What Games Can You Beat Long Term? by Andrew N.S. Glazer
Some casino games can be beaten long term, and some can't. Games that you can't beat over the long run include some pure chance games like craps, slots, roulette, and baccarat, and some that mix chance and skill, like three-card poker or Let It Ride.
Even though you can't expect to beat these games long term, you can certainly win once in a while, so you should still learn as much as possible about these games. A bad craps player will lose a LOT faster than a good one, and will have to get much luckier to have a winning session. These games can be a lot of fun, and as long as your losses bear a reasonable relationship to the amount of fun you're having, well hey, other hobbies cost money, too.
But there are several casino games that CAN be beaten over time, where someone who wants to make the effort can learn enough to gain a long-term edge. Although the path to victory varies from game to game, the reason always winds up as your ability to take advantage of available useful information, and I emphasize useful.
Casinos bar blackjack card counters, but at roulette tables, they post electronically the results of the last 20 spins, and at baccarat tables, they provide you with pads and pencils to record the results. What does this difference in attitude tell you about the casino's opinion of the value of information in blackjack, roulette, and baccarat?
I break these beatable games into three groups:
1) Games where you're facing the casino's money (blackjack).
2) Games where you're really playing against other members of the public, even though the casino is handling your money (the pari-mutuel games like horse racing, dog racing, and Jai Alai; poker; and casino tournaments).
3) Hybrid games (sports betting and video poker) where in some ways you're playing against the casino and in some ways you aren't.
There are other games of mixed luck and skill that you can beat, like gin and backgammon, but you won't find these games in casinos.
GROUP ONE: YOU vs. THE CASINO
Ever since Edward O. Thorp wrote "Beat the Dealer," Americans have understood that card counters can beat blackjack. Why? Because available information about what has already happened (the cards already dealt) can provide reliable information about what may happen in the future.
To demonstrate with an impractical example, suppose you were playing heads up blackjack in a casino that didn't burn any cards and dealt down to the very last card, and you knew the six cards left in the deck were four eights and two sevens. You should bet the table limit, because you can't lose. No matter what two cards you get, you stand, and no matter what two cards the dealer gets, he has to hit, and then busts.
Granted, that particular situation isn't going to come up, but it does demonstrate how tracking dealt cards (and as a result, knowing what cards are left) can provide an advantage. Let's take a more plausible example. Suppose you're playing single deck blackjack, and on the first hand all four aces come out. You instantly know your chances of getting a blackjack during the rest of that shuffle are zero. Because blackjacks are very player-favorable (when the dealer gets blackjack, he just wins, but when you get it, you win 3-2), you wouldn't be very interested in making large bets until the next shuffle.
Extend this to the nth degree, and you can see what card counters do. When the deck is unusually rich in face cards and aces, they bet more, and when the deck is relatively poor in these cards, they bet less. The card counter's knowledge of what cards are out also impacts strategy. Hitting hard 14 is correct basic strategy against a dealer 10, but not when you know all the sevens are gone.
The problem with becoming an expert blackjack player is that if you start taking too much of the casino's money, they will either bar you, or (in jurisdictions where barring is not allowed) use other countermeasures that destroy most or all of your edge, such as restricting you to flat bets or small bets.
Contrast this use of "the past as a guide to the future" in blackjack with similar but misguided approaches in craps or roulette. The color red may have come up seven times in a row at the roulette wheel, but the ball doesn't remember that. The chances of red hitting on the next spin are still exactly the same: 18 in 38.
Dice have no memory either. That streaks will occur is a mathematical certainty, but when they will occur, and for how long, is a complete mystery. There are two possible exceptions. The first is broken equipment: an unbalanced roulette wheel, or unbalanced (loaded) dice. Unless you're a cheater who introduces these broken mechanical devices into the game, you could spend your entire life wandering through casinos looking to find this "edge" without ever doing so.
The second exception is pretty controversial. Some authors claim that lazy croupiers can develop "signatures" (continually spinning the ball and wheel at the same speed), allowing players with lightning reflexes to "read" this information and take advantage of it. I don't buy it, but debunking that claim is another story. Similarly, some crapshooters claim to have such accurate throwing arms that they can use "precision shooting" to impact the results, even though the dice must hit the back wall and its array of rubber pyramids. I don't buy this one either, not unless the shooter is a one-in-a-million athlete who has also somehow bribed all the game personnel to look the other way while he sets the dice and uses a "short hop" against the back wall.
GROUP TWO: YOU vs. OTHER PLAYERS
Casinos or tracks love pari-mutuel action on horses, dogs, or Jai Alai because they can't lose. In pari-mutuel betting, the odds paid to the winners are based entirely on the amounts wagered by the players. No matter what the "estimated odds" are in the racing form, in a two horse race, if the public bets $10,000 on Seabiscuit, and $5,000 on Fleabisquit, Fleabisquit will pay exactly twice as much to win as Seabiscuit, less, of course, the house "takeout."
The track doesn't care which horse wins. They pay out the same amount out to the winners either way. If a 100-1 shot wins the race, the track will be paying a lot of money to a very few bettors; that horse was 100-1 because very little money was bet on it. If a 2-1 shot wins the race, the track will be paying out small amounts to many more bettors. In either case, the track will withhold about 17% of the total amount bet (some of this goes to the track, and some goes to the state as taxes).
Because the track doesn't care who wins (it just wants an honest race so players return), and because a horse's (and to a lesser extent, a greyhound's or a Jai Alai player's) past form is a good guide to its future performance, it is possible to be a long-term winner at the track. You don't have to be smarter than the track (or casino). You don't have to worry about the casino putting heat on you. You just have to be smarter than the other bettors-much smarter, because you have to overcome that 17% bite.
