The Mathematics of Poker - Bill Chen; has anyone read it ?

LsDghLsDgh Red Chipper Posts: 8 ✭✭
edited July 20 in General Concepts
I'm thinking of getting Bill Chen's book, The Mathematics of Poker. Reviews on amazon are pretty poor but that mainly seems to be down to people who can't grasp the math. I've just finished a math-based degree so I'm pretty confident that won't be a problem.

Just want to hear from people here that have read it. How'd you find the difficulty level? Did it help out your game at all or was it more just interesting to see some advanced math applied to poker? Does it take long to get through?

I'd imagine there are other books out there that are more accessible and more directly applicable to actual games but I'm interested in it from a purely mathematical point of view as well as for improvement in poker.
My only concern is that I'm fairly new to poker (only been playing 2 months) and that it might be too much too soon. Too much info when I've just gotten to grip with the basics.

Comments

  • BigDaddyBigDaddy Red Chipper Posts: 70 ✭✭
    Is this a late April Fool's joke?
    Ok, I'll bite.
    LSDGH, I can recommend Poker Workbook, Math and Preflop, by our own James Sweeney. Math and poker strategy in one meal. Available now while supplies last!
  • persuadeopersuadeo Red Chipper Posts: 4,308 ✭✭✭✭✭
    I own and read/agonized through much of it before feeling outmatched at too many key points of detail. It was difficult but would not expect you to have the same issue. The knowledge is valuable, yes - the book is the basis and source for many more popular culture GTO breakdown books. GL!
  • LsDghLsDgh Red Chipper Posts: 8 ✭✭
    Sounds like it’ll be a challenge.

    I just picked up James’ 6 max workbook. Maybe I’ll get his math workbook down the line if I like that and feel I could do with ingraining some concepts.
    For now though liking the sound of a deeper more theoretical look at poker math.
  • blindraiseblindraise Red Chipper Posts: 302 ✭✭
    What's the best branch of mathematics for a new student to study if they want to improve their poker game?
  • TheGameKatTheGameKat Posts: 3,654 -
    blindraise wrote: »
    What's the best branch of mathematics for a new student to study if they want to improve their poker game?

    Bayesian statistics.
    Moderation In Moderation
  • TheGameKatTheGameKat Posts: 3,654 -
    Let me expand on that with a little thought experiment.

    You get called for your seat in a 9-handed live game. You notice a friend of yours is the person vacating the seat. She's in a hurry, but as she walks past you on the way to the cage she whispers "half of them are terrible, the other half decent."

    You get your first hand in MP and it's in your opening range, so you raise. Only the SB calls. You've been told that 4 players at the table are terrible, whereas the other 4 are decent. What's the probability your opponent in this hand is terrible?
    Moderation In Moderation
  • kenaceskenaces Red Chipper Posts: 1,503 ✭✭✭✭
    I read MOP when it first came out, and have revisited it from time to time. I had to recruit my girlfriend(math teacher) to understand some of it, and I think there are a few parts neither of us completely figured out. Now that I have solver it might be interesting to go back to it. FYI - the book was written before we had solvers.

    It still has the best section on BR management I have ever read. It isn't that long but understanding CO bankroll, depleting bankroll, and the importance of WR on bankroll was really helpful.

    @LsDgh - I would suggest you buy GTO+ solver software and Modern Poker Theory by Acevedo. There is still plenty of math in that book for you but it is much more a practical application type of book compared to MOP.

  • LsDghLsDgh Red Chipper Posts: 8 ✭✭
    @kenaces cheers for the advice but I've already pulled the trigger and bought it.

