When games of chance become games of skill, and the house gets an attitude

I don’t understand how “card counting” can be a problem for the casino or a plus for the player?

After a hand, aren’t the used cards pretty much “randomly” picked up by the dealer and placed on the bottom of the deck? Doesn’t this mean that the used cards at the end of each game go through, if you like, a sort of unoffical “semi shuffle”?

I woud imagine that a goodly portion of the deck may be dealt out in one hand and - given that a semi shuffle occurs at the end of each hand - why isn’t reliable card counting rendered impossible?

Legion,

In the Vegas casinos where I have played black jack the game goes as follows.
The deck is shuffled. Games are played by dealing from the top of the deck. At the end of the games the used cards are put in a pile. The next game is dealt from the top of the deck. This is repeated until the deck is about 1/2 to 3/4 used. Then the cards are gathered and the deck is shuffled again and play goes on. Black jack hands are typically only 3 or 4 cards per person plus the same for the dealer. So for a table with 5 people you can get 2 or 3 games per deck.

Casinos are for making money for casino operators through stacked odds.

They are not community based organisations set up to provide a neutral venue for sportspersons.

It never ceases to amaze me the way people seem to post in a manner that suggests they do not understand this.

Enlightenment, like Gazpacho, is best served cold! :slight_smile:

This is a point that bears repeating: There are a lot more people who think they can count than there are who actually can. It’s good advertising for the people to know that there’s a way to bring the odds into their favor, and the money lost to real counters is probably a reasonable price for that advertising.

They do indeed, and in fact I prefer playing at a table using an auto-shuffler than a shoe. Every time you hit the stopping point of a shoe, the game stops for what seems to be 10 minutes (just a guess on my part and probably too long). I think the last time I played using a shoe, the casino was using a yellow stop card inserted somewhere in the shoe by one of the players and I don’t remember what happened when you hit that stop card. I know that that signaled that the current game would be finished and then the shoe reshuffled (which takes forever with 6 or 8 decks, as I said), but I can’t remember if it counted as an automatic bust for the player that got the stop card or not. I’m not sure how many decks they use in an auto-shuffler. However, one night when I was playing at the table for three hours straight (I had a good night, too) play had to stop for a while when a card would get jammed and torn in the machine. Then the card would have to be extracted and switched out with a new one, all with the pit boss watching, of course.

As far as I can tell, that would effectively neutralize card-counting and it works for me and the system I use, which is just to try the make the mathematically-best choice every time as more of the deals are closer to completely independent trials.

The stop card just signals that they’ll reshuffle after completing the current round. It has no effect on anybody’s hand. That might be cool though, you could bet extra for the “stop bonus” and if you get the stop card you get 10 to 1 or something.

Wholeheartedly agreed.

My wife and I were playing a few weeks ago next to an elderly man who, while otherwise pleasant and engaging, was giving us tactical advice based on his “count.” And not just advice, but interfering to the point where he would push my wife’s chips back to her stack when she attempted to double on an 11, saying “you don’t want to do that now. Trust me.”

And then, of course, she would get a ten on her next “normal” hit.

He never admitted to counting, but it was clearly his “system.”

After that kind of advice proved costly more than once, we opted to just leave the table.

It amazes me that people find it acceptable that the house should disqualify people who are, in reality, following the rules. It is the moral equivalent of disallowing non-counting optimal play because the punter is playing her optimal strategy.

Frankly, I don’t understand why they would get kicked out except for publicity. According to Epstein the advantage with optimal counting is still pretty small. Not enough to be a real threat, especially given the remarks about how many people either don’t count or incorrectly believe that they can.

I am skeptical that there is any real profit threat to the casino. I suspect that it has more to do with neighboring players seeing someone getting kicked out for “winning too much” and furthering the myth that one can reliably beat the house.

IIRC, the optimal non-counting strategy is still advantageous for the player, by some ridiculously small amount. Nevertheless, if we take the counting = profit loss excuse, then one exposes oneself to the fact that optimal play is itself inherently unprofitable for the house—yet I’ve heard that dealers are even encouraged to coach players in optimal play. It’s my understanding that some houses even offer cards with the optimal blackjack strategy printed right on it.