While this is possible, it is extremely difficult, and that's probably why the gambler's prayer ("Lord, please let me break even today, I really need the money") is usually associated with the horse track.
Casinos also don't care who wins in their poker games, because you're not playing against house money. The casinos just want games that are perceived as honest and beatable, to keep the tables full. The house bite is much smaller than racing's typical 17% pari-mutuel share.
Casinos charge players in one of three ways. In lower stakes games, they usually "rake" a percentage of each pot, up to a limit of 10% of the money wagered. In some games, casinos collect money from you once each full round (usually nine hands). In most higher stakes games, the casino collects a "time" charge, which might be as little as $5 per half hour per player (higher in bigger games).
For the true expert, poker is probably the most beatable game available (and for the beginner, probably the most difficult game to beat), because there is more information available in poker than in any other game. Not only is there mathematical information available about the probability of a player holding certain cards (information that becomes even more important in seven-card stud, where experts who remember all the folded cards know more about the chances of completing a draw), but there is also a virtually limitless amount of information available about one's opponents.
Most poker players have "tells" that give away vital clues to the veteran eye. Player X slams his chips triumphantly into a pot, and later is found to have been bluffing. A pro remembers that in the case of Player X, slammed chips may be a sign of weakness, (while for Player Z, this may be a sign of strength). A slight tremble in the hand, a shortness of breath,-each of these means something about each player.
At its highest levels, poker almost ceases to become a card game, and instead becomes a "people game" that happens to involve cards. World champions like Phil Hellmuth readily confess that their play is built mainly around their abilities to "read" how strong or weak an opponent is.
Luck can intervene in the short run, but in the long run, the best players take home the money in poker, and unless their presence scares other players out of the game, the casino has no problems with the great player sitting there hour after hour, paying his chair rental fee, and helping to keep the game going as the tourists enter and leave.
Do pros scare tourists away? Usually the opposite is true, as players who have no business playing against a Hellmuth or a Doyle Brunson will sit down just so they can tell their friends stories about when they played against "the great so-and-so."
The casino tournament format means that in a game like craps, you don't necessarily have to outperform the casino's built-in edge. You just have to outperform the other tournament players, and sometimes that means merely losing more slowly than your opponents. Knowledge of the best and worst bets on the table, or of optimal betting strategies against a large or small number of opponents, can give you a significant edge on the field.
GROUP THREE HYBRIDS: YOU vs. THE CASINO "AND" OTHER PLAYERS
Make no mistake: when you place a bet at a casino's sports book, and you win, you're taking the casino's money. If you win too frequently, a casino may eventually tell you it isn't interested in your business. Nonetheless, sports-betting gets hybrid status because the line you bet is heavily influenced by action from other customers.
Football is the easiest example. When Dallas plays Washington, the book posts a line like Dallas +3, and you can bet Dallas plus the three or Washington minus the three, but you must lay 11:10 odds either way. If you want to win $100, you must risk $110. To be a long-term winner against those odds, you must win more than 52.38% of your bets.
Is that possible? Yes. Is that difficult? Yes. The people who set the initial line are not predicting the game's outcome. They are trying to predict what number will draw relatively equal action on both sides. If the public bets 1.1 million on Dallas and 1.1 million on Washington, the sports book knows before kickoff that it will win $100,000.
If the point spread opens at Dallas +3, and everyone bets Washington (-3), eventually the sports book will move the line, sometimes to Dallas +3.5, sometimes much further. If the casino has a ton of action on Dallas, they will welcome your action on Washington -4, and pay you happily when you win, because a lot of people lost on Dallas.
If your knowledge of a sport is sufficiently strong-if you have figured out tendencies that most members of the betting public haven't-it is possible to win long term. However, in this age of almost limitless information access, it is becoming increasingly more difficult for the guy who watches SportsCenter every night and reads the sports section every morning to have any appreciable edge over the betting public.
Winning 52.38% of your bets might not sound hard, but in the long run, it's very tough, and it becomes even tougher if you give away significant money to costly "tout" services that promise winning selections. Some of these provide useful information, but most don't.
Video poker is another hybrid game. Unlike regular slots, you can tell, by examining the pay table on a video poker machine, exactly what its percentage return is. Certain types of machines, with perfect play, offer positive expectation when the royal flush progressive jackpot passes a certain number.
Because the large progressive jackpots aren't coming directly out of the casino's pocket-they've built up because of lots of play by paying customers, and HAVE to get paid out eventually-casinos don't put "heat" on sharp slot players who occupy these machines when play becomes mathematically favorable (although if a team or family tries to monopolize a bank of machines, heat is possible). That's the good news.
The bad news is that these positive expectation situations don't occur all that often, and you can pump a lot of money in trying to hit the royal when they do. If someone else hits it before you do, you're out of luck. A video poker pro needs a big bankroll to withstand these fluctuations.
THE COMMON THREAD
Some games are more beatable than others. Poker is the most beatable, because there is more useful information (mathematical, strategic, and interpersonal) available. The amount of skill necessary to win in blackjack is smaller (there's no interpersonal element), but the rules and the house's ability to bar players limits how much you can win. The more useful information available, and the less threatening you are to the casino, the better your chances of long term success become.
If you want to be a long-term "advantage" gambler, you have to pick a game where there is a lot of useful information available, and study and practice a lot. If that sounds more like work than play, you're starting to understand why many people who enjoy gambling are content to learn just enough to reduce the house edge while looking for thrills on their casino visits.
