I'm surprised at the general tone of response here towards the subject of poker overall (I didn't read the article yet so perhaps something more is in the context, but all the same); I understand there are many in the poker world even regarding the most successful of whom are regarded as living a degenerate lifestyle, but being that I was previously in that world myself and was not a degenerate type; I never gambled outside of "my game" that I had an edge in, I learned and implemented proper bankroll management and I studied the game on fundamental levels and on up, progressing into the meta-psyche game that is NL heads-up.
Which brings me to my point which is that while some forms of poker have proven "beatable" by ai, certain forms ie; short-handed tables of NL Holdem, increase in perpexlity to a point where, in heads-up, there are too many variables at play both "physically" (the cards and corresponding hand ranks) and metaphysically (the story being implied thru the route of actions taken at each street from preflop, flop, turn, to river) for there to exist some perfect approach against a skilled player.
NL Holdem poker is absolutely a game of skill with an element of variance aka luck/lack-there-of.
I agree with everything you've said, and I think we'd have better politics, economics, human relationships and fun, if more people got their heads out of their posteriors and actually understood poker more.
Also used to be in that world and identify similarly in terms of my lack of love for gambling.
I'd suggest that you're empirically incorrect in saying that there is no perfect approach against a skilled player (6handed games which often reduce to a single heads-up interactions by showdown):
1. we know that a Nash equilibrium exists for every two-player zero-sum game such that it’s mathematically unexploitable
2. Pluribus approximated the Nash well enough (didn’t have to search over 10^161 possibilities) to crush high stakes skilled player over a good run of hands
> NL Holdem poker is absolutely a game of skill with an element of variance aka luck/lack-there-of.
I’ve played a lot of Holdem, and I’m not sure I agree. A lot of what passes for skill is just an innate understanding of the odds.
John Scarne writes about gambling that a good bet isn’t one you are likely to win, but one where the payout is enough to be worth the risk. The best players know the odds of pulling a straight and can do math to figure out if it’s worth chasing one.
Also known as Expected Value (EV), as in, how much is in the pot right now compared to how much you’re betting/calling, usually compared to how likely you are to win a hand using the cards you’re holding.
That works well for limit games, where you can’t bet more than a set amount (in relation to the blinds or the current pot), especially when there are multiple people at the table, and you’re in an advantageous late position so others act before you do.
In high-stakes no-limit heads-up (1v1) play, the cards you’re holding matter less, especially before the flop. EV and pot odds are almost useless except for gauging when to bluff / if you’re being bluffed. Hands rarely end in a showdown as opposed to one of the players folding. The hands that do are essentially coin-flips, with both players holding what they believe are strong hands.
I think this is true and why programs like like Pluribus, Libratus and DeepStack have outperformed professionals in both heads-up and multiplayer no-limit Texas Hold'em. It's not reading social cues like traditional players, but just relying on probability. Even when giving perfect knowledge of the computer strategy to humans, they're still unable to exploit.
Humans are improving their game by using solvers and introducing randomness into their decisions. For instance, an optimal strategy given a hand might be "fold 80% of the time". One way to do that in live play is look at the second hand of a watch and fold unless it seconds (in this case) are about 48 (80% prob).
You're not wrong that knowing the odds is a component of the skill, but to suggest that skill in poker stops there is minimizing many of the advanced aspects that require playing at a higher level (information management, assessing a player's likely range, determining the equity of a player's range with cards to come, realizing when your or their range is capped, etc)
> I’ve played a lot of Holdem, and I’m not sure I agree. A lot of what passes for skill is just an innate understanding of the odds.
"The odds", however, are not simply a function of the cards in your hand and the unknown cards in the deck. There are also the cards in other people's hands, and getting a good read on what they may be based on the person's behavior is absolutely a skill.
It's always entertaining to play poker with 1 friend who is very skilled at a table full of novices. They often get frustrated and crash out due to their read on other people's behavior being miscalibrated to the situation.
Hey I've started playing poker occasionally again, wanna have a chat about poker? My email is in my profile.
