João Guimarães's avatar

João Guimarães

52 points

Hey guys, long time no update, sorry for that.

I was talking with a friend about a few mental game themes relevant to me, and she recommended this particular book on Stoicism, which by the way is a philosophy I have had an interest for a very long time, but never went any further than superficial blog posts here and there.

Nov. 28, 2022 | 5:23 p.m.

A bit more about BRM. I think that, in theory, it should be simple. Pick the winrate, standard deviation, go to pokerdope and he'll give me a min bankroll for less than 5% of ruin. I added some margin of safety (it gave me 2440 bbs, which I rounded to 25 bis, added 10 and to the final result added 4 (number of tables I play, as not being able to fully buy-in for all tables is already de facto broke. Thus we have 39, that I simplified to 40.

I wanted this as 25nl is a limit I already know I´m beating, so I don´t want to move down ever again, unless something makes me withdraw the roll again ofc.

To the shots, I thought a little here and how aggressive I can be is more a function of my stop loss, than some actual number of BIs, since I don´t really know if I beat 50nl to begin with. I could even do it now if I´m ok moving down losing, let's say, 3 BIs.

I play my best when I´m not constantly worrying about my roll, and I´m striving to go til Nov, 30th without checking it. This per se should eliminate any possibility of aggressive shot taking, at least right now.

Pros and Cons of being more aggro with the BRM. Pros: 1) I might already be able to beat higher limits, to which point I´m wasting time waiting and hoarding cash. 2) Even if I don´t know 1, the only way to be really sure is trying obv, making a sample and evaluating. 3) Improving faster due to playing tougher regs, which would force me to study harder, which would make me a better player after some time. 4) Playing with too much BIs behind gives a sense of safety that, when it is too much, might turn into complacency. Having some type of pressure over us to not screw up things might be good (but might be a double edge sword also, mental game issues :D ).

Cons: 1) Having to worry about the bankroll while playing. 2) Mental game pressures. 3) Aggressive shot taking seems related, from what I´ve seen in other threads especially on twoplustwo, seems to be related with people having too much this irrational sense of urgency, which leads to recklessness and eventual blew ups. I know I have at least some of these tendencies running deep into my brain, as thoughts about accelerating things jumps into my mind from time to time.

One reason to me thinking about the 50 BIs (40 on, 10 off as a reserve), and 10 BI shots, is that I can actually play sessions without worrying over the bankroll. Sure, -10 or more BI sessions happen (mitigated by the fact I play full ring which is lower variance), but they are unlikely and won´t bust me.

Since downswings happen (I could actually lose 3 bis on session 1, 3 on session 2, 3 on session 3 and then would be playing a final session pretty much underrolled already, which is non-optimal), a better solution imho could be going slightly more aggressive, let's say 40 BIs overall, then moving down at the end of the session if I´m already below this number.

Have to analyze this more, and will deeply appreciate feedback from you and/or the others.

One problem I have with the standard approach like 30 BIs then 5 stop losses etc is that this can´t really be done without some real time watching of the results, which is exactly what I don´t want as it is stressful and worsens my game.

Oct. 27, 2022 | 6:11 p.m.

Up until this year I was really nitting it up without much of a real plan. I struggled a lot moving up from 5 to 10nl and I think this more or less affected my mindset, and brought an almost irrational fear of moving up.

Might seem a bit false, or cheesy maybe, but what people said to me in this exact thread, at the beginning, was the reason I took the risk and moved up, finally, to 25nl. And guess what, I did beat the limit straight away :)

But I was still playing pretty much with 100-120 BIs. At the end of the last month I decided to withdraw most of my roll, which was the reason I grinded 10nl again for a few days. Having 40 BIs now, I´m confident I can play 25 as my main limit and not really worry about busting it.

Not really a plan (maybe that is my problem, lack of structure), but I´m thinking about, from now on, always accumulating 50 BIs for the new limit, keeping 40 online and 10 offline in my account. And moving back down if losing 10 buy-ins at the new limit.

Cheers :)

Oct. 26, 2022 | 10:54 p.m.