When I hear that the prohibition against card counting is premised on lost profits, I can’t help but be extremely skeptical, because blackjack is inherently biased against the house and with millions of players betting billions of dollars every year, even non-counting play guarantees a net loss to the casinos. But, when throwing out a player who is just “too good” every now and again is going to spread the myth that phat stacks of mad cash can be made playing blackjack, it seems more than reasonable that profits from players short on skill, long on hubris, or overfilled with booze who come chasing that myth would be considerable.

If counting is allowed, then counting is allowed. That means whole tables of people betting along with one good counter, because he can freely and openly suggest betting strategy. That means you get many many more people trying to count, since they won’t get kicked out for it. That means counting suddenly becomes easier because you don’t have to do it in secret or try to hide your bets.

You get that type of scenario, Blackjack suddenly becomes a pretty lousy game to have around, other games give the house a pretty nice edge, why would you want a game that is even (or worse) taking up space, players and dealers?

As it is today, they kick out the occasional obvious counter, most people don’t bother with it, and the game stays viable. Very different situation if it becomes openly acceptable. Eventually, they’ll all just go to auto shufflers, and all this counting stuff becomes moot.

Good play is openly encouraged because the house retains their edge, and players feel good about playing the game, so they play more. Nobody wants to keep finding out that they suck at the game. Blow a quick $100 in 20 minutes and you might just get up in disgust. Play well, and you may stick around to lose $200 over a night or two.

I recently picked up a copy of Scarne’s Guide to Gambling by the late John Scarne, who spent decades working at casinos setting up blackjack tables, and was the first person to correctly figure out the house “take” on blackjack (note: that was back in the '30’s–before then blackjack was almost entirely a game for illegal casinos). Scarne reckoned that, if everybody always lost at casinos every time, nobody would go anymore, save a few masochists. Casinos need to win, and win big in the long run, but they also need a few winners here and there to promote interest.

I think the casinos’ fear of card-counters might be misplaced. As you say, the margin of error for a good card-counter is razor thin. If the shoe is shuffed after 60% penetration or less, it’s hard to see how even the average card-counter is going to consistently eke out a win. I would have thought the negative publicity generated by a casino which consistently ejected card-counters would hurt them more than letting bad card-counters play on. Obviously I’m wrong, as I’m not running one myself…

Hmmm. That could be it, too. Throwing out an “average,” and, likely, losing card-counter, might plant an idea in someone’s head that card-counters can actually succeed. But, if it were to work, surely the casinos would periodically throw out card-counters and make maximum publicity about it–with the undercurrent argument of “maybe you might succeed where he failed.” After all, as Scarne put it, “Casinos love it when players come with a ‘system,’” because the “system” is almost always bound to fail. That doesn’t seem to be the case, though, as Bone’s story suggests…sounds like casinos want to keep the ejections as quiet as possible.

BTW, Scarne’s book is a fascinating work…out of date, to some extent, as he passed away in the 70’s, but as an in-depth examination of how casino and other forms of gambling developed, it’s hard to beat. I’ll tell you what, you’ll never play a carnival game again after reading it.

You recall incorrectly. If the player can double on anything, and the dealer has to stand on a soft 17, AND blackjack pays 3-2, a person playing basic strategy (the best strategy that doesn’t involve counting) is still at a 1% disadvantage to the house.

Which is why you’re hard pressed to find a game with all of those rules in favor of the player. The most common one, in my experience, has the dealer hitting the soft 17, then it’s only allowing double-downs on 10-11 and not doubling after a split, and some houses have even started paying blackjacks at 6-5 on low-limit tables.

-lv

js_africanus said:

The edge is small, but you get many hands per hour, and can put a lot of money into play. A very good counter MIGHT get a 1% advantage over the house. More likely half that, if he’s using camouflage and giving up some expectation. But if he’s betting an average of $500 per hand, and getting 50 hands per hour, he’s going to make somewhere between $125 and $250 per hour.

The big risk to casinos is not the ‘recreational’ counter, who might be betting $10-$50 per hand. It’s the professionals with huge bankrolls and the professional teams. If a team of ten counters and a million dollar bankroll enters your casino, they can do some significant damage to your bottom line.