I used to be a winning player at small stakes about 20 years ago, so nothing major but enough for me to show that it's a game of skill.
But yea, for anyone interested why poker is a game of skill, it's due to the law of large numbers. You can easily see the law kick into effect when you simulate a dice roll and you win from 1 to 4 and the other wins 5 to 6 and you both get $1 if you win. I recently had to explain this concept so I happen to have the JS still lying around in my Chrome console.
const rolls = 10_000;
let a = 0;
let b = 0;
for (let i = 0; i < rolls; i++) {
const die = Math.ceil(Math.random() * 6); // 1–6
if (die <= 4) a++;
else b++;
}
console.log(`Player A wins: $${a}`);
console.log(`Player B wins: $${b}`);
console.log(`Total paid out: $${a + b}`);
console.log(`A's edge per game: ${(a - b) / rolls}`);
console.log(`Difference: ${(a - b)}`);
Poker has much, much higher variance than dice though (or weighted coins, which is what you're actually modeling). It takes hundreds of thousands of hands to establish a statistically significant win rate.
At a common online pace of 1.5 hands per minute (live games are much slower) that's over a thousand hours of playing. I.e. even if playing for one hour every day, it takes years before a player knows whether they're profitable or not.
Seems disingenious to compare to dice when you presumably know poker belongs to that class of distributions to which the central limit theorem applies very slowly.
> I'm surprised at the general tone of response here towards the subject of poker overall [...] regarded as living a degenerate lifestyle
Maybe the people who are negative have read to the end of the article where we are let into the not-so-hidden agenda of the parent: Teach the kids to hustle their way through college so they can become a market speculator.
One of the best life lessons I learned was while perusing a poker strategy book in a bookstore as a teen. I’ve never been into poker, not even sure why I picked it up.
One thing it said was the most important thing to remember is that most of your hands will be crap. Don’t get attached to a bad hand and don’t convince yourself that an ok hand is a good hand. If you just fold the bad hands and play the good ones you’re already a better player than most.
I took that to heart and it has served me well in life.
That’s it. That’s the entire strategy. I pray that the Texas Hold ‘Em fad doesn’t come back. That was an insufferable decade of hearing how clever everyone was.
Being a loose aggressive player is far more likely to lead to you losing a lot of money, than winning a lot of money.
Once you consider what the house earns, poker is a net negative for the players. In order for there to be some big winners, there have to be a lot of losers. And a shocking number of those losers will, thanks to our selective memories, consider themselves winning players.
I like this. Most people try to teach card games by listing every rule, but it's much easier to play a simpler version then add in new rules.
I play the Chinese card game Zhao Peng You (Finding Friends, part of the Sheng Ji family of games https://en.wikipedia.org/wiki/Sheng_ji), which is a trick taking game with a trump suit that changes between games, a trump number that changes between games, and a team selection mechanic rather than fixed teams. It's insanely hard to learn everything at once, so we usually start new people with fixed teams and trumps just to get the feel of a team-based trick-taking game, before adding in the complications.
Every game is easier to learn when you start with simple rules and then add new ones as you go.
We teach people Liar's Dice. It's a very simple game, especially if you build it up like this. Everyone gets five Dice. You roll them and look at your own, and then take turns guessing how many of a given number are on the table. Guesses have to "go up" (either the number of dice stays the same and the number of pips goes up, or the number of dice goes up). Instead of guessing you can challenge the person before you. Whoever is wrong loses a die and game play repeats.
After a few rounds, dice showing a one are wild.
After a few more rounds, if anyone in a round bids 1's, then ones are not wild for that round.
After a few more rounds we start discussing the probabilities and strategies.
The challenge is not breaking the game fundamentally while you add rules.
We play Uno like this. Start with the basic (agreed upon house rules), then every time someone wins a hand they get to add or remove any rule they want, as long as it doesn't outright break the gameplay.
Blind Man's Bluff is a great variant: Give everyone a card face-down, they put it on their forehead without looking at it. Bet based on whether or not you think the card on your forehead is higher than other people's. More fun in my opinion.
