4 ms·
Fascinating. I haven’t played or thought about poker in 15 years. Can you elaborate more on the current strategy of poker? A real example would be helpful.
by iknowSFR 4y ago
Fascinating. I haven’t played or thought about poker in 15 years. Can you elaborate more on the current strategy of poker? A real example would be helpful.
- orwin 4y agoPeople use rng generators to take decisions nowadays. Impossible to really exploit.
- RivieraKid 4y agoUsing randomization doesn't mean you're anywhere close to being unexploitable.
- zucker42 4y agoA large idea in modern poker theory is playing the Nash equilibrium strategy. A Nash equilibrium is where no player can improve their expected result by unilaterally changing their strategy, and two player games provably have the property that if you play any Nash equilibrium strategy, you expected value will always be greater than or equal to your EV when your opponent also plays an equilibrium strategy. The application of this is that at any given point during a hand you and your opponent have a "range" of hands (the term in the technical solver literature is an information set) that it would make sense for you to have from your opponents perspective. A range is a probalistic weighting of hands. Every action changes a players range, and for example if a player bluffs too much you can exploit that by calling with a bigger range against them. So poker becomes a game of, rather than figuring out what your opponent has, instead establishing a baseline strategy that can't be exploited no matter what your opponent does and then adjusting that to optimally exploit your opponent. That's the basic framework. Another book recommendations is "Modern Poker Theory".
- RivieraKid 4y agoAt every decision point, you know what your range is and what your opponent range is. Then you choose one or two bet/raise sizes based on the board and the ranges. Then you split your range into categories: value bet, call, bluff, fold. And then you look at your hole cards and play based on its category. Of course, this is very hard to do in your head so, in practice, players rely on various principles and heuristics based on what the solver does in various situations. Here's a pro explaining his thought process: https://www.partypoker.com/blog/en/its-the-same-game-but-it-isnt.html https://www.partypoker.com/blog/en/its-the-same-game-but-it-...
- kuboble 4y agoYou can search youtube for things like "hand analysis with piosolver"
- eof 4y agoI can only tell you my perspective as an ex poker pro who has not studied with solvers. But the basic idea is that you put in a range of starting hands for yourself and the opponent, and possible bet sizes, and the solver will brute force a play that approximates Nash equilibrium. The output of the solver will be something like “45% of the time bet. 55% check.” Or “95% fold. 5% raise X” The general loop as I understand it is inputting tricky/common situations running the solver and basically memorizing it.