Casinos will usually leave counters alone if they are betting an average of under $100 per hand, for exactly that reason. The amount they make is small, and they are good advertising. As a result, it’s quite easy to make $10/hr counting cards. As you increase your betting limits, you will get increased ‘heat’ from the house. To stay in the game in the face of that heat means learning how to ‘camouflage’ your play, and learning the discipline to ‘hit and run’ and not stick around long enough for them to get a handle on your play. This makes professional card counting a pretty tedious existence. The ones who are really good at it and who have managed to stay in the game a long time travel a lot, and spend more time walking between casinos and studying games than they do playing.

Nope. Other than in a few rare situations. Occasionally you’ll find a low-limit single deck game where the odds are very slightly in the player’s favor with perfect play. But the limits will never be high enough to make significant profit from it.

But the vast majority of games have house advantages of between .3% and .8% with perfect basic strategy.

Sure, the casinos give out basic strategy cards, because A) players rarely follow them, and B) the house still has an edge.

Speaking as someone who was a professional card counter for some time, this just isn’t true. There is profit to be made. It’s just hard to do, and most people who think they are winners aren’t. But the casino is genuinely fearful of counters - so much so that they spend millions of dollars a year on new countermeasures and increased security staff. And they are so good at spotting counters that big-limit counters have really been forced to modify their strategy substantially to survive.

When it comes to Blackjack, Scarne didn’t know what he was talking about. His basic strategy was wrong, and when the first counting systems came along Scarne claimed they were all garbage and that counting didn’t work - until it was proven that it did, at which point Scarne tried to claim that he invented it.

Playing Optimal Basic Strategy (without counting) will in the long run never amount to an advantagew rfor the player. Ever. That’s how casinos make money. The best you can hope for while playing perfectly and with the friendliest rules you can find is about a .20 advantage for the house (they profit on average 20 cents for every $100 bet ( or is it $2 ? damn, I can’t remember. A little help?))

Dealers try to help inexperienced players here and there just to be nice (And to get better tips if the advice leads to a win). The PR value is higher than any extra money a non-knowlegable player would gain from advice from a only-slightly-more-knowlegable dealer. The thing is, most dealers have little or no understanding of perfect basic strategy (or they do understand it, but believe that hunches and beliefs are a better way to bet).

Most casinos sell little basic strategy cards in their gift shops and most will let you bring those cards to the table. Problem is the cards just have an “average basic strategy” printed on them. Not the optimal basic strategy custom tailored to the particular house and table rules.

Actually, what Scarne claimed (at least in the version of his book that I’ve got) is that Thorp, the “beat the blackjack” professor who may or may not have invented counting (AFAIK Thorp too may have borrowed ideas from others), was employing less mental means to win–that two of his confederates had dealers in their pockets, and bribing them to tip their hole cards. The version of counting Scarne describes that he “invented” (and, you’re right, he came after the other counters) is pretty primitive.

I have to ask something, though, based on one of your comments, of another poster. Bone, when you were barred from playing blackjack, what were your bet sizes?

In responding to my post you provide the perfect illustration of the attitude I am amazed by.

You talk of morals and rules in the context of some of the most openly venal business interests on the planet. What have these concepts to do with it?

Casinos are not set up to be moral, they are set up to make money out of punters. Casinos are not places in which obeisance is paid to abstract rules for the sake of some sort of chivalrous ideal. Casinos are not the local amateur bridge club.

And besides which it seems to me that you are without a doubt flat wrong: the people disqualified are not in reality following the rules: the rules in question are the rules of the casino (not the rules according to Hoyle) and those rules will say that the casino can to kick out card counters (or indeed probably whoever they want for any reason they want, for that matter).

Your attitude, mistaken as it is IMHO, is why casino owners are filthy rich. Yes, no doubt casino owners are not against the idea of punters having the mistaken attitude that you have about their “fairness”. But that attitude is one fit for mug punters, not GQ.

That’s not entirely accurate. Casinos make money, yes, but typically it’s only about 5% of all the money that circulates throughout their property. 95% of it goes into Paybacks, as in, letting people win. If people didn’t win often enough, they wouldn’t come to gamble.