Article footnote mentions this with the caveat that it requires some dexterity that young children may find challenging. That aside, I think the two games make a great complementary pair and switching between provides a nice contrast for kids.
You play normal one card poker until the kids realize the benefit of seeing other player’s cards - then you play blind man’s and learn that incomplete information can go the other way.
I’ve taught all five of my kids how to play poker, and if they ever sit down at a cash game consider their stack gone and play the cards (remove the dopamine chaos). Learn the math, betting strategies, and look for villain patterns.
These all directly relate to real life.
I believe in it so much that I have a tournament training app startup: https://mach9poker.com/.
There’s a company in Chicago that teaches women poker in relation to business: https://pokerpower.com/.
Bankroll management is a critical skill regardless of the use case.
Life is full of uncertainty. Learning to take calculated risks, where most attempts fail but a few ones pay off big, is an important life skill. Reading other people's behavior to infer hidden information is another one -- Jane Street apparently used to have people learn poker to learn how to infer hidden information from the behavior of other people buying and selling stocks, but invented their own game (https://www.figgie.com/) to teach the same skills more efficiently.
ETA: I would say, when poker is taught correctly, it should discourage anyone from the sorts of gambling which are problematic:
Problem 1: Wasting your money in situations where the odds are "with the house". This would include playing slot machines or basically anything at a casino, the lottery, or even 50/50 raffles (although I can see an exception for the last one).
Poker should teach you to only take bets where the expected value (value of winning * prob winning) is greater than the cost, which is not true in the above examples.
Problem 2: Getting sucked into betting more and more to make up what you've already lost. One aspect of long-term poker should be teaching you is how to manage this effectively.
That’s funny. I’ve played poker but I’ve never gambled a cent in my life. How does that work? Oh yeah, we played poker with plastic chips not backed by any money. We just played for fun.
Likewise, never gambled once even when exposed to the possibility, but I love a good game of poker or blackjack, it's fun for the mind and it's sociable. Our maths teacher a few decades ago used roulette and other games to teach us about statistics, we all loved it and it engaged the entire class, a bonus for slower maths learners. Today I suppose it's not allowed in the classroom?
Once kids get familiarity with odds and probability they will soon realise that casino games they have no edge and the house always wins. Also you cannot bluff a casino dealer which is half the fun
Readit News
Which brings me to my point which is that while some forms of poker have proven "beatable" by ai, certain forms ie; short-handed tables of NL Holdem, increase in perpexlity to a point where, in heads-up, there are too many variables at play both "physically" (the cards and corresponding hand ranks) and metaphysically (the story being implied thru the route of actions taken at each street from preflop, flop, turn, to river) for there to exist some perfect approach against a skilled player.
NL Holdem poker is absolutely a game of skill with an element of variance aka luck/lack-there-of.
I reccomend:
Thinking in Bets: Making Smarter Decisions When You Don't Have All the Facts Hardcover – 6 Feb. 2018 by Annie Duke
Or listen to any of the podcasts she did when promoting the book - Peter Attia or Masters of Business are the two I presonally consumed at that time.
It's a warped puritanism.
I agree with everything you've said, and I think we'd have better politics, economics, human relationships and fun, if more people got their heads out of their posteriors and actually understood poker more.
I'd suggest that you're empirically incorrect in saying that there is no perfect approach against a skilled player (6handed games which often reduce to a single heads-up interactions by showdown):
1. we know that a Nash equilibrium exists for every two-player zero-sum game such that it’s mathematically unexploitable
2. Pluribus approximated the Nash well enough (didn’t have to search over 10^161 possibilities) to crush high stakes skilled player over a good run of hands
I’ve played a lot of Holdem, and I’m not sure I agree. A lot of what passes for skill is just an innate understanding of the odds.
John Scarne writes about gambling that a good bet isn’t one you are likely to win, but one where the payout is enough to be worth the risk. The best players know the odds of pulling a straight and can do math to figure out if it’s worth chasing one.
