Blackjack for Advanced Players
We've divided most of our blackjack pages into 3 categories:
In the section for beginners, we covered the basics of how to play, some of the simpler strategy ideas, and definitions of terms used in the game.
In the section for intermediate players, we covered what a player should learn next: complete basic strategy, game selection, and the basics of counting cards.
Now, in this section, we cover more advanced topics. These include advantage play techniques like card counting and shuffle tracking. We also cover dealer tells.
We think it's a good goal for a beginner to work toward lowering the house edge to 1% or so. We think it's a good goal for an intermediate player to get the house edge as low as possible (to 0% if possible).
But our goal for advanced players is for them to be able to play at a significant advantage over the casino.
And the only way to do that is to learn at least one, and preferably multiple, advantage play techniques.
Advantage Play in Blackjack
Advantage play is a means of getting an edge when gambling. It's distinct from cheating, because advantage play doesn't involve changing the conditions of the game. Advantage players just use better strategies than the average gambler.
Most casino games don't have a means for you to get an edge. Some games might or might not have a means for you to get an edge. Blackjack is one game where you certainly can get an edge, but let's talk about some other games and why you can or cannot get an edge playing them.
We like to start such discussions with roulette, because the math is easy to understand. An American roulette wheel has 38 numbers on it. Almost half of them (18) are red. Almost half are black (18). And 2 of them are green.
When you spin the ball in a roulette game, the probability of the ball landing in any particular slot is the same: 1/38. The probability of the ball landing on a black number is 18/38. These probabilities don't change based on what happened on the previous spin.
Every spin, the same 38 possibilities are there.
The way the house gets its edge is clear.
Suppose you bet $100 on black. 18 times out of 38, you'll win even money on such a bet. That's $1800 in winnings.
But 20 times out of 38, you'll lose $100. That's $2000 in losses.
The difference between the $2000 in losses and the $1800 in winnings is the house edge. Average it out on a per bet basis, and you're looking at a net loss of $200 divided by 38 bets. That's $5.26 per bet on average that you lost.
All the bets at the roulette table have this same house edge. It's usually expressed as a percentage-in this case, 5.26%.
But suppose we made a change to the game.
Suppose every time the ball landed in a spot, it was eliminated as a possibility on future spins.
The probability would change on subsequent spins based on what happened previously.
You're playing roulette, and the ball lands on black 3 times in a row. Each of those black numbers has been eliminated from the game.
Now you have 35 possible results. 15 of them are black. 18 of them are red. 2 of them are green.
The probability of a black result has changed to 15/35, or 42.89%. The probability of a red result is now 51.43%.
A bet on red is now clearly the smart play.
This is what happens when you're playing blackjack, isn't it?
Once a card is dealt, it's no longer in the deck, so the probability of getting a certain combination changes.
It's not likely.
The only legitimate advantage technique related to roulette that we know of is finding a biased wheel.
Since a roulette wheel is a mechanical, not electronic, device, it might have some mechanical flaws. It might be slightly warped, which means that certain parts of the wheel might have a slightly higher chance of being a result than others.
Such an imperfection would be invisible to the naked eye.
The only way to find such an imperfection would be to "clock" the wheel. In other words, you'd have to record the results of thousands of spins. Based on those results, you'd be able to detect any imperfections or bias.
We don't think this is a practical advantage play technique.
First, you might spend hours clocking a wheel only to find that it has no bias at all. Not all-or even most-roulette wheels have a bias.
Second, casinos don't leave roulette wheels in the same place forever. You'll eventually have to go home and get some sleep. At various times, the casinos move these wheels from table to table.
You might spend 16 hours clocking a table's results only to have that wheel get moved to another table in the middle of the night.
Slot machines are an example of a game where you can't get an edge.
Here's a simplified version of how the probability of a slot machine game works:
You have a game with 3 reels, and on each wheel, you have 10 symbols. Each symbol has the same probability of coming up.
The probability of getting a particular symbol come up on each wheel is 1/10 X 1/10 X 1/10, or 1/1000.
But the jackpot for that symbol is programmed to pay off at less than 1000 to 1-usually significantly less. You might get a payoff of 800 to 1 for one symbols, a payoff of 80 to 1 for another symbol, and a payoff of 8 to 1 for another symbol.
