Midlife Crisis? What Midlife Crisis?

Please tell us more about "almost there".
You already have features that PioSolver is lacking, see http://forumserver.twoplustwo.com/167/commercial-software/piosolver-postflop-equilibrium-solver-holdem-1516187/index21.html...
Is everything crucial in C++?
Do you have any commercial plans?

Now with these great sofwares, how to translate what we learn in our games, how to remember some things as don't value second pair 50bb deep in such and such conditions? Bet more if you have backdoor flush draws...

Originally posted by CarrollJeff
Please tell us more about "almost there".

Well, I just need to get away from MATLAB, tidy up a few bits and bobs, add small saves and on-the-fly river calcs.

You already have features that PioSolver is lacking, see http://forumserver.twoplustwo.com/167/commercial-software/piosolver-postflop-equilibrium-solver-holdem-1516187/index21.html...

I have ICM, rake and equity plots. I'm sure Piosolver will have all those very soon.

Is everything crucial in C++?

It's C, not C++, and I can't use the best compiler until I get away from MATLAB. There may also be some other overheads that I don't know about.

Do you have any commercial plans?

I don't have a commercial licence for MATLAB, so I can never make any significant $$ as it stands. Also, being in the Piosolver Skype group has put me off the idea of ever getting involved with customer support. I don't have the time, and especially not the patience. Maybe I'll open source it, or release it for free one day. It's still not as good as Piosolver, which is only going to get better.

Now with these great sofwares, how to translate what we learn in our games, how to remember some things as don't value second pair 50bb deep in such and such conditions? Bet more if you have backdoor flush draws...

Well that's the million dollar question. Not for beginners, that's for sure. I think GTORangeBuilder has the edge when it comes to educational material.

When are you coming to Vienna?

Originally posted by EmanuelC16
When are you coming to Vienna?

Vienna is too monochrome and moody for my liking.

Some progress with my code but I will refrain from giving details in order to appear mysterious and enigmatic, i.e. cba to write about it.

Just back from my hols.

I miss playing poker.

I miss posting in here.

I miss talking shit with everybody.

Maybe I should start again.

Well hello there!

I'd talk shit about cricket. But... yeah.

Yeah, I miss your posts, you should post at least occasionally.

Every man and his dog has a solver now. I dont think it's totally the end of the world because I'm pretty sure the majority of consumers either find it too much work or are pretty clueless as to what to take from them. I wonder how much $$ piosolver has brought in. I'd guess it must 50k+ USD? If you saw Mr Piosolver driving a flash new car would you key it in a fit of jealousy?

I've noticed a few things like, some regs who I thought perhaps weren't the best have started to play exceptionally well in a few isolated spots. My guess is that they've studied these isolated spots with solvers. Or perhaps more likely are benefiting from others who have done the work for them (god damn alex's videos at GTORB are way too cheap, giving away all the secret sauce ). So their game is a little like a patchwork quilt. With some patches made of fine silk and others of dirty old rags.

I think I'm going to have to put in some more serious solver study in again after not doing much the last 6 months. If I don't it feels like I'll slip behind in the arms race..

I've been having quite some fun playing this week. I try to start tables, especially at the beginning of a session. To get going when there's little action, I'll try to play with regs and have some fun playing heads up. Just about always they sit out, but this week a guy from Argentina has been sitting me. And my word have I been running like a god verse him. It's unbelievable! He'll have mid set and I'll have top set, I'll have a shoving hand and he'll have his best bluff catcher. He'l have Aces and I'll river a gutshot. Over and over again. Poker is great. Today I think it finally got to him with this hand. I wonder if he sits me tomorrow...

Hero (SB): 105.86 BB
BB: 100 BB

Hero posts SB 0.5 BB, BB posts BB 1 BB

Pre Flop: (pot: 1.5 BB) Hero has 8♥ 8♣

Hero raises to 2.5 BB, BB raises to 8.5 BB, Hero calls 6 BB

Flop: (17 BB, 2 players) 8♦ 7♦ T♣
BB bets 9.5 BB, Hero raises to 25.74 BB, BB calls 16.24 BB

Turn: (68.48 BB, 2 players) T♠
BB checks, Hero bets 25 BB, BB raises to 65.76 BB and is all-in, Hero calls 40.76 BB

River: (200 BB, 2 players) 5♥

Originally posted by jules97
I'd talk shit about cricket. But... yeah.

Yeah, I miss your posts, you should post at least occasionally.

Every man and his dog has a solver now. I dont think it's totally the end of the world because I'm pretty sure the majority of consumers either find it too much work or are pretty clueless as to what to take from them. I wonder how much $$ piosolver has brought in. I'd guess it must 50k+ USD? If you saw Mr Piosolver driving a flash new car would you key it in a fit of jealousy?

