There'll Be Time Enough for Countin' When the Dealing's Done

Jim Manzi on counting cards:

My experience was that it was very easy to stay under the radar of casinos….Just play solo at the quarter tables, never spike your bet above 5:1, and play no more than one hour at casino before you move on to the next one. There are about 100 casinos in Vegas, so you can play ten hours per day every other weekend and only visit a given casino once every two or three months (for an hour each time). No pit boss will know who you are or care what you're doing because you're so far down in the noise. You can make a lot of money this way. Of course, nobody will ever know that you are taking them, and the emotional satisfaction arises from walking into this multi-billion dollar enterprise and walking out with their money because you're smarter and more disciplined than they are. In a bizarre way, you succeed through classical bourgeois virtues: self-discipline, frugality, ego control and steady work.

Once you realize all this, of course, you figure out that you can make a lot more money in that giant casino called Wall Street.

Comments

[Elemenope]

Once you realize all this, of course, you figure out that you can make a lot more money in that giant casino called Wall Street.

And suddenly, the essay takes a turn into bizarro land.

Geez, if I knew picking winning stocks was just like counting cards, I'd have been rich years ago! (ROFL.)

[Some dude]

So what are your 2?, Jesse Walker?

[Jesse Walker]

So what are your 2?, Jesse Walker?

I have no two cents. I'm not a gambler, and I have no idea how to count cards. But I was entertained by the portrait of the successful card-counter as an exemplar of the bourgeois virtues, bound for a brokerage job.

[Bronwyn]

Thanks, Jesse. I was wondering what would be on today's internal soundtrack. Now Kenny's all cued up.

[TallDave]

Yah, if you're ambitious and intelligent you can make seven figures in investment banking just playing arbitrage and options games. Not that different from a thinking man's poker, really. It's pretty high-stress though.

Personally, I'm more than happy to write code for six figures. But then I'm not that ambitious.

[Someone Who Doesn\'t Want to L]

Geez, if I knew picking winning stocks was just like counting cards, I'd have been rich years ago! (ROFL.)

Actually, MIT (at the time) mathematician Edward O. Thorp, the author of the first big card counting book, "Beat the Dealer," wrote a second book titled "Beat the Market." I haven't read the second one, but in it I believe he uses similar game-theoretic ideas applied to give a consistent edge in dealing with hedge funds. It's probably out of date now, but then again, so is "Beat the Dealer." As the methods in "Beat the Dealer" have been improved, I'd assume those in "Beat the Market" could be as well.

[Someone Who Doesn\'t Want to L]

Counting cards is relatively easy, theoretically. There's really only one number you have to keep track of yourself. You also need to have a good idea of the approximate number of decks remaining. (The method kind of has to be simple since you need to do it nearly perfectly under what could be stressful conditions.)

However, it's also boring.

[ed]

However, it's also boring

Unlike, say, watching poker on TV.

[Warren]

Just play solo at the quarter tables

Quarter tables? As in $0.25 or is that Vegasspeak for $25. Because there's no way the house could justify running a table that's only taking in a dollar a hand. OTOH a $25 minimum, while not 'high roller' territory is high enough to get attention. And if your, playing solo never spiking your bet above 5:1, your edge over the house is so slim you could make better money waiting tables.

I call bullshit, this guy's full of it.

[ClubMedSux]

Once you realize all this, of course, you figure out that you can make a lot more money in that giant casino called Wall Street.

You know, I've always thought the better analogy was fantasy sports. With both stocks and fantasy sports, you're ultimately speculating on somebody else's success in a field of which you (generally) have limited knowledge. And given that I know infinitely more about baseball than stocks, I've always figured that I would be pretty stupid to try and play the market on my own.

[Scott Wood]

Can you make a living at the quarter tables?

[robc]

Warren,

From some stuff I just googled, 5:1 is fine for a single deck table (actually 4:1 is fine). For a 6 deck shoe, you need to be able to go to 12:1.

[Colin]

A friend of mine and I once came up with a system for winning at roulette. We'd each make a modest $300 or so each time we went to the casino. Then, I told this guy I worked with about it. He was a Math major from Harvard and spent a whole week developing a proof why our system was bad.

And wouldn't you know, the very next time we went, we lost.