    I've heard solvers mentioned but haven't really looked into them. Do up just input a scenario and wait for it to output the frequencies at which you should play for an approximated GTO approach?
    I definitely see why a cursory knowledge and access to GTO solutions would be way more beneficial than taking an in-depth look at the maths behind it.
  • Ori13_TTVOri13_TTV PennsylvaniaRed Chipper Posts: 77 ✭✭
    edited July 21
    LsDgh wrote: »
    Do up just input a scenario and wait for it to output the frequencies at which you should play for an approximated GTO approach?
    thats how a lot of people use solvers however that would be the incorrect way to use solvers.(in my opinion of course.) theres a lot you can do with solvers other than simply discovering the equilibrium play. (i differentiate equilibrium from GTO because in its truest sense they are not the same despite the mass confusion that people seem to think they are the same thing)

    I think a really useful way to use solvers is to build somewhat complex decision trees, and you can manually input what you think your opponent will do at each street and discover the optimal exploits against your opponents line, you can see how your strategy adjusts by trialing different strategies for your opponent, and you can also trial your own strategies and run the EV of that strategy (since solvers often output very complex strategies that humans are not very capable of replicating)

    solvers are much better used for understanding on a conceptual level how a multitude of factors effect each other than for simply clicking solve and trying to replicate the strategy it spits out.
  • persuadeopersuadeo Red Chipper Posts: 4,308 ✭✭✭✭✭
    TheGameKat wrote: »
    Let me expand on that with a little thought experiment.

    You get called for your seat in a 9-handed live game. You notice a friend of yours is the person vacating the seat. She's in a hurry, but as she walks past you on the way to the cage she whispers "half of them are terrible, the other half decent."

    You get your first hand in MP and it's in your opening range, so you raise. Only the SB calls. You've been told that 4 players at the table are terrible, whereas the other 4 are decent. What's the probability your opponent in this hand is terrible?

    Bit of a trick question, but if you give away the salient assumption then the game is no fun, I suppose.
  • TheGameKatTheGameKat Posts: 3,654 -
    persuadeo wrote: »
    TheGameKat wrote: »
    Let me expand on that with a little thought experiment.

    You get called for your seat in a 9-handed live game. You notice a friend of yours is the person vacating the seat. She's in a hurry, but as she walks past you on the way to the cage she whispers "half of them are terrible, the other half decent."

    You get your first hand in MP and it's in your opening range, so you raise. Only the SB calls. You've been told that 4 players at the table are terrible, whereas the other 4 are decent. What's the probability your opponent in this hand is terrible?

    Bit of a trick question, but if you give away the salient assumption then the game is no fun, I suppose.

    I'm trying to squash my reputation for obscurantism by simplifying these things.

    Plus it does illustrate how frequentists and Bayesians approach problems.
    Moderation In Moderation
  • TheGameKatTheGameKat Posts: 3,654 -
    Ori13_TTV wrote: »
    theres a lot you can do with solvers other than simply discovering the equilibrium play. (i differentiate equilibrium from GTO because in its truest sense they are not the same despite the mass confusion that people seem to think they are the same thing)

    Intrigued. Isn't a Nash equilibrium a GTO solution?
    Moderation In Moderation
  • Ori13_TTVOri13_TTV PennsylvaniaRed Chipper Posts: 77 ✭✭
    TheGameKat wrote: »
    Ori13_TTV wrote: »
    theres a lot you can do with solvers other than simply discovering the equilibrium play. (i differentiate equilibrium from GTO because in its truest sense they are not the same despite the mass confusion that people seem to think they are the same thing)

    Intrigued. Isn't a Nash equilibrium a GTO solution?

    equilibrium falls under the umbrella of GTO i think. its like equilibrium solutions are assuming that both players are adjusting perfectly optimally. if one of the players is not adjusting optimally, then playing the equilibrium strategy would be leaving money on the table and therefore not exactly optimal, and exploitative strategy would then be more optimal under the confines of game theory, and therefore the optimally exploitative strategy would be considered GTO in the truer sense of the meaning of GTO
  • BFSkinnerBFSkinner Red Chipper Posts: 118 ✭✭
    Ori13_TTV wrote: »
    therefore the optimally exploitative strategy would be considered GTO in the truer sense of the meaning of GTO


    ???

