I'm a few months into a new phase of playing poker. I've previously had a few runs over the years where I played online and live as my primary hobby, generally with modest success. My newest push is going to come with a serious effort to study. I studied hard before, but I'm starting already with a fairly solid base, hoping to go from somewhat good to good to great. I'm going to try to document my thoughts here as I go, because why not!? I invite people to comment and discuss and I will from time to time update with progress, reflections, and results.

I'm going to begin, in this post, with gathering a concise explanation for how I think and approach mastering hard subjects. I have formed and applied these ideas in other arenas, and am now going to apply them to poker. Specifically, my experience as a graduate student in mathematics contributed most to learning how to learn. I've also applied these ideas to chess, my other hobby, where I am a USCF class A player rated in the low 1800s. I use these ideas in my profession, software engineering and data analytics, where keeping up with technology changes puts a premium on continuous learning.

Learning is ultimately a psychological battle with yourself to be honest and fearless enough to identify and discard your faults. It requires tremendous humility, in my experience. I will try to approach my game with brutal honesty, combined with an optimistic faith that the journey is the destination.

I think there are 5 levels/states of learning. These apply at different scales, to individual techniques and to the whole:

• Novice: unconscious incompetence
• Enthusiast: conscious incompetence
• Practiced: conscious competence
• Fluency: unconscious competence
• Mastery: conscious expertise

Cutting across these levels of learning are various approaches to conscious improvement. Most activities we do will fall into one or more of these categories:

• Ideation: cataloging the ideas, concepts, and terminology of a field so you can discuss and think about it efficiently
• Review & Assessment: looking back at your performance and results with a critical eye to find, identify, and quantify improvement opportunities
• Deliberate Practice: repeatedly performing planned activities with the intention of gathering feedback to improve a specific associated skill
• Analytics: using data driven approaches to refine assessment and guide deliberate practice
• Experimentation: trying things to see what happens with an open mind

Ideation is what differentiates enthusiasts from novices, but it's necessary at all levels above. In poker, when we read blogs, watch videos, and interact with each other in discussions on 2p2 or discord, we are learning through ideation mostly. Having the right mental model and vocabulary is simply a requirement for success as a learner. In poker it's even more important than most settings, since as dnegs says "everything you do at the poker table conveys information". Confuse a set with trips or say "I put him on AK" and you've capped your range on poker knowledge.

Occasionally, someone actually posts a hand for discussion and then they are doing the second approach: review and assess, where peers, experts or tools can be used to reason "in situation X you did Y, but the alternative Z was preferable". It seems like a classic 2p2 pattern is someone moaning and groaning about losing due to bad luck and bad beats only to be counseled to post hands so they can actually learn. Learning requires a "no excuses" brutal realism. We must review and assess what we can control so we can improve it. Bitching about things we can't control wastes the most precious resource we have: time. I've been using poker snowie to review my sessions and quickly find mistakes and dubious plays I make across all the hands I play. Flopzilla and solvers are for the deep dives.

In most subjects, deliberate practice is the meat and potatoes of improvement. In some sense, I use online play as practice for live, where the stakes I play will be higher. That isn't entirely fair, because online play matters some to me, just not as much, at least at present. Practice is how competence is honed. I don't believe "practice makes perfect" but rather "practice makes permanent." There is a whole world out there for how to decide what to practice. I recall learning to get better at golf and being amazed at how refined their set of drills are. An expert trainer can make the translation from swing fault found on video review to a drill to solve it. We need more forms of deliberate practice in poker. The ones I see in poker are things like interactive quizzes, "play against the solution" vs a solver, or the kind of hand range quizzers that a few sites have. In poker, we think in terms of leaks. Play, find leaks, practice new behaviors until they are automatic, repeat. When you've done this a little you are practiced. When you've done it a lot, you become fluent.