That works well for limit games, where you can’t bet more than a set amount (in relation to the blinds or the current pot), especially when there are multiple people at the table, and you’re in an advantageous late position so others act before you do.
In high-stakes no-limit heads-up (1v1) play, the cards you’re holding matter less, especially before the flop. EV and pot odds are almost useless except for gauging when to bluff / if you’re being bluffed. Hands rarely end in a showdown as opposed to one of the players folding. The hands that do are essentially coin-flips, with both players holding what they believe are strong hands.
Humans are improving their game by using solvers and introducing randomness into their decisions. For instance, an optimal strategy given a hand might be "fold 80% of the time". One way to do that in live play is look at the second hand of a watch and fold unless it seconds (in this case) are about 48 (80% prob).
"The odds", however, are not simply a function of the cards in your hand and the unknown cards in the deck. There are also the cards in other people's hands, and getting a good read on what they may be based on the person's behavior is absolutely a skill.
It's always entertaining to play poker with 1 friend who is very skilled at a table full of novices. They often get frustrated and crash out due to their read on other people's behavior being miscalibrated to the situation.
I used to be a winning player at small stakes about 20 years ago, so nothing major but enough for me to show that it's a game of skill.
But yea, for anyone interested why poker is a game of skill, it's due to the law of large numbers. You can easily see the law kick into effect when you simulate a dice roll and you win from 1 to 4 and the other wins 5 to 6 and you both get $1 if you win. I recently had to explain this concept so I happen to have the JS still lying around in my Chrome console.
At a common online pace of 1.5 hands per minute (live games are much slower) that's over a thousand hours of playing. I.e. even if playing for one hour every day, it takes years before a player knows whether they're profitable or not.
Seems disingenious to compare to dice when you presumably know poker belongs to that class of distributions to which the central limit theorem applies very slowly.
Maybe the people who are negative have read to the end of the article where we are let into the not-so-hidden agenda of the parent: Teach the kids to hustle their way through college so they can become a market speculator.
Deleted Comment
I took that to heart and it has served me well in life.
Once you consider what the house earns, poker is a net negative for the players. In order for there to be some big winners, there have to be a lot of losers. And a shocking number of those losers will, thanks to our selective memories, consider themselves winning players.
I play the Chinese card game Zhao Peng You (Finding Friends, part of the Sheng Ji family of games https://en.wikipedia.org/wiki/Sheng_ji), which is a trick taking game with a trump suit that changes between games, a trump number that changes between games, and a team selection mechanic rather than fixed teams. It's insanely hard to learn everything at once, so we usually start new people with fixed teams and trumps just to get the feel of a team-based trick-taking game, before adding in the complications.
We teach people Liar's Dice. It's a very simple game, especially if you build it up like this. Everyone gets five Dice. You roll them and look at your own, and then take turns guessing how many of a given number are on the table. Guesses have to "go up" (either the number of dice stays the same and the number of pips goes up, or the number of dice goes up). Instead of guessing you can challenge the person before you. Whoever is wrong loses a die and game play repeats.
After a few rounds, dice showing a one are wild.
After a few more rounds, if anyone in a round bids 1's, then ones are not wild for that round.
After a few more rounds we start discussing the probabilities and strategies.
The challenge is not breaking the game fundamentally while you add rules.
These all directly relate to real life.
I believe in it so much that I have a tournament training app startup: https://mach9poker.com/.
There’s a company in Chicago that teaches women poker in relation to business: https://pokerpower.com/.
Bankroll management is a critical skill regardless of the use case.
ETA: I would say, when poker is taught correctly, it should discourage anyone from the sorts of gambling which are problematic:
Problem 1: Wasting your money in situations where the odds are "with the house". This would include playing slot machines or basically anything at a casino, the lottery, or even 50/50 raffles (although I can see an exception for the last one).
Poker should teach you to only take bets where the expected value (value of winning * prob winning) is greater than the cost, which is not true in the above examples.
Problem 2: Getting sucked into betting more and more to make up what you've already lost. One aspect of long-term poker should be teaching you is how to manage this effectively.
Deleted Comment