[R C Dean]

Actually, MIT (at the time) mathematician Edward O. Thorp, the author of the first big card counting book, "Beat the Dealer," wrote a second book titled "Beat the Market."

And I beleive he also did some consulting and trading on Wall Street, and made himself a crapload of money there.

In a bizarre way, you succeed through classical bourgeois virtues: self-discipline, frugality, ego control and steady work.

Once you realize all this, of course, you figure out that you can make a lot more money in that giant casino called Wall Street.

See, also, Warren Buffett.

[Raymond]

Tens are good.

One if it's bad. Two if it's good.

Counting ... counting is bad. No counting.

I'm an excellent driver.

[Bob Smith]

Don't forget that to win in Vegas you must start with a significant bankroll. Even if you have a statistical advantage over the house, you must be able to weather the statistical variance in losing streaks. "I'll make money in the long run" only matters if you can get to the long run. When a short bankroll requires you to stop betting in the middle of a long losing streak you've essentially lost everything, even if you would have made up for the losses had you been able to continue.

As others have pointed out, I question what the author terms "a lot of money". Given the long time between paying hands and the modest advantage you have over the house, making a significant "wage" per hour is tough without playing at a table with a large enough betting limit that they'll pay attention to you. Then again, playing cards for a living, even at minimum wage, is probably more fun than a regular minimum wage job (and has better tax benefits, since unlike regular folk career gamblers get to deduct their gambling losses).

[Someone Who Doesn\'t Want to L]

Colin:
Roulette is bad for the player (VERY bad) unless you cheat (with some sort of mechanical device, maybe), or maybe if the casino tries to cheat. If the Casino rigs crooked wheels, you can use that knowledge to win, but the fairer the wheel, the more certain your eventual loss.

The house edge in Roulette is 5.26%, the worst of all the casino games for the player.

[Episiarch]

What the hell is this guy talking about? I haven't seen the movie, but I've seen a special on the History Channel regarding the MIT team that inspired the movie. This isn't a one man operation; you can't count a six-shoe deck unless you are Rain Man* (I'm a very good driver...driver).

It required a number of participants; one to play the big hand (the one that bets the real money), at least one counter, and I think a signaler. It wasn't based on actually counting cards, but rather keeping track of how many high cards and low cards have fallen unsing a single count that would range from -5 to +5. Whenever the count got near either end it meant it was either time to bet big or assume the dealer would bust, something like that.

The casinos are far from stupid, and share information, and after a while they do get wise to you. Once they do, they spread you to all the other casinos and you get shut out universally.

So these teams, using smart, ambitious kids, could make a go at it for a while and make a profit, but eventually they fail from stress, recognition, or a catastrophic screwup.

* Yes I know Tom Cruise was actually "Rain Man/Raymond"

[Episiarch]

* Yes I know Tom Cruise was actually "Rain Man/Raymond"

Whoops, goof by me. Tom Cruise actually said "Rain Man" for Raymond as a kid.

[Warren]

robc,
True, but how many single deck tables are there? Certainly not one in every one of those 100 casinos.

Using a six deck shoe and reshuffling after only four decks have been played, cuts the percentage available to card counters to minuscule amounts. You need a team of people working together to make it pay, and you'll be spotted and shown the door before you recoup your gas money.

Single deck tables, and tables with less frequent reshuffles can be found, but have tight min/max limits. It's theoretically possible to turn the edge to your favor, but not enough to make it preferable to a straight job.
The days of counting cards are over.

[Warren]

SWDWTLHJ

Let me introduce you to Claude Shannon. A man whose tremendous genius has been largely overlooked. He also was part of the team that beat the casinos, but he did it at roulette.

Read:Fortune's Formula: The Untold Story of the Scientific Betting System That Beat the Casinos and Wall Street

[Scooby]

My experience is that the quarter ($25) tables can be good for an average $500-1000 per 12 hour day payoff, but my sample size is small. I've had weekends where I was up $7-8k, but those have been counterbalanced by weekends of dropping $3-4k.

It's not exactly money for nothing- it's not productive, but it is draining work. I tried for a while to make a living gambling. While I made almost as much doing that as at my chosen profession, it was mentally draining (due to the boring, repetitious tasks involved) and soul crushing (you see a lot of broken people). One upside was being able to shock the squares by telling them of my "job" as a "professional gambler."