Not many people come OUT of a casino and have MADE money or even kept their original sum, but that’s not because they lost all of it hoping for the big payoff.

People who DO win a lot, be it slots or blackjack or whatever else there is to play are encouraged to spend it in. A very tricksy lot, they are.

For instance, the layout of the casino is deceptive. There is no clearcut aisle or way to the exit. It’s a directionless maze intent upon keeping people inside. Notice also that there are NO clocks, there are no windows. No way for you to tell time short of bringing your own watch. Next, say you’re playing the slotmachines and you win a golden jackpot. In some casinos i’ve witnessed waitresses flocking to your sides when you win to offer you drinks and snacks on the spot, readily taking your winnings from you… It’s noisy too. The bin clatters loudly, full of coins, bells and whistles go off signaling a victory.

They thrive on people being silly with their money, casinos. I’ve known people that can make money off of casinos too. It’s not hard, you just have to keep calm and level headed when you win. Don’t spend your winnings. Keep them.

I have a question for Sam Stone, or anyone else knowledgable about counting.

I first read about card counting in one of William Poundstone’s Big Secrets books. Poundstone wrote a brief expose of Ken Uston’s (sp?) team play techniques, among other tidbits of counting info.

Anyway, later on, when I played blackjack in a casino, I tried a super-simple method of counting. All I did was keep a tab of how many 10-through-King cards had been dealt. IIRC, the games were dealt from four decks, with the shoe being about two-thirds of the way down the combined decks.

The thing was, I found it very easy to keep track of the value-10 cards as I played, chatted, ordered drinks, etc. I wasn’t anal about it – if I missed some cards, I didn’t make an effort to figure out what those cards were. But I wasn’t missing cards all that often … my count was pretty good, and I think it helped me sometimes. It’s not like I kept a betting log, though. But I remember more than a few times walking away from the $5 or $10 tables ahead about $20-40 or so. Other times I ended up down to the house, while some other times (rarely) my take would be just over a bill.

The question is this – if I’m able to count the 10-value cards more or less effortlessly, am I on the right track to successful counting on small-stakes tables? Or would I have to be more exacting in my counts somehow to benefit more fully?

Sam, I didn’t know you played blackjack for a living! How long did you do it for?

Bet sizes varied widely, as they have to in order to maintain the return. The general rule was that on a single deck, there needed to be an approximate 4 to1 ratio between big and small bets. At double deck, the ratio needs to be 8 to 1. Six deck shoes needed 16 to 1, or thereabouts. The reason the increase in bet size was required was because of the increased statistical variance which caused bigger swings. This means that on a typical 5-1000 table, a 10 starting bet, going to a max of two hands of 75 at true of 3 or better worked. At negative counts, a bet down to 5 is necessary. A person betting as little as $5 and then expanding to two hands of 75 tends to draw attention, thus as Sam remarked, a certain amount of camouflage was necessary. A lot of times, me and the people I went with would just watch tables, jump in for a few hands when it was good, and then leave. It involved a lot of standing around and doing nothing. At these bet sizes, I think rule of thumb says that you need 50 big bets as a bank roll to reduce the “risk of ruin”. I think that number is actually a bit low, as we maintained about 70 big bets. Never went down to zero, but more than a few times went home feeling like crap.

Casino’s allow the little color coded chart for basic strategy. It’s pretty funny actually. Often times as a dealer I’d see people with the card, and the player would choose to do something that I knew was against basic strategy. I’d ask them, “Is that what your card says?” “No, but I have a feeling!” Funny stuff.

However, if you see someone with a cheat sheet for card counting, it would look similar to a basic strategy chart, except much larger. It would have about 4 tables, with +xxx and –xxx numbers on it. Hit if below, Stay if above. So not only do you have to maintain an accurate running count, you have to remember the chart of actions based on that running count. You also have to take the running count and divide by the remaining number of decks, to get a true count. Thus, in a multi deck game, the running count is generally reduced. However, in a single deck game, after the first hand, dividing by the fractional remaining deck, multiplies the running count. This can lead to wilder fluctuations from hand to hand.

