Recently I've decided to dedicate myself full time to poker. I'm still a beginner, but I'd like to give my best effort towards it for the foreseeable future. I can post a smallish sample from about a month ago, but sadly I managed to brick my hard drive between this sample and now, so it's not quite updated.




This is on ignition mostly at 25nl, some shot takes at 50. Since this sample, I've definitely made some changes. Fixed my preflop ranges up quite a bit using Snowie, and I've been trying out a bit more 1/3 cbets in position, stuff like that.

However, I find that I have much more difficulty studying reliably, especially without feedback from others. So the goal is simple. I want to do an in depth study of at least one hand a day, for a full year. To that end, I decided to purchase a solver. I went with GTO+ because it seemed to perform as well as PIO and is much cheaper.

I want to focus most of my effort in the beginning on two major spots; SRP IP and 3bet OOP. I'll go ahead and post the parameters I use for my solves. All my preflop ranges are pretty much copied from snowie. I'm willing to change that in the future, it's just easy to implement for now.



It's pretty primitive, but usually I solve for these bet sizes and then if the solver vastly prefers one or two, I'll re-solve. I'm certainly open to the idea that my bet sizes could be incorrect, but the reasoning for the 110 132 150 line is to get all in by the river in a SRP. It actually leads to be a bit more than 200bb but that's ok, I think similarly large bets will perform about the same. Once I've done this solve, sometimes I will node lock myself changing up the strategy slightly. For example, I might node lock a range bet of 1/3 size, and see what opponent is supposed to do in response. If I think my opponent is incapable of doing that, I might take the line exploitatively.

Let me know if theres any huge errors in my process or decision making here. I'm completely open to the idea that I'm a fish, I'll just be glad to be getting some feedback. And I'll be back daily to post my thoughts on a new hand. I have been using the tipton subset as a guide for which flops to study, located in this thread: https://forumserver.twoplustwo.com/1...-game-1408420/

And as a quick followup, because I know people will come in here arguing about exploitative play, my stance is that I would like to learn the theory thoroughly and then I can more easily learn the exploitative side of the game! I don't fault either play style, I just think that's what's gonna work for me!

One problem I definitely have with my strategy parameters is that I have no turn leads! It just seemed so overwhelming up until now, I'll have to work on figuring those out as well.

gl

GL GL

Alright boys! Today is the ladies Bday (Terrible day to start a challenge) but I've managed to get some time in for study. Gonna look at SRP IP, flop of 765r today. Gonna solve using my original params for now and see what the solver thinks about bet sizing. I'm fairly certain I'd prefer to split my range between 2/3 and overbet here, but we'll see what the solver thinks.

I fear turn leading ranges for the BB will remain a mystery for now, one step at a time

So I was wrong about sizing! The solver doesn't like overbets at all. After some consideration, I think it's because the board is too connected. It might put you in a weird spot if you just opt to overbet straights, and also you don't have any particular range advantage vs the BB on this flop.

However, I will say, Snowie range for BB vs Btn is DEFINITELY way different than the average villain at my stakes. Snowie goes for a 3bet any pair above 4s, for example, where as my nitty villains tend not to 3bet JJ even.



Anyways, in this image, light blue is a 1/3 size and purple is a 2/3. The solver likes me to split most of my range between those two sizes, with the exception of going small or checking w/ some hands like AKs.

When ranges are split between two bet sizes like this, do you guys have a system for simplifying it? It reminds me of how Doug Polk always says sometimes call, sometimes raise, sometimes fold in a lot of spots. How does one go about actually emulating that?

Looking at turns: there are some turns where the solver likes to develop an overbet range, like a T. There's also turns where it likes BB leads, like 3, which makes a good amount of sense. Sizing on the turn lead is still a mystery and I didn't tell the solver to choose between multiple options

Turn overbet range makes a lot of sense, it's like weaker Tx that did cbet the flop, some sets, and it mixes in some bluffs with hands like QJo and A9s. Not sure what the logic is there.

I'm gonna have to get off at my wifes adamant request, but I'll keep posting about this hand when I come back

In GL!


Sent from my iPhone using Tapatalk

Hey guys! Sadly I already skipped a few days, but oh well! Patterns take a while to stick. Considering this 567r hand, I was checking out the response that BB should have to a 2/3 cbet. It's pretty straightforward the hands you should be folding, because you have a lot of unsuited overcards that miss. It's a bit strange that you should evidently call all the suited overcards, even if there's no BDFD.

My bigger concern is w/ the check raising range in spots like this. First of all, with this sim I was 3betting 55-77 pre, so I have no sets on this flop. So that makes my check raising value range straights, since I do have 43s pre and 89s. Then it throws in a lot of pair+open enders as well as some pure open endedrs like K8s. I think it's pretty cool that it wants me to go 50/50 with hands like K8s, it makes a lot of sense. If I c/r all hands with an 8 in them i'll end up far overbluffing, but if I only do the pair/draws I'll be underbluffing.


