My thinking is that microstakes players are bad. Like, really bad. Even the "good" regs are bad. Sure we can aim to play as close to GTO as we can and we'll make money, but given that there are certain exploitable tendencies that are present in the vast majority of your opponents, I can see an argument for defaulting to a slight deviation from GTO against an unknown microstakes player until you're given enough info to do otherwise.

Some ideas off the top of my head:

• Your average microstakes opponent is bad at folding. So we bet more with our value hands than solver says, and bluff less.
  
• Your average microstakes opponent is too passive. So we fold more bluff catchers than solver says (when when facing aggression), and try & induce less with our value hands.
  
• Your average microstakes opponent doesn't like folding preflop. So we adjust by playing more hands preflop, especially when in position.
  
• Your average microstakes opponent 3bets preflop way too little. So to adjust we call 3bets & 4bet less often than solver says
  
• Your average microstakes opponent calls 3bets pre too much, even out of position. So we reduce our light 3betting range and increase our value 3betting range.

Just wondering if this is a thing any players actively do? Like, sit down at a 10nl table and already have some idea of the ways in which they'll deviate from a perfect GTO strat. Obviously if an opponent gives you reason to think that they're competent and might even be capable of exploiting your deviations from GTO, you would tone it down vs them and play as close to GTO as you can.

Ironically, the strategy you outline would end up having you play like most microstakes regs. In order to exploit them, try to exploit what you came up with.
As a general rule that's kind of funny/paradoxycal, whenever you feel you are 1 level above your pool, you're usually exactly at your pools level. Similar to how everyone thinks they're smarter than the average.

Even this is too simplistic though. Microstakes players tend to way overfold certain spots and not fold enough in others, for instance. Or even call 3bets too much in some situations but not nearly enough in others.

Nit it up until you know what kind of player they are then adjust rapidly (assuming regular tables and a pool similar to PS). You can of course do what aner0 is suggesting but imo there is no need for that 'cause the unknown doesn't stay unknown for long so you can rapidly play them as an individual (of course this will be harder if you play a lot of tables).

In a sense, deviating from GTO isn't really 'deviating'. It's just applying the principles of GTO to better and updated perameters. It's deviating in that it eschews standard GTO v GTO output, but even a **** GTO exploit *feels* crusher is still incidentally applying GTO to max exploit all of villain's accidental deviations.

Do you have a solver or access to one? Try playing with nodelocking. That'll give you some ideas of how ranges should be adapting to each others' mistakes. To use MDA/exploits effectively you kind of need a baseline understanding of GTO and how solvers are using pot odds to exploit. A long list of juicy 10nl spots one player might compile wouldn't be worth sharing because they'll be a) misapplied - given much of the understanding of how to deploy them would require theory and bare minimum understanding of MDA/pot odds/gto etc and b) instantly of less value to the person sharing them.

Exploits are usually shared quite freely on here when relevant to the hand or context though because in that scenario b) doesn't really apply given only dedicated players will pick up on them and we're not really optimising against each other so it doesn't matter.

At micros your hardest opponent is the rake. Nit it up pre--avoid just calling like the plague--and value bet thinly across multiple streets.

Sent from my SM-A146U using Tapatalk

🤯🤯🤯🤯🤯🤯🤯

I have the starter GTO wizard plan, I think that includes nodelocking. I'll look into it, thank you!

After playing all of yesterday while referencing the solver after every spot that wasn't super obvious, I think you all are right that much deviation without reads isn't really necessary or productive. The only default adjustment I think I'll make moving forward is value betting wider and more often.

Solver wanted me to just check down flop and turn on 9xxx with A9 after opening pre then check/fold on a Q river! Normally I'd be going for 3 streets of value there against a random 5nl opponent and I think I'll continue to do that.

Yeah if you can make a good case for it follow your logic. Solvers are not arbiters of the truth, merely an idealised p.o.v. and 99.9% of the hands we play at micros are so so far away from that it's very often worse EV play.

Add to that the unrealistic ranges we feed into them (creating lopsided nut distributions) and you can see how unrealistic they start to become.

But the trends and ideas are still quintessential and make us better so study always beneficial. The magic is when you start to understand what they're doing and why which leads to internalising/incorporating into decision making.

I found Andrew Brokos' Play Optimal Poker was my gateway into understanding the bare-bones of GTO and what's going on. Relatively easy to understand. Even if you only do the first few chapters, it does a nice job of explaining indifference and why it's the source of EV/profit. Once I got that locked in solvers made sense.

An exercise that helped me out was mapping out the game tree into manageable chunks to study.

I just counted spots in my database and how often they occur. Then I grouped certain spots together that are similar (e.g. UTG v BB and HJ v BB) and kept whitling it down into about 6-10 categories that cover the whole game. IP/OOP and SRP/3b/4b etc. Found it much more effective to study spots within subcategories and keep group insights together to see if they correlate, and easier to remember of course.

Yes, this is exactly what I'm doing! Usually accompanied by a bit of Googling if I can't figure out exactly why the solver is doing certain things in certain spots. E.g. yesterday watched a video on how to decide whether to cbet (and what sizing) when you open IP and get flatted from BB. That gave me some starting categories (e.g. single broadway flops, high paired boards etc) and then I could take those categories and go plug example boards into GTOw to compare our range to opponent's range and each player's equity and use that information to try and understand why we're cbetting certain boards more than others, and certain boards bigger than others.

It's absolutely fascinating, I was a mediocre mass-tabling rakeback pro back in the day and have just come back to poker recently, so relearning the game from the ground-up with all the new developments in GTO is just sooooooooo interesting to me.

yeah it's great innit. I'm a visual learner with a questionable relationship to math so the modern era/approach is a godsend

Man this by a mile!

This is what flopzilla did for me pre solver days