All told, I'd much rather do my day job for a living, and go to the casino for entertainment.

[robc]

Warren,

From what I was googling, it looks like on a 6 deck shoe, playing $5-$60 spread bets, you can make $7/hr (over the long haul, with extreme variation).

Multiply that by 5 for $25 min tables and its $35/hr (but also needs a 15k bankroll). There are better ways to make that kind of money.

If you are playing the $5 tables, you will probably never be kicked out. One screwup of a $60 bet will wipe out a whole days worth of profit. I doubt at that level you would ever even be noticed.

It sounds like the places with single deck tables do it to cater to counters - 90% of which lose in the long run because they suck at it.

[Scooby]

Oh, I'll add that I did the majority of my counting on double deck tables. Single deck only plays out two or rarely three rounds between shuffles, so there is no way to vary the bet to correspond with the count.

Six and eight deck shoes are no more difficult to count than single or double deck. The only difference is that the raw count is a lot more volatile in six/eight deck shoes.

[ChicagoTom]

Using a six deck shoe and reshuffling after only four decks have been played, cuts the percentage available to card counters to minuscule amounts

When I was in Vegas last, I played BlackJack at the Bellagio...and they had a never ending shoe.

The shoe was mechanical and seemed to be constantly shuffling the cards. Used cards just got put back into the machine and shuffled back into the mix. We would play for hours non-stop. I really disliked it

Also, now many Casinos have adopted rules that don't allow mid-show buy ins.

[Scooby]

We would play for hours non-stop. I really disliked it

What's next? "It was horrible. We couldn't stop." If you don't like it, don't do it.

Continuous shuffle machines make counting impossible. But even without counting, blackjack has one of the lowest house advantages (depending heavily on table rules). It's a long term loser, but more slowly than most games in the house.

[Citizen Nothing]

If it weren't for blackjack I wouldn't bother going to casinos. I can lose my money very slowly and usually have a good time jawing with the other players and the dealer. It's cheap entertainment.

[Russ R.]

I don't gamble in casinos, but I do like to visit and observe. You learn a lot about human psychology, but even more about business operations management and profit optimization.

One thing I've noticed is that more and more casinos are adopting the continuous shoe. After every hand, the dealt cards are loaded back into the shoe and the decks get automatically reshuffled. This has two benefits... first, it makes keeping a count impossible, but second, and more importantly, it also decreases the average cycle time (since there is no down time for shuffling), meaning more hands dealt per hour.

The profit formula is simple: Average margin per dollar bet x Average dollars bet per hand x Average number of hands dealt per hour per table x Average number of table hours spent per player x Average number of players per table x Number of tables in the house.

[Dee]

With the advent of the contant shuffling card deck machines it is impossible to count cards these days. Sure you could still count them but the count you have will be of no advantage.

Even the best card counters in the past only had a mere 1-2% edge over the house. That is not much to have any errors and still be up at the end of the night. Counting is easy, staying focused was the hard part when the waitress with her boobs hanging out walks around to get drinks all the time it is a bit distracting.

With the 1 and 2 deck games you really have no time to count before a reshuffle and even ifyou could get a good count it would only be for one hand or two at most before reshuffle happens. They also change the rules allowing the dealer to not have to stand on a set number etc. Believe me if it looks like they made it an easier game to win it only means they made it easier for you to lose.

For those that think they have a Roulette system your nuts. Its fun and slow moving when the table is full so you can get lots of drinks but odds wise it is about the worst bet you can make in a casino.

Last time I went to a casino I brought $100 expecting to blow it and leave. After walking around for an hour and playing a few slots I decided I was better off sticking the $90 I had left in my pocket in my wallet and leaving. For all intensive purposes you might as well walk into a casino, go straight to the cashiers desk, slap your money of the counter and ask if you have won. All the rest is just a distraction to make you feel like your getting something for your money.

[ChicagoTom]

What's next? "It was horrible. We couldn't stop." If you don't like it, don't do it.

"it" = the continuous shuffle shoe -- not gambling. The whole paragraph is about the shoe not about gambling. The gambling we LOVE.