Comment | João Guimarães commented on Onklebs 2020

Henry Lister, one of his first live play vids on runitonce, I was reviewing them recently :)

Didn´t mark the exact vid tho, sorry. Just saw this thread pop up again and wanted to point at the irony that I believe OP winrate is considerably better than Henry's. But don´t get me wrong, I like Henry and his vids, otherwise I wouldn´t be rewatching him.

Oct. 26, 2022 | 12:22 p.m.

Not anymore :)

Withdrew the majority of my bankroll and had to rebuild things for a while. Now I have 40 BIs for 25NL again, which by my calculations at full ring keeps risk of ruin very very low.

Being a level I already know I can beat, I can have a safe $ and then play without looking at my results and any worries until I want to do so (put a target for Nov, 30th). Then just work on my game, be disciplined with the hours, and hopefully improve a ton along the way.

Oct. 25, 2022 | 5:54 p.m.

I always knew there is a marked difference in how the games play during the off-peak hours (here in my country, anything from midnight up to 4 PM), and peak, or what I like to call, the happy hour :) But never cared to quantify it, and even considering I sucked (I´m still bad nowadays, but way better than at the beginning), the winrate doubles. At 10nl for example, this year, it goes from slightly less than 5 bb/100 to slightly less than 10 bb/100, and I'm quite lazy with my selection rules in any case.

Why do I play earlier then? Convenience. To adapt to my IRL routine. Still, since I'll usually play the evenings, I could just drop all other sessions, slightly increase the evening volume and get the same result with less physical/mental drainage. Or be more professional and play where my EV is the highest, and stop wasting precious time? Maybe if I want to dedicate it to poker, study in the mornings instead of playing?

Even not being a pro, my approach should be way more professional, so the logistical part of the game is my #1 priority for fixing.

Oct. 25, 2022 | 3:45 p.m.

Comment | João Guimarães commented on Onklebs 2020

Great blog (all 3 of them were great btw). Hope everything's going well.

Was watching an older video here and had to laugh pretty hard when the coach said smth like well, "...this guy's this and that, and this guy (you haha) is just crazy, so..." LOL :D

Oct. 25, 2022 | 3:29 p.m.

Did you guys get a response on this? I fully watched the video before it was removed, thankfully, but what was the issue?

Oct. 25, 2022 | 12:35 a.m.

+1 for Demondoink making videos here.

I don´t necessarily share the same view of you guys about CFPs (about it not being maybe the best option for a player climbing the stakes), as pointed in MatoStar thread, but I have a feeling the drop in quality here may be related to the fact many coaches are either owners or players or related parts of these businesses, due to incentives and sometimes even the good old conflict of interest.

And this might not even be due to some evil purpose, on the contrary, I think most of these guys genuinely want to help, but they never will share the same they share to their players, and we will always have this feeling here.

So, RIO bringing good coaches not related to CFP (my vote goes for OP :) ) would likely improve the quality of the content again.

My 2c, and sorry if I sounded disrespectful with anyone reading this. Not my intention, I was just being 100% honest giving my own personal opinion.

Oct. 24, 2022 | 11:53 p.m.

Not much happening. The "main event" of these last few months was I deciding to withdraw close to my full roll because I freaked out. There are risks in keeping more money than necessary on poker sites, sure, and using a local bank account for this purpose is not always the best idea (foreign exchange volatility due to living in an emerging market country).

Anyway, I redeposited (less), meditated a bit about future plans, then went back to 10nl and will hopefully move up again to 25nl in the end of this month or beginning of november. I plan on keeping online only the # of BIs necessary for less than 5% risk of ruin, then withdraw the excess and keep it in a low cost dollar investment fund in my bank (the equivalent of buying and storing dollar myself without the safety risks of physical currency), which by all means will still be considered my bankroll and completely segregated from my IRL. Only moving funds back online if roll gets short, let's say, 10 bis or less due to big downer.

I toyed around using crypto, or dollar denominated bond ETFs like SHV etc, but all these options add either a market volatility that the bankroll can't really have (crypto haha), or brokerage and slippage costs, less liquidity here, slightly higher cost etc etc.