I think I may do a quick resolve and throw those pocket pairs back into my preflop range to see how it would change things.

Imo you should include weight to hands that are on the fringe of a calling range instead of all or nothing. If you do something drastic like remove pairs it’ll dramatically change your sim on the boards with those cards

Ahh ok, I'll try that in the future! Do you recommend very small weights or just leaning towards 50/50 type stuff

In for this.

Hell ya! I was doing some consideration last night, and I think that in my own little thread, I can't expect people to comment about every thought I have. So when I want a specific question answered, I'll go to other parts of the forums and just try to bring what I've learned back to here.

I made a thread, for example, about that 567r flop where the turn comes a T. I remember in the solver it likes me to overbet some of my weaker hands, and someone replied with:

"The basic idea is that on dynamic boards, your opponent is heavily incentivized to not slowplay. This means that if they check or call, they are heavily capped. They almost always have some sort of weak pair or draw.
What this all means is that, if the turn bricks, your TPWK are actually very close to beating most of your opponent's range. However, they are still very vulnerable, your opponent has plenty of draws and pairs that can 2pair.

So they make a lot of sense to overbet, despite not being that strong.
Just like we can attack our opps capped range on the turn, he can do so too after we check the turn, and thus, if we had an invulnerable hand, it would make sense to check it to "protect" our weak capped check range. This is why it's not OBing your flopped straight"

That's pretty interesting to me. I want to go back to my solve and see if the solver ever checks back the nut straight on the flop like he said. I also wonder if being two toned would change this type of thing.

I’d just try to make it look reasonable and not worry too much. For example you have no mid pairs, suited broadway or suited Ax in the bb range but the population does. You want to be able see how the spot works from both perspectives. Just make sure hands that are considered standard calls are in there as well as the borderline ones have some weight

Yah this thought had definitely crossed my mind. I personally am trying to play around snowie ranges, so I think a snowie sim is good. But I totally agree with you that the population is not playing like this at all. My pop probably doesn't even 3bet JJ in BB vs btn that often.

Yesterday I went over to the micro stakes threads and ran the solver for some of the hands. One guy went ahead and 4bet EP vs MP (a bit loose) and then flop came K55.

My conclusion to the hand is that if you think opponent flats AK, you should just check and call any underpair with a spade. However, if you know that opponent will just jam AK, the flop is actually fine for your JJ and the solver likes a range bet.

I think the safer option for sure is to go for the check call line. It's just interesting to see how AK is the entire crux of the hand here

I'm gonna put some effort into this post: https://forumserver.twoplustwo.com/6...-line-1731048/

I find the spot to be super interesting! The question is basically what to do with ATdd, 88, and 66 in a 3bet pot CO vs Btn, board comes 975r. Hero makes a loose call with ATdd facing half pot, and then gets half potted again on Jd turn.

Here's the solve for it:

I thought it was super weird that it came up with raising the 8s. You would think that the combo draw is a way better candidate for it, because check calling with no showdown value seems bad. 88 At least has the possibility of going c/c on river and winning.

Ahh, I realized later that it's because I'm being too strict w/ preflop stuff like was mentioned before in this thread. Adding in some weighted JJ allows us to add in the combo draw as a raise as well.

Very interesting blog!

I like your approach of doing the research first to see how you should be playing each spot optimally before going to exploitative lines.

I see solvers as a tool of finding the "base line" for each spot and scenario and depending on our reads and population tendencies we can add different jams and riffs to that base line ---> exploiting and making our winrate higher. I have PIO basic for these studies and usually do first the gto approach and after that node lock villain´s perceived actions and ranges to see what should I do to exploit and maximize my WR.

Do you study purely 100bb cash games? I play tourneys and occasional cash, so if you are interested studying hands with different effective stack sizes, I have posted hands on my blog.

subbed

We're definitely on the same page. I do the same type of thing with the node locks.

As far as the tournament formats, I think that I will play some live tournaments in 2019 but for the first few months at least I want to study cash hard, to give me better fundamentals for my overall career. I'm honestly more interested in learning some deep stack play cuz that stuff is bananas.

What's the difference between gto+ and pio?

I wouldn't know myself, but this is a post from another thread regarding the topic: "PIO has some better visualization of overall EV and frequencies across run outs and rapid switching between run outs and flops. GTO+ is really quick make small changes to villains strategy on turns and rivers and solve with the changes. GTO+ also has pretty good visualization of how to play range overall while PIO is a little exact combo focused, but PIO also has range explorer. GTO+ works poorly on small monitors. GTO+ is somewhat slow to change run outs and flops compared to PIO and SPF, and it's not quick and easy to change the game tree on later streets without solving everything from the flop again."

It seems to me they perform really similarly, and GTO+ is muchhh cheaper

Very interesting as I just bought GTO+ too. Will be following!