[ChicagoTom]

If it weren't for blackjack I wouldn't bother going to casinos.

Craps baby! You stay away from the hard ways and the prop bets, stick to pass/dont pass or the come line, take advantage of the free odds and you can really lower the house edge.

Again though, this is one of those games that you need a bankroll to withstand a couple of bad shooters.

[Scooby]

ChiTom,

I get that. What I'm saying is: don't encourage those bastards. Don't play at tables with CSM's, because they suck.

[Edgehopper]

If you're going to gamble, stick to multiplayer games rather than games against the dealer. Poker rooms, sports betting, and off-track betting are your friends, or at least more so than the table games and slots. Low-limit games have large edges available (relative to the bet size) for good players with huge variances, high-limit games have smaller edges and smaller variances, and no-limit games are most comparable to playing a game of skill for money.

The MIT folks made it work with a team that could count multiple tables, watch for huge player advantages, and signal the team's "high roller" to head to the right table and place huge bets. This approach lets you place bigger spikes without being obvious about it, and not have to wait for a single table to develop enough of an advantage to make your profit.

[Someone Who Doesn\'t Want to L]

Warren,
I was aware of Thorp's work in Roulette, which was apparently done with Shannon. I don't know how it actually worked, but it's the "electronic device" of which I spoke earlier. Some sort of wearable computer, I believe.

[Someone Who Doesn\'t Want to L]

This isn't a one man operation; you can't count a six-shoe deck unless you are Rain Man* (I'm a very good driver...driver).

Episiarch: Counting is the same regardless of how many decks you have. It isn't really much harder at all. You only have to keep track of one number:

A new deck starts at a count of 0.

Every 10, face card, or ace you see go by subtracts one from the count.

Every 2,3,4,5, or 6 you see adds one to the count.

The higher the count (especially compared to how many cards remain), the better the deck is for a player generally (because of increased chances of getting a Blackjack, and increased chances of the dealer busting). There are some play strategy changes you make when the count is high or low to improve your odds as well, but mainly to change the take you change your bets. Roughly, bet more when the count is higher and less when it is lower.

[Anon]

Thorp's book on gambling (his later one) is available here. The relevant chapter on roulette (PDF) is here. One thing to note (as Thorp does) is that his method required being allowed to place a bet after the ball was launched. I have no idea what casino policy is on this issue these days, nor if wheels have changed in some way to make Thorp's analysis invalid.

Anon

[Ryan Jensen]

For all intensive purposes ...

I love it! It sounds so dramatic!

[Ska]

Craps is efinitely the game I love to play the most.

And Citizen Nothing - blackjack at $25 (avg. minimum you'll find on a weekend) every half minute is far from cheap entertainment. If you want cheap entertainment, go sit in a park. Thinking of gambling as entertainment is the best way to lose.

[prolefeed]

This isn't a one man operation; you can't count a six-shoe deck unless you are Rain Man* (I'm a very good driver...driver).

Episiarch: Counting is the same regardless of how many decks you have. It isn't really much harder at all. You only have to keep track of one number:

What I think Episiarch was getting at was that with a six-deck shoe and reshuffling after four decks, counting cards doesn't give you an edge in the odds. You can count the cards, sure, but you won't make money because you'll never get to extreme situations that are highly favorable to you.

[JM]

Haha when was this article written? 1990? There are no quarter tables left in Las Vegas, lowest you will find is 5 dollar minimum bet.

[Scooby]

JM,

In Vegas, "quarter" means $25. It might also surprise you that there was never a time when dime-bags cost ten cents, nor have nickel-rocks ever been available 20 for $1.

Epi & prolefeed,
The math for counting single deck, double deck, six deck and eight deck are the same. I've never seen a single deck table that gave more than about 50% penetration (i.e., half the cards dealt out before a shuffle)- playing head to head they only go about 25% penetration in most places (three rounds or so). I have seen six deck shoes that played down to less than 1 deck- more than 83% penetration. I'd rather be at that table than a single deck table. Sure, there's a chance that all 96 ten cards could be behind the cut card (giving an astronomical true count of between +44 and +48 with 2 decks remaining, but not dealing out any ten cards), but that's not very likely.