Lifetime results so far (noticed I´m not posting any progress ITT since the beginning lol):

Oct. 22, 2022 | 10:53 a.m.

A major goal/dream coming true today: first 50nl shot. Wish me luck :)

Aug. 11, 2022 | 8:37 p.m.

Just checked the .it site and this is true for 2nl wow. Such a rip off. Just to put in perspective, 2nl rake is 3.5% with 15 bbs cap. Unfortunately our italian friends don´t have much option unless they want to move to another country :(

Aug. 11, 2022 | 1:47 p.m.

Yes, I made a mistake. If both players check, by my premises, it has to be 0.5, since I put the condition that the 100% rake wouldn´t come into play unless there is an all in bet and the other player calls.

Aug. 9, 2022 | 10:43 p.m.

Think I got a way to model this with GTO+:

OOP range=only KK
IP range=only KK
Eff stacks=100
Board Texture=22223
Only allowed sizing=all in
Both players can bet or check
Rake=10% cap=100bbs

Solution is to OOP always bet and IP always fold

Solution was ran til 0% accuracy (perfect I guess :) ).

Then I nodelock IP to always call when facing a bet.

In the new solution, OOP always check, IP always bet and OOP then always fold.

IP EV=18

But if IP improved its EV (and OOP lost EV) with this deviation from the equilibrium solution, then the first solution can´t really be a nash equilibrium. Thus, GTO+ didn´t really arrive at a nash equilibrium the first time I ran, by the definition.

Last, I lock OOP to always call when facing the IP bet. And, no one bets anymore, it goes x x and everyone wins 9 in EV. Since anyone who deviates from this by betting loses, then this has to be the true equilibrium. But GTO+ can´t arrive to this conclusion.


Aug. 9, 2022 | 10:08 p.m.

This is a bit odd in the sense that the solver must reach a strategy where he has among his options the losing call. But this call will never really happen, because the player to act, knowing it exists, will never really play any strategy leading to this particular node. Can a solver reach this? Not sure, GTO+ doesn´t allow me to model this particular toy game. Maybe PIO does, but I don´t have it yet :(

Aug. 9, 2022 | 8:47 p.m.

Sorry Dr. Maximilious Exploitopoulos MD , it was the first example where the pot was 1 but you wouldn´t be raked by winning it, just if any player bets and the other calls. I should have made it more clear. Btw, I know this is a huge manipulation and very far from reality, it was just made to really get to an extreme situation.

The solver always knows perfectly both strategies, so he always knows the remaining player will call when facing a bet. If I understand correctly, this is what people mean by the clairvoyance principle right?

I think the 6-max table/100% rake/no flop no drop, was a better example, when I first came to the conclusion that the first player to act would always want to shove (he can´t really open a normal sizing, as if he faces a 3bet shove, he can´t call, if he face a normal 3bet, he shoves and the other player can´t call etc).

But, if the equilibrium solution of the solver is that, when seeing any open, from a min raise all the way to a shove, the next player to act will always call, then no one can ever open anything, as the ev of opening anything, even AA, is negative compared to folding.

Aug. 9, 2022 | 8:34 p.m.

So, the solvers strategy is to always call when facing a shove, but he then, being clairvoyant of this fact, will never choose that strategic option and both players will always x for the 0 EV?

This probably solves the "paradox" of the 6-max game being +EV pre rake, as the actual solution will actually be for no player to ever open anything from any position when the rake is extremely high.

No groundbreaking discovery nor any interesting counterintuitive solution coming on, just play tighter the higher the rake, sorry haha! And my brain is melting lol :D

Aug. 9, 2022 | 7:43 p.m.

Just remembered one example that came on the 2p2 thread:

Still 100% rake. Let´s assume IP is nodelocked to always call when facing the shove. The solver being clairvoyant, OOP always checks, IP shoves 100% and OOP now always fold. OOP ev=0, IP ev=1.

But, if IP departed from GTO, how can he be winning not the same, not less, but actually more? Unless the first solution in the OP is not GTO?

