QJT9 ss

Solver assumes other solvers. Once you get so many calls it plays a 3b or fold strategy. This is correct against other solvers.

Vs idiot live players you can play this way and print as well, but it may not be the MOST profitable or the MOST fun. Emphasis on most. You will print playing this way, but most don’t have the patience to do so or the drive.

Live players make egregious mistakes pre and post flop. If you nodelock the solvers opponents to play like live players, you will see a solver deviate massively to exploit them, but the solver will in turn then become exploitable.

The output you are seeing is the most positive EV strategy you can implement while being completely unexploitable. That’s different than having the most EV strategy based on your opponents tendencies and leaks.

As far as study. Every day or few days pick a topic. Always start with preflop. I’d go in this order.

RFI from every position
Cold calls from Button and BB
3b from Button, SB, BB
Squeezing
Calling 3b
4betting
Calling 4b

Drill the trainer after the lessons and then again the next day to see if the concepts stuck. Hit 90% over 100 hands and if not, retake the lesson and focus on the hands u made mistakes from and understand the concepts and hand properties that you are making the mistakes on.

Once ur preflop game is down I’d work ok postflop in this order.

IP-SRP
Btn vs bb (wide ranges as PFR)
Btn vs ep (tight ranges as CC)

OOP - SRP

Bb vs button (wide ranges as CC)
Ep vs button (tight range as PFR)

Multi-way SRP
IP as the button CC vs multiple people both when their is a bet and when checked too

OOP as the EP raiser when donked into, when checked too, and when you check and face a bet from IP

This is the foundation you need to become a winning player. Take 2 weeks to a month studying here and then come back.

Very solid advice and explanations, appreciate it Sir !!

I am not 100% sure, but I am quite certain this is a limitation that is made to simplify the simulation. Edit: To be clear: Its not because Solver will not have a flat range, its because who ever ran the simulation limits the action to fold or squeeze after a certain number of flatters to simplify the simulation - I am guessing because of computational limitations

Yes, you are never seeing anything like 'pure solver output' in these things, you are seeing a very rough version of it at best, misleading strategies that lose against other solvers at worst, and very little in the way of helpful data points to strategise versus live players. I think it's a mistake to assume these things are always an improvement on poker theory. Where solver says 'this very bad' or 'this very good' we should pay attention, but so many of these are close calls according to the solver and we should just follow poker theory in those spots rather than a distant cousin of the actual GTO solution.

ah, interesting .... thx for the input guys.

i was starting to really doubt my overall poker skills, like, how on earth is QJT9ss not a call in this spot

also: just watched another video from on of the coaches at mastermind, where he emphasis how much he actually deviates from "pure gto" in certain live game scenarios and why, and when i look at one of the games i regularly play in, there's so much passivity (limping and coldcalling a wide range of shitty hands), that to me i would feel like a crime almost to not call a hand like the one mentioned above in those spots

This is a similar spot but a different simulation. Here they allow for a flat. I am almost 100% sure the reason it says 3b or fold in the other simulation is because of limitations set by the person who ran the sim (to ease the computation), not because solver wont have a flat range in that particular spot

This image I am guessing is the simulation OP refers to

I am not the greatest expert on these things, but I get the distinct impression that altering the parameters of a preflop simulation will potentially have a profound effect on the output. Therefore it is of utmost importance to know the parameters of a preflop sim before you use it to whatever games you play in. Stuff like rake, ante, straddles, stacksize and whether or not to allow open limping are all of huge importance

Wow …. From 0% flats to 37,4%
That’s some difference;-)))

its a lot easier to study standard strategies at 100bb, 6max the 200bb, 8max

eventually you can learn intuitive ways to translate strategic changes as you get deeper, by position

in the scenario you described the I imagine the main reasons the solution prefered to fold would be

1) your hand performs poorly against additional raises and
2) the deeper you get, the more important it is is to have nut blockers and avoid being dominated (in freeroll type spots)

now in an actual live game, calling with your hand might still be ok / +ev

With my limited experience using holdem preflop solvers, I would imagine this has to be due limitations on computational power and/or program limitations. Especially when you are 200bb deep 8 handed. Usually you have the option to have the player closing the action flat in addition to anyone else being able to flat, so if you allow a player to flat there, then you end up allowing it to go 5 ways postflop multiway. Then being 200bb deep and 8 handed the game tree starts getting impractically large. Memory limitations either won't permit it or the postflop models have to be simplified so much that the accuracy is limited. IE, fewer abstractions will be used, limitations on bet sizing and donk betting will be used, sufficiently multiway pots may have to assume that the last betting round will be on the turn.

There is not many scenarios in which I am folding QJT9ss preflop in a cash game lol. What solvers cant equate for is the interpersonal dynamics of live poker. Especially if these are invite games, no one likes to play with nits just be comfortably rolled to ramp up the variance. DS bombs away

Mastermind rents and uses Amazon servers I believe and runs solves for months at a time last I heard. I doubt this is solver noise due to lack of computational power.

This is most likely the optimal non exploitative way to play. A solver will absolutely demolish loose ranges and in deep stack situations put hands like this in a blender.

People won’t tho and folding this on the button in a live game is prolly worse than calling giving live dynamics.

It's literally not allowing flats in this spot with anything though. It seems to me that this is a choice made by the person running the solve either because of hard limitations or time limitations caused by exponential growth in game trees. They are running these solves for many different stack depths, rake structures. Maybe allowing a 4th cold caller not closing the action in an 8 handed game blows up the computation time from 1 month to 2 months and they deemed that it is just not that useful.

At a certain point, you have to wonder, after 2 people flat, is it really that important to know what the flatting range from a 3rd field caller looks like to solver? Won't it be similar to what the 2nd field caller's range looks like? But then you have to ask how the parameter limitations affects the ranges of those callers since they will get cold called less and 3b more. It probably overestimates the strength of the 2nd field caller's range because unless they get 3b, they end up with position, when in reality there can be multiple callers in position to them. If anything, that might give us an argument for flatting less and 3betting more in the 2nd field caller's position.

At the end of the day, preflop solvers are never going to be perfect for practical application. But they can still be really useful for building a foundation. A much more sound foundation than just playing like everyone else does.

Would you run the same hand but DS and see what the result is?

Its filtered for single suited rundowns. So its a 37.4% flat frequency with ss rundowns and not overall. Overall flat frequency in the spot is much smaller

Also I should add: If you are a paying customer to PloMastermind there are excellent channels (both public and hidden) on discord where the good people of Mastermind answers questions like this. And they obviously know a lot more about the sims they have run than I do.

ds its a pot reraise (4bet)... which is why i was so astonished, that same hand ss is an immediate fold (and not at least a flat call)

ah, didnt know that (just started with mastermind), will do that and ask, thx

more good insight, thx.

and yeah, i thought i had somewhat of a foundation (playing live successfully for 15+ years now), but i wanted to refine that, get some good gto back-up under my belt so to say (and i've never in my life folded QJT9 (even only ss) preflop for any amount ;-)), which made this so hard to believe for me

PS: as for the scenario itself (200BB, multiple limpers/callers):

i wanted to get an idea for a situation that is very similar to the cashgames i usually play, and thats how it goes there (4-5 way preflop is absolute standard)