I actually learned to count in an environment that Cheesesteak supposed. It was a local Indian Casino. At the time, the casino did not discourage open card counting. The legal age for gambling there was 18. The winnings and losings from players was maintained in a house “pot”. Based on the structure of the house, if players won or lost, it didn’t effect the house take. This is because they charged a “collection” on each hand played, win or lose. The collection went up incrementally as bet size varied. By doing this, they eliminated the risk of big winners on the bottom line. To them, card counters were of no effect. This was a great environment to learn in, since players could openly converse about strategies and keep tabs on what count was to check against each other. The odds and return change based on collection, but even with that added cost, the expected return was still very positive due to other factors the house implemented. The biggest of all was that blackjack paid 2 to 1, rather than 3 to 2. This made a huge difference. There were other miscellaneous “bonuses”, but that was the biggie. In that environment, with a lot of cash, and time that should have been spent studying in college, I learned to count.

Either a study of statistics, or faith, but there is money to be made in counting, as long as you don’t get caught.

Bone: I’ve counted for pocket money since about 1980, but around 1993 I decided to go ‘full time’. I played blackjack for about six months, during which I discovered that poker was more profitable at the limits I played, and you didn’t have to skulk around the casino. So I played poker for a couple of years before finally getting bored with it and going back to work. I thought I had found the perfect job, but it was amazing how rapidly I got tired of just sitting at a smoky table playing cards.

I still play blackjack occasionally, and poker more often, but I’m happy actually creating stuff for a living again.

bordelond: The ‘tens count’ you use no doubt reduces the house edge, but probably not enough to make you a winning player. You might be close to breakeven with the house, though.

For shoe games, a highly effective count is the simple plus-minus count. Here’s how you do it - count the 2-6 cards as +1, face cards and Aces as -1. The 7,8,9 are neutral. So all you have to keep in your head is one number.

For ease in counting, learn to count in pairs. For instance, if the dealer deals out a 2T, you should recognize instantly that that pair has a count of zero because they cancel out. Hands like KQ have a count of -2, and a hand like 23 would have a count of 2. Practice with a deck of cards, flipping over cards two at a time and counting the pair. Eventually, it should be pretty much instant. When I was practicing, I could glance at a 7-player table and count all the face-up cards in a second or two while maintaining a conversation with a pit boss and not having him realize I could do it. It’s really not that hard.

So now that you’re keeping a count, there’s one more step you need to do when deciding how to use the information: You want to divide your ‘running count’ by the number of decks still left to play, to give you a ‘true count’. So if you’re playing a 4-deck shoe, and roughly one deck has been dealt out, there are three left. Now let’s say the hands you can see are 2T, 23, 55, 4A, AQ, 44, 68, and the dealer is showing a six. What’s the running count? Start at zero, and count the pairs and keep the sum only in your head, like this: 0, 2, 4, 4, 2, 4, 5, 6. You should be pretty much saying that in your head. So you have a running count of six. But you look in the discard, and estimate that one deck has gone by. So there are three decks left, and the true count is 6/3, or +2.

Depending on the rules, a true count of +2 to +3 is the dividing line where the odds shift in favor of the player. So what you want to do is bet the house minimum whenever the count is below that, and the table limit when the count goes over that value. But that’s pretty easy to spot, so you might want to double up at +3, and double up again on the next hand if the count stays high, and do it again. After all, losing gamblers often ‘let it ride’, so it’s not necessarily suspicious.

The other half of a counting system is learning to vary your playing strategy based on the count. For example, you should stand on 16 vs a dealer’s ten at any positive count, and hit at any negative count. You should always take insurance at counts greater than +3, and never take it below that number. There are 18 basic strategy changes that you should memorize - after that, the return for memorizing the more obscure runs into diminishing returns.

Really, that’s all there is to counting. It’s not hard. I could teach anyone how to count cards in a weekend. Poker is much, much harder.

That’s not to say that there aren’t a lot of things outside of the counting system to learn to make you a successful counter. You need to learn how to recognize good and bad games. You need to learn how to calculate your bet sizes based on your bankroll size and the Kelly criterion. You need to learn how to hide the fact that you’re counting. You need to learn the discipline of never deviating from the math. But none of this is difficult.