I've noticed a few things like, some regs who I thought perhaps weren't the best have started to play exceptionally well in a few isolated spots. My guess is that they've studied these isolated spots with solvers. Or perhaps more likely are benefiting from others who have done the work for them (god damn alex's videos at GTORB are way too cheap, giving away all the secret sauce ). So their game is a little like a patchwork quilt. With some patches made of fine silk and others of dirty old rags.

I think I'm going to have to put in some more serious solver study in again after not doing much the last 6 months. If I don't it feels like I'll slip behind in the arms race..

I've been having quite some fun playing this week. I try to start tables, especially at the beginning of a session. To get going when there's little action, I'll try to play with regs and have some fun playing heads up. Just about always they sit out, but this week a guy from Argentina has been sitting me. And my word have I been running like a god verse him. It's unbelievable! He'll have mid set and I'll have top set, I'll have a shoving hand and he'll have his best bluff catcher. He'l have Aces and I'll river a gutshot. Over and over again. Poker is great. Today I think it finally got to him with this hand. I wonder if he sits me tomorrow...

Hero (SB): 105.86 BB
BB: 100 BB

Hero posts SB 0.5 BB, BB posts BB 1 BB

Pre Flop: (pot: 1.5 BB) Hero has 8♥ 8♣

Hero raises to 2.5 BB, BB raises to 8.5 BB, Hero calls 6 BB

Flop: (17 BB, 2 players) 8♦ 7♦ T♣
BB bets 9.5 BB, Hero raises to 25.74 BB, BB calls 16.24 BB

Turn: (68.48 BB, 2 players) T♠
BB checks, Hero bets 25 BB, BB raises to 65.76 BB and is all-in, Hero calls 40.76 BB

River: (200 BB, 2 players) 5♥

Spoiler

BB shows 3♦ 7♥ (Two Pair, Tens and Sevens)
(Pre 13%, Flop 5%, Turn 0%)
Hero shows 8♥ 8♣ (Full House, Eights full of Tens)
(Pre 87%, Flop 95%, Turn 100%)
Hero wins 199 BB

I only watched the highlights of the Test matches we won, so we had a great summer as far as I was concerned.

I'm not really jealous of Piosolver Piotr, because I'm in the PioSolver Skype group and, if I was him, the customer support that everyone would have received would have been along the lines of 'Why don't you all fuck off and leave me alone?!'.

I have other plans for my software. You should add me on Skype since we're allowed to put it in our signatures now. It was Top Secret before.

Yes, the GTORangeBuilder training material is awesome. PokerStrategy should look in to partnering with GTORB somehow.

Can a deep knowledge of game theory help one to flop lots of sets? I certainly hope so.

Originally posted by jbpatzer
Can a deep knowledge of game theory help one to flop lots of sets? I certainly hope so.

A deep knowledge of game theory won't help you flop more sets than you would otherwise.
What it will do is help you recognize the flops you can represent a set. Which of course is all of them on which you don't actually flop one. You will also recognize that when you don't flop a set, and don't represent one, you are bluffing that you're trapping.

Best of luck,
VS

Seems like the place for the cool kids to hang out :f_drink:.

As the beginning of term looms (see earlier third week of September posts in this blog for my views on the return of the undergraduates), I thought I might cheer myself up by posting something. This is always a miserable time of year, but the good news is that I have fully converted my code into C and have found someone who wants to use it commercially (no I'm not going to tell you who, so don't ask). I just have a few more finishing touches to make and it'll be done. It's been work on this that has stopped me playing or posting for a while. It's been a bit of a slog tbh, and maybe writing a GTO solver is not the ideal first ever code to write in C (I skipped the 'Hello world!' stage), but it's worked out OK. I've also decided that trying to grind out some $$$ online is not for me. I am a recreational player (or 'fish' as I believe you poker players call us), and I want to play live. I'm therefore going to take my online BR and top it up with some $$$ earned from my code and do just that. Should be fun to play properly bankrolled.

Family update: Number one son - not prepared to discuss him, but suffice to say he remains still at home, nearly 21, and without a sensible plan for the next 12 months. Number two son - much improved. Looks and sounds frighteningly like me. There's nothing like a mini-me to present you with all your basic character flaws in all their glory. Scary stuff. Daughter - a flawless princess, as ever, and don't try to tell me any different. Mrsjb - long-suffering, both from various illnesses and from having a crap husband. I'm going to try to be nicer to her.

Work trundles along. I'm up to 9 PhD students this Autumn, which is far too many, and I'm hoping that some will be culled, hopefully by viva success, but there are always other options......

Musically speaking, having Spotify has been fantastic recently. It keeps suggesting really awesome stuff that I wouldn't otherwise have listened to. Even though it knows my appalling taste in music, it seems to be able to find stuff from other genres that I enjoy. I think it's tapped into by love of Kate Bush to give me the Norwegian equivalent. This is great, as are all of her albums that I've listened to.

Nice update, shame your not around here often but obv for very good reasons!

