Card counters

In the world of casino gambling, card counters are legendary. Most “systems” promising to make it possible to beat casinos are ludicrous. However, blackjack, if played very well, can be winnable by the player. For every successful counter, there are probably hundreds who fail at it, because one mistake per hour can annihilate even an optimal player’s edge. Casinos love the legend of the card counter, because it encourages so many people to do it ineptly, and because there’s a lot of money to be made in selling books on the subject. They don’t love actual card counters, though. Those get “burned out”. Casinos share lists of known counters, so it’s pretty typical that a too-skillful player will be banned from all casinos at approximately the same time, and therefore lose this source of income.

There’s danger involved, as is documented in Ben Mezrich’s Bringing Down the House. In Vegas, they’ll just ban you for counting. Shadier outfits in other jurisdictions will do a lot worse. It’s important to note that card counters aren’t, in any sense of the word, cheating. They’re skilled players who’ve mastered the rules of the game and disciplined their minds well enough to keep track of what is happening; nothing less, and nothing more. Even still, they face physical intimidation.

Lousy players are money-makers, and good players are seen as costs. How damaging are good players to a casino’s economic interests? Not very, I would imagine. Card counting is legitimately hard. Don’t believe me? Try it, in a noisy environment where cards are dealt and discarded rapidly. Of course, most people know they will lose money, because gambling is a form of entertainment for them. Casinos will always make money, but it’s not enough to have 98 percent of the players be lousy ones. It’s better to select lousy players exclusively and toss the skilled ones out. “You’re too good for us. Don’t come back.”

In other words, the possibility of earning an edge through skillful play is used as a lure. Most people will never acquire such skill, and casinos can hardly be faulted for that.

Play too well, however, and you won’t have a spot at the table. Lousy players only. Sure, you can say that you beat the system. It might make for interesting discussion at a party, but your playing days are over. You’ve won, now go away.


15 thoughts on “Card counters

  1. 25 years ago, I did a lot of probability computations and simulations on card counting, while playing Blackjack. Unless the code I wrote has a big flaw, my conclusion is, that even though you have a moderate advantage over the dealer, its not worth while to earn a living out of it. The risk to lose your initial capital, the time you spend at the table and the insane environments in casinos simply doesn’t pay off.
    You can test, simulate and download the code used to build that site from

    • That seems accurate. Poker (where casinos don’t need to pay attention to individual skill, because they take the rake regardless of what happens) is winnable but casino poker is a pretty boring and volatile way to make a living. It’s certainly not glamorous.

    • What's the relation between the reduction in FAS and SEAS?And given last year's hiring binge, does that mean Harvard CS has efflitcveey shuttered its doors for the near future?

    • The parallel is certainly there, but I don’t think it’s limited to technology.

      Create a complex game that is hard for outsiders to decipher. Make it possible for people to win and to lose, with more losers than winners. Control the game. Reward skill that can be bought and subordinated. Punish skill that cannot be.

      Venture-funded technology certainly follows this pattern, but I doubt that it’s the only thing that does.

      • The gentleman from Kentucky who just passed away is a good example of this. Muhammad Ali the boxer could be bought and subordinated. Muhammad Ali the draft resistor was a harder guy to control.

  2. counting cards are an old story, I think now, casinos shuffle cards more often and it will not work. regarding the betting systems, not “most”, but all of them don’t work in the long run, but it doesn’t mean that you can’t make any money in short run, because the reason why these systems lure people into thinking it works, is that you “win” small many times, and when you lose, you lose big, and studies show that humans feel better to win more times in number than in ammount, like in trading where preferable strategy is to “lose” many times small amounts then win big small ammounts but in rare cases(betting against the crowd), which goes against human nature.

    • Card counting isn’t shuffle tracking. Shuffle-tracking can be beaten by having machines do the shuffling, or just having decently skilled dealers.

      Card counting works, both against single-deck and multi-deck blackjack. The thing is that, as Jacob Rief correctly noted, it’s very hard to make a profit at all and the edge is usually small. On top of that, you get banned if you start consistently winning.

        • Oh, sure. That’s true.

          Blackjack is relatively easy to count because it’s suitless and because so many of the distinctions don’t matter. Most systems treat 2-6 (low cards) as a group, for example. It’s much harder to keep perfect count of say, a game like Bridge where every detail matters.

          Single-deck dealers tend to reshuffle when ~1/3 is left. Multi-deck is more variable. It’s a huge pain in the ass to manually shuffle 6-8 decks of cards (it can be done) so if it’s manually shuffled, that happens rarely. If there’s a machine to handle it, then it can be done more often.

