If I could memorize anything, could I nearly guarantee success in Poker?
April 25, 2013 1:41 AM   Subscribe

I recently listed to a This American Life episode about professional poker players, who learned how to accurately estimate pot odds in Texas Hold-Em based upon their hands and the revealed cards, rigidly bet/folded based upon those odds, and won a lot of money. Is there any reason I can't do the same if I knew how to memorize the calculations they use? If so, where would I start (what books/websites contain this info?) and why can't I just automate that process (or consult a table of calculations) and win a lot of money in online poker?
posted by sdis to Grab Bag (10 answers total) 11 users marked this as a favorite
 
Everyone does this in online poker. And you used to be able to make a lot of money at it, but there are so many bots and people running apps that figure out the calculations, that there's no easy profit to be made any more.
posted by empath at 1:47 AM on April 25, 2013 [4 favorites]


Also, the calculations are actually relatively simple, if you just want to do decently well at your average 'just friends' poker game.

Basically you need to know a couple of numbers -- your odds of drawing particular cards, and the pot odds.

A close-enough approximation to figure out the odds of getting the hand you want is to multiply the number of cards left in the deck that can complete your hand * 4 if there are still 2 cards to be drawn, and * 2 if there is only one card to be drawn. So, for example, if between the flop and your hand, you have 4 hearts, there are still 9 more hearts in the deck that can give you your flush -- which gives you (roughly) 36% chance. If the next card on the table is a spade, you have an 18% chance of getting your flush.

Now you just need to know what the pot odds are, which is basically the amount of money that will be in the pot if you win, vs the amount of money you need to bet to stay in.

If there's two players, and $80 in the pot, and the other play bets $10, you'll have to bet $10 at a chance to win $100. That gives you a 10-1 pay off.

What does that tell you? Basically that you need to have a better that 10% chance of winning the hand to make it worth playing. If you're 4 cards to a flush, you have a 36% chance, so that's definitely a bet you want to make. OTOH, if the guy bets $200 into a $100 pot, you'll end up having to bet $100 on a 4-1 pay off, which means that you'd lose money taking that bet.

That's basically all the math you need to know. Other than that, a lot of the skill comes from playing a lot of games and some memorization -- you need to know which hands you can bet, raise and call on in the first round of betting -- there's a LOT more folding than most newbie players think. You'll also need to be able to get a feel for what other peoples bets are saying about their cards, because you need to know if they're drawing to a straight or if they have pocket pairs, because that makes a huge difference in terms of figuring out what cards you need to draw to win (if any), and so that impacts what your odds are.

One thing you SHOULDN'T do as a new player is try to read faces, and you should never bluff. Just play very conservatively, fold a lot, play the cards that you see, and you'll figure out when a bluff is a good idea once you figure out betting patterns -- something that takes a lot of games.
posted by empath at 2:02 AM on April 25, 2013 [23 favorites]


It's all very basic conditional probability. There are piles of books on basic probability with all the info you need. In fact, the 'pot odds' bit was one of the original applications envisioned by Pascal when he invented the study of probability in the first place.

So, while poker's pretty played out, as Empath says, you might study hard, practice your arithmetic, and look for other games where there's a buck to be made.......

And remember: the lottery is a tax on people who don't know math. And most casino situations are similar. Like the mighty lion, pick no fight which the law of large numbers says you won't, on average, win.
posted by kaibutsu at 2:06 AM on April 25, 2013


rigidly bet/folded based upon those odds

Also, they don't do this. There is a lot of intuition and guesswork because you have imperfect information. Just as an example -- there are a lot of people who really like betting big if there is a pair on the board -- which is a classic bluff -- people will think you have three of a kind and fold. If you've played this guy a bunch, then you know that and you will be more likely to call with two pair than you would against an unknown player. But then again, he also knows that you've played him a bunch and will call him on it, and so he might have been playing cautiously which makes it more likely that he wouldn't be bluffing you, and so on.

It's absolutely critical that they know the pot odds and their odds of having the winning hand. The first is trivial to calculate to 100% accuracy. The second is where all the guess work happens. You have to figure out what everyone else has, how much they're likely to bet, who is likely to call you and so on.
posted by empath at 2:21 AM on April 25, 2013 [4 favorites]


why can't I just automate that process (or consult a table of calculations) and win a lot of money in online poker?

Simply put, because that is the least interesting part about poker. The saying goes that you aren't playing against the cards, you're playing against the other people at the table. Online poker tends to take a lot of that out of the analysis, particularly when so many of the other payers are literally programmed to simply play the odds. The reason poker is unpredictable isn't necessarily that the cards are random, because as you've intuited, the probability calculations aren't that difficult. It's because (1) most people are bad at calculations of this sort in their head, but even more (2) the meta-game of bluffing and counter-bluffing is entirely divorced from the math.
posted by valkyryn at 3:43 AM on April 25, 2013


Knowing the odds is necessary but not sufficient for awesome poker play. If you want to progress beyond basics of pot odds and expected value calculations (which is the meat of most good poker books) into more geekier levels of math, I'd suggest The Mathematics of Poker.
posted by rmd1023 at 4:05 AM on April 25, 2013


Knowing the odds was maybe good enough to win at online poker a few years ago, but probably not anymore.

There's a section in Nate Silver's book The Signal and the Noise about how poker, particularly online poker, has become vastly more competitive over the past 10 years or so. Basically, it used to be that a lot of dumb people played online poker casually and it was not that hard to memorize the odds, become a "shark," and prey upon these people. Nowadays, especially with the government (assuming you are in the US?) cracking down on online poker, the lower-quality casual players have been driven out of the pool, and you're more likely to be playing against people (or computers) who are also very good at consulting that table of calculations/memorizing odds.

Nate Silver briefly made a living playing poker, but found he couldn't make a go of it anymore and became a wunderkind statistics blogger instead.
posted by mskyle at 5:17 AM on April 25, 2013


Short answer: knowing this stuff is necessary but not sufficient to win at tough poker games; winning at poker requires you to play with people who are worse than you, and the vast majority of people you'll play against will know these odds too.
posted by MoonOrb at 6:13 AM on April 25, 2013


Like everyone is saying: what you are describing is the very first step in getting good at poker. Except for the rigid bet/fold algorithm, because even mediocre players will learn to exploit that pretty fast.

And while getting pot odds and discipline down pat is nowhere near good enough for online poker anymore, it's still more than good enough for the $1/$2 no-limit tables at brick and mortar casinos, especially on the weekends, provided your bankroil is deep enough (I made a living playing low-stakes holdem in Atlantic City for a while).
posted by 256 at 6:28 AM on April 25, 2013 [1 favorite]


For the record, the show you heard was Radiolab, not TAL. To be fair, it was a particularly Ira Glass-esque episode of Radiolab.
posted by Keith Talent at 9:58 AM on April 25, 2013 [1 favorite]


« Older How do you cheer up when you're sad?   |   What are the prerequisites to learn HTML5? Newer »
This thread is closed to new comments.