what youre describing is some sort of card game strategy from the middle ages where the cavemen just shoved their chips in with brute force because it was the most immediately +ev play.

Yeah, look at the right part of his screenshot Doodoo, solver practically never takes a lesser ev line, when it does (TT for instance 0.6bb) it's just noise which would dissapear if you solve it completely. As for the left part, the ev difference is so small that it increases the noise, it would take longer to solve completely.

You can try running the solver, stop it, check the result and run it again. You'll see how it converges as you get closer to 0%.

Tbh, I tried telling you this a million and one different ways. Nash equilibrium doesn't result in the highest EV; it's purpose is to be unexploitable which is why deviating from GTO to exploit will often result in higher EV if villains can't or don't counter-exploit.

The different mixed frequencies are just created such that your ranges are unexploitable over the set of all possible counter-strategies not the highest EV

Solver's are only given you pseudo-GTO approximations based on input assumptions not true Nash equilibrium solutions.

Better yet, create a super simple situation so the solver will get close to 0% very fast. 2s 2h 2c board, 1bb stack, 1bb pot for instance.

Ok nevermind it doesn't work, it jumps from 0.02% to 0.03% and doesn't solve beyond that. I would suggest getting Janda's books.

Okay so who is correct?

Pallidan or Ojune?

They can't both be correct.

Ask brokenstars or someone competent in the run it once forum idk

Both pallidan and worldzmine are wrong