          Most people who think they can count cards (and, on top of that, play optimally) are so bad at it that it’s rarely worth it to shuffle constantly. It slows down the dealing and thereby the casino’s profits.

  3. I haven’t explored card counting in many years so it is indeed possible that as someone else says it is an old story. However what I recall from many years ago when I did look into it is that part of the reason why card counting is difficult is because it has to be done in such a way that it doesn’t alert the casino. If you could utilize card counting to the full extent that the rules on their face allow–betting the maximum table bet when the card counting algorithm says to bet high and the minimum bet when the algorithm says bet low–it would be easier to beat the house. However the need to avoid making your card counting obvious to the casino and getting banned or worse requires that one keep the spread between one’s high bets and low bets much narrower than the table theoretically allows. That makes the card counting much less obvious but also makes it much harder to beat the casino.

    • Right, the point is that you typically have a disadvantage to the house of some small percentage. In these hands, you want to bet smaller. In the minority of hands when the count is right (which, as mentioned, requires many hours of constant practice to be able to keep up with), you have a very slight advantage over the house, and you try to casually bet higher without making it too obvious. On top of this, you have to keep up with all the variations of rules per table and per casino, like number of decks (you have to continually divide the count by the number of decks left), soft 17s, and surrendering. Being able to surrender certain hands and counts gives you a good advantage, but it always instantly flags you for increased scrutiny because so few people do it. And as mentioned, you will eventually get kicked out. Generally, a hard way to make an easy living.

      Anecdotally, I hear that if you’re able to pull off the persona of an Asian high-roller, you can get away with more, because they tend to bet wildly all over the place and do crazy moves whether they have the advantage or not.

  4. I’m here, Mr. Church, because of your well-done jeremiad agin Agile-Scrum. Your take on the depravity of it is wholly in line with what I was taught in in a class on teaching where the instructor took us into the black arts of how to make any student, even the best, look and feel, feel and become a complete failure, an ignoramus, a slobbering, choke-sobbing basket case of a person with no hope at all.

    And where better to make my comment on workplace and profession turns for depravity than in a post about card counting! In the summer months just after Atlantic City opening it’s first casino in May of 1978 — Resorts International — while living in a carriage house apartment behind a Asbury Park Victorian in the halcyon days of the Stone Pony, I was learning to count blackjack. Memorizing the count tables, playing through hands. After three months, in September, I went there. They required a jacket! Did I own a jacket? No. But the bouncer suggested that if I were to buy a zip-front hooded sweater from the nearby boardwalk vendor, that would pass. So I did and went in.

    At the tables I quickly found that the dealer played way too fast to keep the three counts needed in my head (cards played, face cards, aces, at the least) I hadn’t practiced alone that fast. A mistake for me. If I tried to play into that condition I was no better than a non-counter, worse I would be confused. I back-tracked to the tables of possible hands (yours and the dealers) that do not require a count. I BECAME A LOSER. Less of a loser, but as the statisticians and mathematicians employed by the casinos know, at that point I was playing a losing game, although losing at a slower rate than the players ruled by “luck” and “hunch”. I lost a few dollars over a couple of hours and left. I some other casino adventures unrelated to gambling, it was overall a fun trip down the Garden State Parkway.

    To say that one can count and beat the dealers is wrong. The ones who have done it “successfully” use counting devices and spotters, computers and discrete hidden signaling, each keeping one of the counts (more sophisticated counting tracts tens and lower cards as well: five counts.) The casinos watch like hungry hawks for them, and the winnings give them away. They are losers too. Both the syndicates of cheaters and the casinos. Losers. Why? Those are not super productive endeavors for a human seeking to grow in good ways. Yeah, Blaise Pascal learned a lot of math by his fascination with games of chance. A fascination that was productive for mankind because a real high-living gambler, Monsieur Chevelier de Mere, the client, the end-user, sought him out to gain a edge. Otherwise Pascal and his works we have gained may have not survived his poverty.

    Okay, so even the worst off of human endeavors, such as casinos, there’s some spiritual and intellectual growth opportunities to be found. It’s a judgement call. One that begs the sweat of one’s brow to make, morally and intellectually. “Ethically”. What can one live with?

    Methods that remove human judgement in any arena are losing methods. THAT’S the big reason that card counting is a loser. There’s little growth in it. Once practiced to the extreme, it removes judgement, one is a automata.

    And thus so too is AGILE SCRUM. It reduces the value of human judgement, it doesn’t elevate it.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s