Analytics, on the other hand, is an area where poker is highly advanced. Playing online and capturing the hands in pokertracker or holdem manager or equivalent is just amazing. I love geeking out on the stats in PT4 and finding and quantifying a leak with the right filter excites me because I know that's the first step towards making graph changing improvements. I recall many years ago (before black friday) working with a coach on my HU SNGs and combining hand history reviews with database work, making some changes like playing tighter OOP, barreling some more, and checking instead of betting marginal hands. Each of them saw my graph take an angle up. This is where deep improvement comes from. When you know what most of your leaks are and can help others hone in on theirs (even ones you never had) that's when mastery is at hand.

Experimentation doesn't necessarily get you from any level to the next, but it's how people come up with new ideas. What does the solver do if you offer it a 2X overbet in this spot. How do tournament players react if I raise 5X occasionally when I steal instead of the standard size? How do LAGs handle it if I take this weird line. These kinds of questions may not matter for honing best play in standard spots, but the thought process required to answer them is a critical component of mastery. Playing around with things, solving them with the tools you've learned through ideation and constantly looking for new exploits is the hallmark of poker greatness, I think.

Plan

With that as a backdrop, here's how I plan to improve at poker:

I'm going to try to balance play and study. Something in the 50-50 or 60-40 range seems about right for a while. Study will be a mix of ideation, review, analytics for now. I'll try to incorporate deliberate practice, but that has to be driven by what I'm trying to improve and the best way to practice it.

I will study content on a couple training sites I have and with books. I'll deep dive into them in follow up posts in this thread maybe. My primary focus right now is on 6max cash, but I'm also feeling the bug to get back into tournaments. I like some of the "How to Study Poker" content from Sky Matsuhashi and may use some of his techniques for feedback and review.

I'm reviewing my play most often with poker snowie. I just try to put all my hands in and have it highly every spot where it thinks I got something wrong. I nearly always agree with it, but the few places where I don't are often interesting too. I have a discord group where I post hands from time to time to discuss. When a hand that I or someone else played is particularly interesting, I'll break out flopzilla and sometimes my solver GTO+. At some point, I want to do my own systematic exploration of flop textures with GTO+.

I spend a good amount of time looking for my leaks in my database. At the moment, I seem to have some issues in 4bet pots generally. I have some issues when I've called a 3bet in position. I need to work on my ranges facing raises preflop. I need to improve my blind vs blind play. In all of these areas, I want to refine my exploitive adjustments, so that my approach is to start with GTO as best I can replicate it, but to deviate deliberatively by following planned changes from my exploitive bag of tricks when conditions are right.

I'd really like to expand my exploitive poker bag of tricks, too. This is an area where I'm searching for a way to do deliberate practice. I'm pretty good at identifying player types quickly when I play online with a HUD. It's harder but I do OK with this live, too. But I want more weapons in my arsenal, so that I can go from observing a non-GTO pattern repeating to putting a hook into my thought process where I say "but, this guy does X too much, so I'm not going to the standard play Y, but instead I will do Z".

I never thought that there is a next level after "unconscious competence". Can you explain what is your concept of "conscious expertise"? I'm curious to know about it.

good luck

It's the difference between a native speaker of English and the editor of a major publication. Both are fluent, but the editor has supplemented fluency with deep and complete knowledge through comprehensive study. Both can express their ideas without effort, but the editor will find more efficient ways to do so and can guide others, even those who are accomplished, on how to do so. Mastery means you know "why" things are better and can explain it.

That's my answer and I'm sticking to it, but really, this is all made up, so if some people merge fluency and mastery, that's OK too. I find it useful to subdivide the category, because there are subjects where I feel fluent, but am very aware that there is another level above. I didn't invent this idea either. If you google "fifth stage of learning", you'll find it's a common augmentation of the original four stage model. Many define it in terms of a reflective ability over the other stages. The cynic in me thinks that teacher's unions added this so they'd put instructors on a pedestal. I actually think of researchers and innovators at the top of the learning hierarchy. For me, mastery is the ability to learn things that can't be taught yet.

For a poker example, think about when overbetting started being a thing. Novices didn't realize anything special was happening. The enthusiasts and practiced players scoffed because "educated" poker players didn't bet more than pot and strange plays must be mistakes. They often reacted badly and were confused. People who were fluent quickly became able to interpret this and adapt to it. People with mastery were the ones who introduced it, or having seen it were able to figure out what conditions make it work, and proactively incorporate it into their games.