"OOP x, IP x: OOP EV=0, IP EV=0.
OOP allin, IP folds: OOP EV=1, IP EV=0.
OOP allin, IP calls: OOP EV=-100, IP EV=-100.
OOP x, IP allin, OOP folds: OOP EV=0, IP EV=1.
OOP x, IP allin, OOP calls: OOP EV=-100, IP EV=-100."

Aug. 9, 2022 | 7:40 p.m.

Just made a thread here thinking about the extreme case :)
Anyway, on a more practical standpoint, as your rake looks similar to 50nl on stars, GTOWizard 50nl preflop solutions (free) might be your best choice?

Aug. 9, 2022 | 5:47 p.m.

I always preferred to have an actual hypothesis before looking at the data, and either prove it or disprove it. I don´t really like the way so many people, even in high level scientific circles, gets the data and, only after, try to formulate some explanation for the results they got. That´s why I came to this toy game with something already in my mind (the main catalyst was actually the assumption our absolute 3bet frequency should be higher with higher rakes, even if RFI ranges are a bit tighter).

Second is to test the usual assumptions about solvers liking to open smaller OOP and bigger from LP, 3bet sizings in general etc.

Aug. 9, 2022 | 5:28 p.m.

Hey Dr. Maximilious Exploitopoulos MD , thank you for your comments. To me, the best way to understand difficult (to me, and right now, maybe in the future all of this will be second nature) concepts, is to either invert or, in this case, think in extremes.

I was actually thinking about preflop since it´s easier to model. For an 100% rake, no flop no drop policy, 6 handed, the solution seems to be UTG shoving 100% range and everyone else always folding. We win 1.5bb every round we´re UTG, and always lose 1.5bb since we never fight when we´re SB or BB, for a total post rake winrate of 0bb/100.

Interestingly to me, if the conclusion is that, if a normal raked game by (real-GTO) bots implies everyone losing eactly the rake, then a 100% raked game is more beatable than a 5% raked game? Or else everyones assumption that (real-GTO) winrate is 0 bb/100 PRE is false, and it should actually be positive pre-rake when everyone in the table is playing (again, the real) GTO? Or maybe it always matches the rake? A little bit of a paradox to me here :)

Aug. 9, 2022 | 5:22 p.m.

0% rake if no one bets, if anyone bets, it´s 100% rake (everyone always lose). Pot is 1, 100 bbs stack, only bet sizing allowed is the all in.

Board is 22223.

OOP range is KK. IP range is also KK.

My solver doesn´t accept absurd rake (GTO+), so I can´t check it.

OOP x, IP x: OOP EV=0, IP EV=0.
OOP allin, IP folds: OOP EV=1, IP EV=0.
OOP allin, IP calls: OOP EV=-100, IP EV=-100.
OOP x, IP allin, OOP folds: OOP EV=0, IP EV=1.
OOP x, IP allin, OOP calls: OOP EV=-100, IP EV=-100.

I believe a solver would always go allin here OOP, as it should always maximize his own EV, and IP would always fold for the same reason. Yet, when this came out on twoplustwo a few months ago, the consensus was that we should keep callin til villain knows we will always call if he bets, until he always checks, and we do too, and the nash equilibrium will always be both players x for 0 EV.

My understanding of the Game Theory topic (the broad one) is pretty much nonexistent, but from what they told us back then, a solver will in fact always bet OOP but this is not the true equilbrium, as it would require the, let´s say, future planning knowledge that GTO+, PIO etc don´t have.

Not sure if I stated correctly, or clearly enough, and if you guys understand what I wrote above. If not, but you´re still interested and wanna give your 2 cents, feel free to ask for clarifications.

What I´m trying to teach myself here with this toy game is if high rake environments really lead to high aggression levels in most nodes, as even optimally, villains need to overfold.


Aug. 9, 2022 | 2:52 p.m.

GLGL Alexandre!

Any particular reason for choosing GG? Like country restrictions or we? Never played there, but from hearing/reading what other players say, the rake is insane and might be detrimental to your progress at the higher stakes.

Aug. 2, 2022 | 10:37 p.m.