It's easy to see how in the long run the casino is going to come out ahead.
You have no way of affecting those probabilities on a slot machine without cheating. Remember, you can't change the conditions of the game. This means you can't use any kind of electronic or mechanical device to affect the outcome.
In fact, only a handful of gambling activities have advantage techniques available. These include:
- Sports betting
- Video poker
You can get an edge at poker by playing better than the other players.
But it's not enough to be just a little bit better than the other players. The house takes 5% of each pot as a rake, so you have to be so much better than the other players that you can beat the rake.
You can get an edge at sports betting by being better at estimating the odds and handicapping the games than the sports books are. This is unlikely, as the Vegas books use state of the art software and expert handicappers to set the lines.
And, like poker, you have to get enough of an edge to overcome the rake. Only in sports betting, it's not called the rake-it's called "vig". When you place a bet on a sporting event, you usually have to bet $110 to win $100. So you have to win your bets more than 53% of the time to get even a small edge over the book.
With video poker, you have to find a game with a pay table that enables you to get an edge over the casino. These are unusual, but they're not impossible to find.
But that's not all. You also have to be able to play that game using perfect strategy.
And even if you do, most video poker games with such pay tables are played for such low stakes that it wouldn't be worth your time, anyway.
Suppose you find a full pay Deuces Wild game. You have an edge of 0.76% over the house, assuming you're playing perfectly.
But the game is only available as a quarter machine. Which means you're betting $1.25 per hand.
An average player places 600 bets per hour at a video poker game. That means you're putting $750 per hour into action.
0.76% of $750 is $5.70 per hour.
No one's going to become a professional video poker player making $5.70 per hour.
Let's assume that you're good enough at your various advantage play techniques to get a consistent 1% edge over the house.
And let's assume you have a large enough bankroll that you can bet $50 per hand on average.
You stick with tables where there are other players, but not a lot of other players, so you're averaging 80 bets per hour.
That's $4000 per hour you're putting into action. With a 1% edge, you're expected to win $40 per hour.
That's a little more interesting than $5 or $6 per hour.
But if you get a large enough bankroll to withstand the swings of fortune, you could play for an average of $100 per hand or even $200 per hand.
Now you're looking at $80/hour in expected winnings or $160/hour in expected winnings.
That's a more attractive prospect.
Counting cards is the most obvious way, but pros have multiple means of getting an edge.
We cover the basics of the most popular advantage play techniques in blackjack below, but some of them are covered in excruciating detail in their own subsections on our site.
The premise behind counting cards isn't as complicated as you might think. Card counting works because of 2 things:
- 1You get paid off 3 to 2 if you get a natural.
- 2The deck has a memory.
These 2 factors are so closely related that it's hard to talk about one without talking about the other.
Here's the thing about a natural:
There are only 2 cards which can make a natural-the ace and the 10. (Of course, there are 4 different cards worth 10 in the deck-the 10, J, Q, and K.)
The deck has a memory, too-at least most of them do. There are exceptions, though.
If you're playing in a blackjack game with an automatic shuffler that's going continuously, you're playing from a fresh deck every hand. You can't get an edge at such a game.
But if you're playing in a game dealt by hand or from a shoe, the composition of the deck changes based on which cards have already been dealt. This can increase or decrease your probability of being dealt a blackjack.
Suppose you're playing in a single-deck game, and all 4 aces have already been dealt. Your probability of getting a blackjack is now 0%.
It's impossible to get a blackjack if there are no aces left in the deck.
On the other hand, if half the deck has been dealt, and none of the aces have appeared yet, and lots of low cards are already gone (like the 5s and 6s), your probability of getting a natural goes up.
Now let's suppose you raise the size of your bets when your probability of getting that 3 to 2 payout goes up, and you lower the size of your bets when your probability of getting a blackjack is lower.
If you're betting $100 when the deck has a proportionally higher amount of aces and 10s in it, and you're betting $10 when the deck has a proportionally lower number of aces and 10s in it, you're going to win more money, right?
It's similar to how tight-aggressive poker players manage their poker hands. When they have a hand which is more likely to win, they bet and raise aggressively. When they have a hand which is more likely to lose, they'll check and fold.
Bet more money when you have an edge. Bet less money when you don't.
That's a surefire way of getting an edge at gambling.