I've been studying with a buddy who is profitable at higher stakes than I play and makes me think a lot about why I do the things I do. He sweated me in a session and commented that I'm not check/raising enough, so I've started to do a deep dive into check raising, generally. He suggested I start with c/r lines when I defend the BB. So I'm looking at BB vs BTN in a single raised pot. It makes sense that since BTN is often opening pretty wide here and we get a subsidized call preflop due to the blind, that we should be working to take the pot away after a cbet by using a balanced polarized range of check-raises.

I looked at some different solves for bb vs btn single raised pots and was pretty amazed at what a balanced c/r range really looks like. There are a good amount of bluffs, probably more than the pure value 2 pair and sets type of hands. This solves the biggest concern I've always had with check raising, which is that I worry if I c/r my big hands, my opponents will just fold and I'll lose value with them and my call range will also be weaker. But the solver is taking hands I'd otherwise give up on and balancing them with the monsters, so if you fold, you are rewarding my gutshot+bd flush draw hands, but if you call, I'm going to smack your pairs around a lot. I definitely think this is a play that will up my EV and also can let me make some money OOP with more hands I normally wouldn't. Since villian has to call more to avoid being run over by my bluffs, I'm going to get paid more by my monsters in the inflated pots.

Great work. I've been meaning to study this spot for ages but never got around to it. I was going to start from BB vs UTG and work down to the SB. I feel like my BBvBTN xr% is decent enough although I'm sure I could always use some brushing up on the subject. Did you use gto+ to study this?

Yes, I haven't invested the big bucks in Pio, so I use GTO+ as the affordable option. I'm over the initial learning curve, but not yet a power user.

I started with BB vs BTN because it's pretty common and it's a position where I can improve a lot. The BB generally loses money, because being forced to put in money when you'll be OOP against all but the SB is a structure EV sink. But the BTN is the attacker with the widest range, so it's the easiest opportunity to fight back. The cutoff is what I'll study next.

BTW, @BobbyPeru, (or anyone else) if you are interested in this spot, dm me and I'll send you an invite to a discord server for micro and small stakes cash I like a lot where people like to deep dive into spots like this.

I'm lucky enough to live in Texas, where live poker is a new phenomenon and the games are very juicy. I get most of my volume online, but I expect to monetize the poker skill I develop most in live play.

Last night I played live $1/$2 at Rounders in San Antonio, which is my regular room. I had a mildly winning session (up $187 from a $400 buy in). Late in the session though, I folded an open ended straight draw to a large turn bet after having chased one street for a small bet. I calculated the pot odds and didn't have enough to justify the call. My straight would've hit on the river and it would've been well disguised (QT on J97xx board). On reflection, it was a place where if I'd considered the implied odds correctly, I realize I should've called. If I miss my 8 outs, I'm out the turn bet only but don't have to put any more in. It'd be wrong to bluff the missed draw because my opponent liked his hand too much. In $1/$2 in Texas, people call way too much, so I don't bluff missed draws much unless I have reason to believe my specific opponent can find a fold, which many cannot. This was a typical 4-6 people seeing each flop and calling down with top pair or better.

But it escaped my reasoning in the moment that these are exactly the conditions that make the implied odds better. If I hit my draw, and could get him to put in the same size bet, I would've had odds. There were no flush draws, so I expect he had a made hand. He would've been all-in for about double his turn bet and maybe that would be an interesting decision in maximal extraction.

Because I probably misplayed this spot, I'm going to do some work on implied odds spots. Per the learning state levels in my first post, it appears I'm practiced in implied odds, but not fluent, and it cost me a few hundred dollars last night because I botched a spot when I was a bit tired near the end of a session and folded when I should've called. This is such a fundamental skill, so it's important for me to be honest with myself when I make a mistake with it and restudy it carefully so I can have correct thinking be more automatic.