Since discussing the seeing vs not seeing $ results with Demondoink in his thread, I´ve been thinking a lot about some persistent mental game issues I have. I´m a competitive guy, I must admit. Very competitive. I feel bad when I lose any game, to the point of resentment (a very good friend pointed how I change after losing a board game we were playing). Don´t get me wrong, I don´t do anything bad, don´t even talk about it, and it pass after a few minutes, but my reaction is pretty obvious to anyone, it affects how I play and I know I´m wrong (this tilt doesn´t affect the "behavior self awareness" part of my brain).

At the same time, I get very excited when I win, to the point of bragging and being annoying. Feelings of invincibility. All of this obviously play a big part in my poker sessions, and is probably one of the most detrimental issues to progress. Interestingly, I only get competitive when I know I´m playing a game vs other human players, and maybe the whole reason I do this in poker is because I always thought of our game as a sport. And it has its advantages of course. We study more. We focus way more on the improving part. On leak finding away from the tables. But still, poker is a random game, where we may win after losing money because we played really well but got unlucky, and we may lose after winning money because we played pretty bad but kept luckboxing the whole session.

To me, the best way to think about the game of poker that I came with after meditating, is to think of it as investing. Ok, it´s a more hands on investing activity as we have to keep clicking buttons and constantly thinking/analyzing spots. But still, we have to study/analyze/improve off the tables, especially as we move up to more efficient markets (higher stakes). But the same short/mid term vagaries and fluctuations apply both to the markets and to poker, and we´re better off just ignoring it and focusing on what matters (develop a strategy that gives us a decent edge, executing it well, and grinding our asses off til the long run).

I´m quite young, and decided to start investing a portion of my net worth in index funds, planning for my retirement in a few decades. I keep adding money regularly, and all goes well, not much volatility or any other "surprise". The indexes over the long run seemed to do reasonably well, close to 10%, and I´m aware the return comes with lots of volatility and bad market conditions, which don´t matter as I´m in for the long haul.

Then, all of a sudden I have a very good month, 20% or so. I feel extremely excited, almost feel like a rich guy. Decide to even spend some extra money, going out to an expensive restaurant to celebrate my success. I forget the 10% yearly long term expectation, after all, I´m getting rich right now.

Then, after some time, comes a big crisis and I lose about 50%. I feel a lot of pain, and consider even moving out of the market, as it´s going to 0 and the world will end very soon.

It´s human nature, and might be one big reason why most people won´t ever succeed as investors over the long run. But, just as with real life investments, I need to approach things differently if I want to succeed in poker.


Aug. 2, 2022 | 3:24 p.m.

emsterdad I´m not Demondoink and still have no experience in 50nl, but 10 to 25nl I already have a relatively big experience. At the softer times the difference is nonexistent imo. If you´re playing on reg-heavy times, then yes, but that should be expected. Important points to notice:

At my games a lot of the regs are constantly moving up or down, so I face a lot of the same faces at 25 that I used to at 10nl. I like to look at some of the more familiar ones and see they are still doing the same.

Too many nits at both stakes (pfr less than 8, vpip less than 12). A bit less at 25, but not uncommon for a 9-max table to have 3 of them, some days I find 4 playing the same table.

3bet picks up quite a bit at 25 compared to 10, so need some mental adjustments if you´re not used to it. I played a lot before the cap and lots of the regs I see were 24 tabling 5 and 10nl back in the day doing just that (3betting a lot as their gameplan lol), so it´s not something new. But someone who started to play after 2019/2 might find it a bit odd. Pretty sure they are still not 3betting close to optimal, just that nowadays 10nl and below they almost don´t 3bet as a bluff at all, and 25nl the regs do it :D

July 20, 2022 | 12:15 p.m.

I´m a big believer of the concept of minimum effective dose. Unless you really love studying, maybe you´re a rec with an intelectual passion for the game but no desire to really play that much, and has some free time. So unless this applies to you, it´s better to do the most rational thing, which probably is the MED. We still have to play right? Or do other more important stuff.

But ofc this dosage is an individual thing, and it´s up to us to test it and find what works for us. The biggest problem in putting so many study hours not worrying about actual quality, is that it most happens with beginners (in the game, or in the studying the game for players not used to do it and only playing) who do it by seeing other people doing without putting actual thinking into it. Or, still worse, doing it as procrastination, to avoid more important stuff, or uncomfortable, like actual playing during a downswing.