Hello everybody. It's me again. I thought I'd check in to tell my three and a half remaining readers that I've finally achieved my ambition of writing a serious maths paper on poker. You can check it out at JB paper . Ever the populist, I'm going to predict that if you enjoy poker but don't like maths, you'll hate it, and if you like maths but don't like poker, you'll hate it too. If you like maths and poker, you should probably get out more, but you can try to enjoy my paper anyway. It's about a toy three player poker game and attempts unsuccessfully to decide who should be doing the bluff catching when somebody bets with a polarized range. It's a real page turner, and there's another one on the way that tackles a bigger three player game.

In other news, I'm now what you might call an infrequent regular in the £1/£1 and £1/£2 cash games at Dusk till Dawn in Nottingham. I manage to go there about three times a month and usually win some money having initially been bankrolled by my GTO solving activities, as described in my last post. I enjoy it and have gotten used to the whole live cash game thing. I'm also really enjoying Doug Polk's Youtube channel. He knows his stuff I think. This one made me laugh a lot at the end.

I'll check back in when I've written another paper.

Nice to get news!
I enjoyed your paper, great hard work!
I did not get how you can change both L and k independently as I thought that a big L (many samples) allowed for a small k (adjusting slowly). That was before I red appendix D. Then I became even more confused when you say you don't consider the cases where player3 is dealt a Q or K as he will always check. But how do player1 and player2 adjust after player3 checks in real game as they don't know if he was dealt a J and checked or was dealt Q or K ? Do they compare check frequency with a 50% checking frequency corresponding to never checking a J or 75% frequency for always checking it?

Do you think a player could have a better estimator if he manages to spot the frequency of opponents estimators, by trying to go into quadrature (or right phase) to exploit the opponent he can maximally exploit?
Do you plan to go further, and add decisions of full Kuhn poker (not algebraically, but using a solver) to unveil equilibrium dynamics nearer to the game of poker?

Originally posted by CarrollJeff

Then I became even more confused when you say you don't consider the cases where player3 is dealt a Q or K as he will always check. But how do player1 and player2 adjust after player3 checks in real game as they don't know if he was dealt a J and checked or was dealt Q or K ? Do they compare check frequency with a 50% checking frequency corresponding to never checking a J or 75% frequency for always checking it?

When Player 3 checks back, there's a showdown, so Players 1 and 2 see the J. Dealing the cases where he checks Q or K doesn't add any information for them.

Do you think a player could have a better estimator if he manages to spot the frequency of opponents estimators, by trying to go into quadrature (or right phase) to exploit the opponent he can maximally exploit?

I suppose if a player keeps track of the other players' estimators, he might be able to exploit this by anticipating his adjusted frequency. I hadn't really given that much thought.

Do you plan to go further, and add decisions of full Kuhn poker (not algebraically, but using a solver) to unveil equilibrium dynamics nearer to the game of poker

I plan to do everything analytically as far as possible. I haven't read much of the literature, but I'm not sure there's an algorithm that reliably finds all equilibria in a multiplayer game. CFR certainly doesn't.

I'm working on full street simplified three player Kuhn poker at the moment. There are 11 strategy parameters and a very complicated solution structure that I can get with either computer algebra or pencil and paper. I'm hoping to be able to do the same for the unsimplified version, but there are 23 strategy parameters. My computer can do 3^11 combinations, but not 3^23, so I need to be a bit smarter and use a combination of my brain and my PC.

Originally posted by jbpatzer
You can check it out at JB paper . Ever the populist, I'm going to predict that if you enjoy poker but don't like maths, you'll hate it, and if you like maths but don't like poker, you'll hate it too.

I guess I hate poker because I don't fancy reading the article after reading the abstract... please tell us: in which year is your work going to kill Spin & Gos in terms of profitability? Should we start being worried right now about transitioning from 3-max games to 4-max ones?

Originally posted by tonypmm
in which year is your work going to kill Spin & Gos in terms of profitability? Should we start being worried right now about transitioning from 3-max games to 4-max ones?

Never! Three player NLHE isn't on my radar. If you take anything from my paper, let it be that three player poker isn't just harder than two player poker, it's a conceptually completely different game. You can't just 'play GTO'. You have to work out what your opponents are doing and try to exploit them (and hope they don't decide to gang up on you).

However, once it gets down to two players DeepstackAI has already killed it I think.

I wrote another paper on a three player toy poker game.

https://arxiv.org/abs/1707.01392

Better than the previous one imo, but then what do I know.

You also get a video from my new favourite band.

I'm off to a school reunion in London this weekend, celebrating the 40th anniversary of starting secondary school (omg I am so old). I think I'm guaranteed to be meeting lots of people with whom ancient enmities run deep and, after all this time, incomprehensible. I expect lots of passive aggression. Should be interesting after a few G&Ts. Hangover notwithstanding, I may also be meeting up with young Emanuel(C16) as he's in London this summer, so I should have a fun weekend, one way or another.

Over and out for now.

JB

JB and Manu together again

Originally posted by DannyG13
JB and Manu together again