After further review, my fold was correct. There just wasn't enough left in my opponents stack to justify a call even if I win all of it every time I hit my 8 outs.

I got a copy of How To Study Poker by Sky Matsuhashi. He uses a similar approach to what I present above as to the states of learning. This is (as advertised by the title) a meta-book. It's not really about poker, it's about how to study poker. I'm convinced that poker improvement is defined most by the methodology you use to seek it.

I'd say 80% of poker players don't put any serious effort in to improve. They learn "on the job" at the tables and are fitting their play to what works in that context without any theoretical framework for it. Maybe 20% of poker players study at some level, but probably 75% of these jump around haphazardly and are basically "enthusiasts" (defined in the first post). They talk the language, but make standard mistakes and don't realize it and aren't doing what's required to systematically find their leaks and even if they found one, they might study it for a while and become "practiced". While they are thinking about it, they'll get it right, but eventually they stop thinking about it and the leak creeps back into their game.

When you find a leak, you need to resolve to destroy it. Not merely to learn the right way, but to rework this portion of your game entirely so that you change your intuition, your habits. You want to know the right way so thoroughly you could doing three other things and so that if somebody made the wrong play your reaction would be "What are you doing? That's wrong!"

When I play online 6max, my vpip/pfr is usually something like 29/25 or maybe 32/27 if I have nits to my left whose blinds I can steal liberally. My opening stats are dominated by my RFI ranges, which are based on GTO, plus some exploitive plays: I raise when no one will 3bet. I've been playing live some lately. It's a very loose game $1/$2 game, with the majority of the people at the table loose passive or loose aggressive. Because of this, I basically don't get the chance to open in late position in these tables. Usually, if I'm in the cutoff, say, I'm facing two to four people who put money in the pot when it gets to me and unless the pot is 3bet, I'll expect one or two more to join behind.

I generally set my range based on the seat of the raiser. If it's limped to me I open with the range I would if I'm first in. If there's a raise and some callers, I'm mostly thinking about squeezing with the same range as if the raiser is the only player. I will call a little more with lower pockets and mid/low suited connectors, but for most seats, I'm mostly making raise or fold decisions. I do sort of expect that although players call much more, the raises still happen with reasonable hands, mostly.

As a result, in terms of vpip, I'm trading my late and middle position RFI ranges for my 3betting ranges and this makes my vpip go way down and I feel like I'm nitting it up. Because more people see the flop, I win fewer pots when I play, but the ones I win are larger. I don't win as many pots with bluffs because I'm the PFR only when I 3bet and also because these players call too much, so I focus on value betting. My strategy is pretty simple: play good starting hands, hit made hands or strong draws, build the pot, make a big hand and bet.

My range advantages and ability to play post-flop, generally mean I'm profitable in these games, but I just can't help to think I'm leaving money on the table. My intiution is I should be seeing more flops than I am, so that I'm not at such a risk of being card dead for long stretches and feeling pressure to make something happen when I finally get a preflop hand to play.

So with that as context, how do you play when facing a 2bet and some callers in very loose live games?

A lot of variables like who raised, stack sizes, position, who is on my right and left, what position I am in, also my own hand, before I can decide what I am going to do. Sometimes I will be in the same spot and do something completely different based on some information I got from past hand. I read blogs and want to answer poker questions. But I can't come up with an answer. Its hard to answer poker questions cause of so many variables involved. You seem like a smart gto solver player. Just do what a solver would do. That would probably be the best way to go about it. I don't use solvers so I am not sure best way to go about it. I figure it out as I play to see how they respond to 3b, iso raises, and limps. Maybe try different strats to see what works best as you play. Cool blog. I like how you go deep into solving poker. Will be interesting to follow your grind. GL grinding.

Sounds like a interesting book to read. I am intrigued.

I dont often browse this sub but when i do a gold like this suddenly appears.

some books that have helped me. each targeting somewhat specific concepts

Metal Edge / really just a real good book abou5 time management.
Poker Therapy / i learned to see and understand many underlying emotional problems that occur.
Exploitative Play in live Poker /how to manipulate your opponents into making mistakes

last week started reading Gambling and other theories.

honorable mention would me Limitless by Jim Kwik.

all of those books made me bit better at poker but much better at lifewise approaches.

glgl, subbed.