  • TheGameKatTheGameKat Posts: 3,654 -
    Ori13_TTV wrote: »
    TheGameKat wrote: »
    Ori13_TTV wrote: »
    theres a lot you can do with solvers other than simply discovering the equilibrium play. (i differentiate equilibrium from GTO because in its truest sense they are not the same despite the mass confusion that people seem to think they are the same thing)

    Intrigued. Isn't a Nash equilibrium a GTO solution?

    equilibrium falls under the umbrella of GTO i think. its like equilibrium solutions are assuming that both players are adjusting perfectly optimally. if one of the players is not adjusting optimally, then playing the equilibrium strategy would be leaving money on the table and therefore not exactly optimal, and exploitative strategy would then be more optimal under the confines of game theory, and therefore the optimally exploitative strategy would be considered GTO in the truer sense of the meaning of GTO

    Hmmm. I'd agree that exploitative strategies invariably make more $$$ against non-optimal opponents, but a Nash equilibrium is a GTO solution.
    Moderation In Moderation
  • Ori13_TTVOri13_TTV PennsylvaniaRed Chipper Posts: 77 ✭✭
    edited July 21
    TheGameKat wrote: »
    Ori13_TTV wrote: »
    TheGameKat wrote: »
    Ori13_TTV wrote: »
    theres a lot you can do with solvers other than simply discovering the equilibrium play. (i differentiate equilibrium from GTO because in its truest sense they are not the same despite the mass confusion that people seem to think they are the same thing)

    Intrigued. Isn't a Nash equilibrium a GTO solution?

    equilibrium falls under the umbrella of GTO i think. its like equilibrium solutions are assuming that both players are adjusting perfectly optimally. if one of the players is not adjusting optimally, then playing the equilibrium strategy would be leaving money on the table and therefore not exactly optimal, and exploitative strategy would then be more optimal under the confines of game theory, and therefore the optimally exploitative strategy would be considered GTO in the truer sense of the meaning of GTO

    Hmmm. I'd agree that exploitative strategies invariably make more $$$ against non-optimal opponents, but a Nash equilibrium is a GTO solution.

    theres a vlogcast with matt berkey and nick howard where they get into it


    EDIT:
    i think this might be the podcast i was thinking of rather than the one above, but really both podcasts are extremely deep and interesting conversations about poker and the nature of the game:
  • Ori13_TTVOri13_TTV PennsylvaniaRed Chipper Posts: 77 ✭✭
    edited July 21
    TheGameKat wrote: »
    Ori13_TTV wrote: »
    TheGameKat wrote: »
    Ori13_TTV wrote: »
    theres a lot you can do with solvers other than simply discovering the equilibrium play. (i differentiate equilibrium from GTO because in its truest sense they are not the same despite the mass confusion that people seem to think they are the same thing)

    Intrigued. Isn't a Nash equilibrium a GTO solution?

    equilibrium falls under the umbrella of GTO i think. its like equilibrium solutions are assuming that both players are adjusting perfectly optimally. if one of the players is not adjusting optimally, then playing the equilibrium strategy would be leaving money on the table and therefore not exactly optimal, and exploitative strategy would then be more optimal under the confines of game theory, and therefore the optimally exploitative strategy would be considered GTO in the truer sense of the meaning of GTO

    Hmmm. I'd agree that exploitative strategies invariably make more $$$ against non-optimal opponents, but a Nash equilibrium is a GTO solution.