Btw, I had this friend in school who played chess, and before important tournaments he would obsessively study more than 10 hours/day for a few weeks. Now he was a real smart kid, the smartest kid at school, quite special, very eccentric, let´s put it :) But it worked for him.


July 18, 2022 | 12:15 a.m.

Hey Demondoink , first of all, this is a subject that really interests me a lot, so thanks for bringing it here and keep the discussion going a bit longer :)

I believe the actual first sentence that triggered my "disagreement" for the lack of a better word, was "sticking your head in the sand". It brought to me a scene from some The Simpsons episode where Homer daydreams about the house being on fire with the children trapped, Marge desperately shouting at him to help, and he literally with his head below the ground as n ostridge :) Which I couldn´t agree as I still believe knowing our results doesn´t accomplish anything related to our technical game, while studying does.

But, I changed my mind and I´m now agreeing with you, because, unfortunately (to me) we aren´t bots and will never be, and, as humans, we know we ran good or bad during a session to a certain extent regardless of the graph, esp if it was a terrible session. So, is it really better to not look at the results, with this deep feeling that we ran bad, we might have lost a lot, etc? A feeling that won´t go away? Because if the goal was to not let it affect our mental (and technical game as consequence), it will not only do it, but carry it for a very long time. While simply confronting the actual reality, maybe allowing us to get a little bit emotional at first, will bring closure. So, the next day, or week if it was a very tilting one, the emotions will have gone and we'll be back in shape. And the constant exposure to this will train our minds to become stronger.

We´re (for some reason lol) playing with a knife and, in the process, end up with a small cut in the skin, in a particularly hairy part, and put a bandage adhesive there. When it heals, we need to take the bandage off. It will hurt a bit regardless of how fast or slow we do it, so we normally do it fast, to have the pain gone as quick as possible.

Cheers, thanks for bringing the subject and sorry for the derail :)

July 13, 2022 | 1:09 p.m.

I don´t think I can entirely agree with this: "50% of poker is about losing. so you better get used to it instead of closing your eyes and pretending that it never happened.".

Yes, we shouldn´t ignore that we´re playing a game where variance can be huge, and we should know we will lose a lot regardless of how good we are. But, what will we accomplish by always checking our results? Will it make us play better? Isn´t the HH's the most important part? So my point of view is that, if a small sample graph won´t tell us anything about how good or bad we´re playing, won´t improve our techincal game, and will either do nothing in the best case, or do harm in the worst case, for our mental game, we shouldn´t look at it, unless we´re just curious or we´re taking an aggressive shot.

My 2 cents :)


July 12, 2022 | 10:26 p.m.

Dr. Maximilious Exploitopoulos MD Thank you very much for the very well thought answer, hugely appreciated.

HawksWin Yeah, very good idea. On average 25nl pool where I´m playing still passive.

I´m trying to get a good grasp on what optimal play is much more to have a good understanding of the equilibrium, than bc I want to apply it vs most (or all) of my current opponents, as what´s incentivized is exactly the simple strategy (100% open to be proven wrong). Not really looking to know how to mount a pseudo GTO strategy in game, it would take a lot of time and study and I´m not really in a hurry anyway (I´m a very patient person), but at least understanding the broad ideas as soon as possible.

There was this old saying that we shouldn´t worry about GTO at the micros bc ppl are deviating so heavily that we would actually be hurting our progress, so instead, look at exploiting them the most. Which is totally true even nowadays, and I believe it won´t really ever change as I can´t see the incentive for ppl that good to be playing the micros instead of moving up at least to smth reasonable. Yet, unless we really want to play the micros forever (which is 100% valid if we´re playing seriously but still as a hobby), blindly following exploits without trying to understand the theory behind them seems a good recipe for hitting a wall eventually.

July 10, 2022 | 3:28 p.m.

Gonna check him, ty very much Kalupso

July 9, 2022 | 7:55 p.m.

Load more uses cookies to give you the best experience. Learn more about our Cookie Policy