Thanks! You're motivating me to keep trying to put good content here.

I got the kindle version of Exploitative Play in live Poker a few weeks ago. I really like Alex Fitzgerald. I'm looking forward to working my way through this book. Although, I kind of think most of his poker advice applies when you can actually play poker as opposed to $1/$2 in Texas, which is more like bomb pots with a range advantage. I laughed when I wrote that, but that's exactly what it is, now that I think about it.

Ten years ago I would have brushed aside a book like Poker Therapy, but after realizing that variance does not give a damn what I think I'm entitled to or how bad it hurts when it takes my toys away, I've come to appreciate things like this. In fact, I think the top poker players in the post-GTO poker world will be differentiated by their mental awareness.

Limitless by Kwik looks very interesting.

I just watched an interesting video from splitsuit called 3BET RANGES: Hand Reading & Beating Them.

Here's the short summary. In a live $1/$2 game with 100bb stacks, he opened AQs UTG to $6, was called by UTG+1, and then the BTN squeezed to $24 or something. He 4bet jammed. He showed the GTO ranges, which wouldn't jam, but reasonably asserted he didn't expect a $1/$2 button player to squeeze with GTO ranges. He thought they squeezed with standard nutted hands (KK+ AK and half of QQ) and some lite hands, which he put at A9-2s K9s, 2-gappers T7s-74s and A2-A5. Some of these were half weighted.

I'm pretty skeptical of this range. At best it's a wild guess and my wild guess would be different. I just don't think random $1/$2 players are squeezing with 74s and A4o. That's a pretty sophisticated idea to use your weaker connectors and A-rag because it blocks aces, whereas I think most $1/$2 players would use their stronger suited connectors and some of the standard lite suited aces (A5s A4s A3s) I also think people will do this with some of the better broadway suited connectors.

My guess is the squeezer's range is something like KK+ AK+ 1/2 QQ AJs+ KJs+ QJs and A5-3s and JTs - 98s. I ran some numbers and AQs is probably a profitable 4bet against this range. But he ignores the other really interesting thing here: it's a profitable 4bet, but is it best to 4bet-jam? This is where I'm not on the same page as him at all.

It seems like if we find some range where villian has enough lite stuff to justify a 4bet, we have to recognize the 4bet is only true because we're exploiting their crummy ability to balance this properly with their nutted hands. Those nutted hands autocall our 4bet jam and crush us from an equity viewpoint. The crappy stuff mostly gives us fold equity and this is what makes our AQs 4bet possibly profitable. Said differently: AQs is not a nutted 4bet hand either in the core value range and it can fold to a 5bet-jam and save us some EV against villians value range. This comes at the cost of play it vs villians lite range. The only hands we worry about calling are some AK, and some QQ. We flip against any lower pair, we dominate suited aces, and QJs and any other stuff like T9s or KJs is just fine to play postflop against. We can just play the 4bet pot with the range advantage and initiative. We give up position, but if we jam postflop, that's gone.

In summary, I like 4betting AQs here to exploit a $1/$2 players poorly constructed lite 3betting lite ranges, which are really all over the map and impossible to accurately guess. The video title tries to leave us with the idea we can somehow read this situation, but meh. Even if I like 4betting AQs here, I don't like 4bet jamming it at all, as it makes our opponent play a little too perfectly and loses the maximum to his nutted value range.

I hadn't heard of this book before now. Have you had a chance to start in on it yet? Any thoughts so far if so?

Subbed & good luck!

I've had it for a while, but rediscovered it. I think the biggest thing it taught me is that it's best to pick a skill you need to improve using data, really study it hard, and don't stop until you see the data change. This may sound simple, but it's easy to brush off. The way not to do this is to skip around, read a random blog because it found its way to you, watch a new video just because, pick up a random book, and just generally consume poker content without staying focused on a plan.