That's where the various card counting systems come in. All card counting systems share the same goal:
They try to estimate how much of an edge you have.
They do this by assigning a value to each card. You add or subtract that value from the count, and you raise and lower your bets according to that count.
Here's one of the most basic examples of such a card counting system. Every time you see a 5, add 1 to the count. Every time you see an ace, subtract 1 from the count.
When the count is 0 or negative, bet 1 unit.
When the count is positive, bet 1 unit + the count.
So if the count is -4, you'd bet a single unit. If the count is +4, you'd bet 5 units.
If you're playing for $100 on average per hand, you'd bet $100 if the count is -4, but you'd bet $500 if the count is +4.
A more accurate count is the hi-lo count, which is one of the most-used and long-standing counts. Here are the values for that one:
- All aces and 10s are -1.
- 2s, 3s, 4s, 5s, and 6s are +1.
- 7s, 8s, and 9s count as 0.
This is a little more complicated but a lot more accurate. It's an example of a balanced card counting system-you have just as many cards valued at +1 as you do at -1, so if you count through an entire deck, you'll end where you started-at 0.
But it doesn't stop there.
Card counting systems are aimed at single deck games. When you get more decks into the mix, the effect of a card being gone from the deck is lessened.
Think about it this way.
If you're playing in a single deck game, and 4 aces have been dealt, the probability of getting an ace as your next card is 0.
If you're playing in an 8-deck game, and 4 aces have been dealt, the probability of getting an ace is lower, but it's far from 0. After all, you started with 32 aces (4 in each deck). You still have 28 aces left.
To compensate for this dilution effect, card counters take the "running count" and convert it into a "true count". They do this by dividing the running count by the estimated number of decks left in the shoe.
They increase and lower their bets by using the true count.
Let's assume you're playing in an 8-deck game, and the running count is +3. You're not going to raise your bet until the count is +1 or higher. There are 6 decks left in the shoe, so the true count is 3/6, or +0.5. You continue to flat bet until the count improves some more.
As you can see, with more decks in play, it's harder to get an edge. Since you'll have an edge less often, you'll probably need to increase your betting range (the amount of your high bets compared to your low bets) to compensate.
A dizzying array of card counting systems are available. Each offers something different. Some provide a more accurate gauge of how much of an edge you have over the casino. Others provide you with a better gauge of how you should change your strategy. Some are easier to implement than others.
The goal is to find a card counting system that works for you. You might not be interested in adjusting your basic strategy decisions based on the count. You might just want to get your edge by increasing the size of your bets.
If that's the case, you'll want to find a simple system that does a good job of estimating your edge over the house.
We cover various card counting systems in detail on this site. Some of the more popular card counting systems we cover in detail include:
We cover dozens more.
But counting cards is only one tool in the advantage player's arsenal.
Sometimes cards clump into sequence or packs in a deck even after they're shuffled. If you can track these groups of cards throughout the game, you can get an edge over the casino. The simpler the shuffle, the easier it is to keep up with these groups.
The reason shuffle tracking works is because shuffling is an imperfect, inexact way of mixing up the cards. It's a mechanical process. You can't use shuffle tracking with an online casino, for example, because the random number generator really does randomize the cards. You'd have to hand shuffle a deck many times before approaching true randomness.
What a shuffle tracker does is combine card counting with "zones" of cards. Think of these zones as miniature decks within a deck. When the cards get shuffled, these zones tend to stay together. If you identify a zone that has a lot of 10s and aces in it, you can raise your bets when that zone comes into play.
If you know anything about poker, you've heard of "tells". These are nonverbal cues that give you a hint at what cards your opponent is holding. For example, in poker, if a player's hands tremble when he bets, he probably has a strong hand. This varies based on the individual, though, and reading other players is one piece of a poker player's strategy.
In blackjack, you can pay attention to the dealer's behavior as he looks at his hole card. This gives you a hint as to what he's holding. You can adjust your playing strategy accordingly.
In fact, some dealers are sloppy enough to expose their hole card when they're checking for blackjack. That provides an obvious clear advantage to the player.
Becoming an advanced blackjack player means learning enough about the game and how to play that you consistently have an advantage over the casino. This might involve becoming an expert card counter, but it can also mean taking advantage of other advantage play techniques like shuffle tracking and dealer tells.