    to expand on it, i think it comes down to a language problem. in my opinion, 99% of the people who use the term 'GTO' dont understand GTO and misapply the term, so do we use the term in its truer more original meaning, or do we use the term in its more commonly accepted (by people who rarely understand it) evolved definition?
  • TheGameKatTheGameKat Posts: 3,654 -
    edited July 21
    I guess my problem is I used to teach game theory, so inevitably when I use "GTO" I'm referring to the mathematically-rigorous definition, rather than how it is commonly misunderstood.
    Moderation In Moderation
  • LsDghLsDgh Red Chipper Posts: 8 ✭✭
    Ori13_TTV wrote: »
    I think a really useful way to use solvers is to build somewhat complex decision trees, and you can manually input what you think your opponent will do at each street and discover the optimal exploits against your opponents line, you can see how your strategy adjusts by trialing different strategies for your opponent, and you can also trial your own strategies and run the EV of that strategy (since solvers often output very complex strategies that humans are not very capable of replicating)

    That's actually really interesting and seems way more useful.

    As for whether using the solver for exploitative strategy is "game theory optimal" or whether game theory optimal is just nash equilibria, that really just seems to be semantics. I feel that game theory optimal may even be a poker term rather than a mathematical term.
    I think game theory generally means we're dealing players that are making optimal decisions. All the times I've come across game theory has been in economics, where it's assumed participants in the games are acting rationally in their own self-interest. I don't think if you remove the assumption that players are acting logically if its still technically game theory.
    But the bottom line is if your opponent isn't playing optimally and your assumptions about them are correct, a solver strategy which incorporates those assumptions is going to increase your winnings compared to assuming the player is optimal.
  • TheGameKatTheGameKat Posts: 3,654 -
    edited July 21
    Right, so the point is to determine how an opponent deviates from optimal play and maximally exploit that. Which requires one knows the GTO solution. A mathematical term used long before some intelligent biscuit applied it to poker.
    Moderation In Moderation
  • TheGameKatTheGameKat Posts: 3,654 -
    To wrap up (?) this discussion, I should note that our GTO Crash Course delineates all these ideas clearly and rigorously.
    Moderation In Moderation
  • kenaceskenaces Red Chipper Posts: 1,503 ✭✭✭✭
    TheGameKat wrote: »
    You get called for your seat in a 9-handed live game. You notice a friend of yours is the person vacating the seat. She's in a hurry, but as she walks past you on the way to the cage she whispers "half of them are terrible, the other half decent."

    You get your first hand in MP and it's in your opening range, so you raise. Only the SB calls. You've been told that 4 players at the table are terrible, whereas the other 4 are decent. What's the probability your opponent in this hand is terrible?

    LsDgh wrote: »
    @kenaces cheers for the advice but I've already pulled the trigger and bought it.

    I've heard solvers mentioned but haven't really looked into them. Do up just input a scenario and wait for it to output the frequencies at which you should play for an approximated GTO approach?
    I definitely see why a cursory knowledge and access to GTO solutions would be way more beneficial than taking an in-depth look at the maths behind it.

    Yes with solver you input all the variables, and then can view the solutions. If I were ever to reread MOP, I would do it with a solver because I remember there were a lot of toy games. So for me it would be interesting to look at the solutions, and maybe tweak some variables to see changes...

    As others have mentioned you can use solvers to find minimally or maximally exploitive strategies if you know how opponent will play. You can also explore lots of interesting things by changing variables in a solution one at a time to see the impact on the strategies...
  • kenaceskenaces Red Chipper Posts: 1,503 ✭✭✭✭
    Ori13_TTV wrote: »
    to expand on it, i think it comes down to a language problem. in my opinion, 99% of the people who use the term 'GTO' dont understand GTO and misapply the term, so do we use the term in its truer more original meaning, or do we use the term in its more commonly accepted (by people who rarely understand it) evolved definition?

    I agree many don't understand what they are saying - which is a good thing :)

    However, if you care about understanding things taking the time to learn, and use language/jargon correctly is worth the effort. It helps both you and anyone who reads your posts :)

Leave a Comment

BoldItalicStrikethroughOrdered listUnordered list
Emoji
Image
Align leftAlign centerAlign rightToggle HTML viewToggle full pageToggle lights
Drop image/file