He has a couple great techniques to help stay focussed. He recommends having a weekly learning plan. So you are explicit with what you are working on and why. He uses a concept called FOCUS sessions. FOCUS stands for Follow One Course Until Success. The idea is that you play with a specific learning idea in mind that you want to apply. Maybe the idea is to check raise bluff. So play a session and look for spots to demonstrate the idea. At the end, go back and evaluate how you did. Not whether you did it, but how it worked and why.

Another technique I've done a couple times that I want to do more of is game tape. The idea is to record yourself. Capture audio and screen. Maybe video of yourself. While you are playing, explain your thought process. Then go back and watch the video and evaluate how you did things. Check your assumption. Look for important things you missed. See how well you articulated your reasons. Take several passes through reviewing it: first for your own play, then for your opponents and how well you understood them and how you used the available information about them. The last thing stood out at me the times I've done this. I'd see myself bluff at somebody with a really high WTSD and I'd be like "No!!" or I'd call some loose passive player's big river bet because I'd convince myself he was bluffing, when his whole reason to exist in this universe is to river unlikely straights over a set or two pairs and get paid.

He has a couple chapters on using your HUD and database effectively, and on using flopzilla to learn postflop play. It's good stuff. The book is short. It's an easy read, but it really comes down to one main idea which is to be organized and systematic in studying. Pick the most impactful thing, improve it systematically, prove it's better, then repeat. Easy game.

Cool. Yeah, I've used a 'topic of the week' type plan in the past, though I haven't been very disciplined with it lately.

The 'game tape' technique sounds interesting. I'll have to try it.

Thanks for the info on the book! I'll check it out.

You should check out his podcast. Here's an episode on study techniques. It includes some info on game tape. smartpokerstudy.com

Here's a podcast on game tape, specifically.

Thanks! I'll be sure to check it out.

Ah, that feeling when you've got the highest possible hand in poker and your opponent shoves the river.

So I'm building my bankroll up to take a shot at 25NL.

I have some BR rules I follow that I think set me up for success and will keep me out of trouble in the inevitable downswing. Here are my rules:
- I always have, two stake levels I'm playing, a higher one and a lower one. I switch back and forth and try to get some volume in both.
- When my bankroll reaches a fixed threshold number N of buy-ins for the next level above, the two stake levels change.
- I can only put money onto a table if I have a bankroll that is more than N, so if I lose that buy in, I can't play the level any more
- Only money on the site where I will play counts.

For example, right now I am playing 5NL and 10NL on ignition, and I've set my threshold at N=16, so that when I have $400 I can make a $25 buy in at $400 but if I lose it, I can't rebuy. When I hit $400, my top level becomes 25NL and my lower level becomes 10NL. I will still play both each month and strive to be profitable at each.

What this does is constantly has me switching up levels, so that I'm comparing my win rates at each. It gets me out of the mindset of "going back down" since I'm playing multiple levels regularly either way. The rule makes it impossible for me to go busto by variance, unless I'm forced all the way down to the lowest level. I generally want to optimize my win rate and not care what stake I play at, so if I'm most profitable at the lower level, I'll end up playing more there naturally for a while, but eventually I get more comfortable at the higher level, get my win rate up there and it becomes more profitable. This is when my BR usually heads up to the next threshold.

I've taken one shot at 25NL so far. I ran a big bluff and got picked off and am trying to get above the $400 again for my next try, which should come hopefully this weekend. Wish me luck.

GL with the study and improvement.

In my experience, most live 3b ranges skew towards being overly tight like QQ+ AK. I wouldn't even consider 4b AQs without seeing the squeezer showdown a light holding or 3bing 3 or 4 times in an hr.

I believe Ace wheel suited hands are slightly better to 4b light in general as the straight cards will unblocked when called.

Yeah, 4 betting AQs is an exploitive play. That means we should have some read to justify it, like the ones you mentioned. Now that I think about it, this AQs is a hand at the top of our range for calling a 3bet, so I think we'd want to 4bet it against somebody we thought called 4bet jams too much and not somebody whose polarized range has too many light holdings.