Also, 6 deck shoes rarely have a mid-shoe entry restriction. That is why team methods can work against them- one guy counts and plays flat betting at table minimum, a teammate jumps in with big bets only when the counting player signals that the count is favorable.

No matter the size of the shoe, the probabilities are similar, when corrected for true count. One doesn't need more than elementary arithmetic to work a counting system. OTOH, actually understanding how it works (beyond tens are good, small cards are bad) is more suited to people who have mastered college-level probability and statistics.

[gorak]

If your really that smart you could play poker and sit all day at one table making oodles of money.

[george]

"Isn't poker a game of chance?"

"Not the way I play it."

[Scooby]

Smart has very little to do with it. Anyone who can add and subtract small integers, estimate the number of decks remaining to be dealt with some degree of accuracy by looking at a stack of cards, and multiply and divide by fractions can successfully execute a card counting system over the short term. To successfully work it over the long term takes the virtues pointed out in the original post: self-discipline, frugality, ego control and steady work. If you have those qualities, you have a decent chance of being successful in any venture you pursue.

As far as poker goes- live poker is (to me) boring as hell- too slow to keep my attention. I'm tight folding preflop and waiting for the next deal 80% of the time or more. On a full table, you're only seeing a few dozen hands per hour, and seeing the flop on a dozen or so.

Poker's less mechanical than counting cards & blackjack, but it's still immensely dull. Any change from boring means that you're probably in trouble. It also shares the same pisspoor working environment- nowhere close to every gambler is a degenerate, but casinos have a disproportionate number of addicts and scammers amongst their clientele.

Making a living at gambling takes all of the fun out of it- it becomes a boring, unrewarding job with no benefits and no security.

[eric taylor]

hi jesse. you need a team to make money playing blackjack. The problem is that you have to make tiny little token bets until the count is right, so you are sitting there making $5 bets until the count is right then you have to play the table maximum. So right there that is a problem, because the dealer will start shuffling more often and you won't get as many big bets on a good deck if he notices you change your betting behavior. He doesn't even have to know you are counting he just sees that you go from little bets to big bets, and he shuffles, now you have to wait again for a good count.

Second if you use the "Kelly Criteria" to determine how much to bet, you'll find that it is proportional to the size of the bankroll divided by the odds. So for a game like blackjack where you have a very tiny edge, and you need to make big bets, your bankroll has to be really big to prevent bankruptcy due to pure chance.

The good thing about the kelly criteria for avoiding bankruptcy is that if you get a group of bettors together, pooling the money will let them have a bigger bankroll. So for instance, suppose you have $5,000 and your edge is 1 percent, your max bet should be $50 to avoid bankruptcy, but at that rate you'll barely be making minimum wage.

If however, you get a group of 30 people together and pool all their money, each putting up $5,000, you'll be able to place a maximum bet of $1500 at 1 percent odds in your favor what is more, if you are using a big bettor, you will normally be able to keep a deck with a good count up for several more hands than otherwise, maybe double or triple.

So, if you decide to go into bankjack and do it solo, you will be making somewhere around 100 times less money per hour than going into a group.

If you decide to go solo, you should play poker, because in poker, you get a much better edge, so you don't need as big of a starting bankroll to make a decent living.

[Jim Walsh]

However, it's also boring

i considered professional gambling as a career once until I decided it was too much like work. Blackjack is boring as hell - pure numbers. Plus, like any house game, it's a sucker's bet; the house will always find a way to change the rules to their advantage. Invent a better mousetrap, they build a bigger mouse.

At least with poker, you have the human factor...

[eric taylor]

in my experience jim walsh, the house doesn't care if people win money. In fact, they kind of like it when people sometimes win. A guy sitting at the blackjack is fine making money.

that's because they make so much more money from the suckers, the slots especially. The blackjack the poker, those games arent money makers, they are enticements for slots players so they choose that particular casino. Everyone casino would make a bigger profit if they cut all card games and went to slots, but they dont because the card games are a draw and give slots players variety.

The real reason that casinos dont like blackjack players is because they are taking up space in the casino that could be occupied by real gambler.

[wizard of oz books]

With many new announcement about the wizard of oz movies in the news, you might want to consider starting to obtain Wizard of Oz book series either as collectible or investment at RareOzBooks.com.