Although counting cards can potentially get you banned by a casino, it isn’t actually illegal and is widely considered a legitimate strategy for increasing your odds of winning. Here we explain the rationale behind how and why card counting works.
In blackjack, there are multiple methods for counting cards. On this page, we will dissect some of the more common methods including the high opt 1 card counting system, the knockout card counting system, the omega 2 card counting system, the red seven count, the unbalanced zen count, the zen count and the Wong halves method.
Most casino players think of card counting as an esoteric skill requiring feats of superhuman memory and computer-like mathematical prowess. The reality is that thousands of players know how to count cards in blackjack, and most of them have average intelligence. Counting cards is one of the only ways to get a consistent mathematical edge over a casino, but it does require significant amounts of dedicated practice and study before you can reliably count on that edge.
Entire books and websites have been written about this particular method of advantage gambling. The purpose of this page isn’t to provide a comprehensive guide to card counting, but it does aim to provide a detailed discussion of the basics.
Edward Thorp published “Beat the Dealer” in 1962, and this date is considered the genesis of advantage play at blackjack. According to multiple sources, including Thorp, but also according to Stanford Wong’s “Big Book of Blackjack” and Andrew Brisman’s “Mensa Guide to Casino Gambling”, early blackjack players intuitively gained an edge over the casino game by using primitive counting techniques. The idea wasn’t popularized or a concern of the casinos until the publication of Thorp’s book, though.
The casinos’ reaction to the publication of “Beat the Dealer” was a comical overreaction. They were afraid that mobs of blackjack players would start taking them for all they were worth, and so they made dramatic changes to their house rules in order to make it impossible to get an edge. Before “Beat the Dealer”, it was common for a casino to deal blackjack from a single deck and to deal all the way to the bottom of that deck. This practice was eliminated, and several other rules designed to eliminate a counter’s edge were put into place. This backfired, though—the rules had become so restrictive that even the people who weren’t counting stopped playing blackjack.
The reality is that the casinos probably didn’t have much to worry about in the 1960s. Thorp’s methods were a little on the unwieldy side, and even though blackjack was the fresh new “beatable” casino game for intelligent players, most players didn’t study or practice enough to actually get an edge. What had happened instead was that a lot of clueless players started playing, which probably made more money for the casinos’ blackjack tables than they would have before. In fact, one of the reasons that blackjack makes up 50% of a casino’s table game revenue is that it’s considered a game that an intelligent player can beat.
Card counting systems in blackjack are a means of advantage play in which you gauge how favorable or unfavorable the deck is, and then betting more or less based on that estimate. This estimate is arrived at, not by memorization, but by a heuristic system which assigns a point value to various cards in the deck. Once a card has been removed from the deck, it either makes the deck more favorable for the player or less favorable for the player. This effect is tracked by adding and subtracting numbers from a mental tally called “the running count”.
Imagine that you’re playing roulette, and you place a bet on red, and you want to know what your chances of winning are. The calculations are simple enough; you simply divide the number of red slots on the wheel by the total number of slots on the wheel. On an American roulette wheel, you’d have 18 red slots, 18 black slots, and 2 green slots, for a total of 38 slots on the wheel. 18 divided by 38 is 47.37%, which is your chance of winning. No matter whether you win or lose on that bet on red, if you bet red again on the next spin of the wheel, your odds of winning are the same: 47.37%. They don’t change based on what’s happened before.
Now suppose you’re playing a card game similar to roulette, where the dealer turns over a card, and you bet on whether or not the next card is going to be red or black. The deck of cards has 18 red cards in it, 18 black cards in it, and 2 jokers. If you bet on red, you have a 47.37% chance of getting a red card. Now suppose that card is discarded, and you’re dealt another card. Have the odds of getting another red card changed? The answer should be obvious, because now there are only 37 cards total in the deck, and only 17 of them are red, so the odds of winning become 17/37, which equates to 45.95%. The odds of getting a black result have increased to 18/37, or 48.65%.
Now suppose you get red five times in a row. If you’re playing roulette, then your odds of winning your next bet on red is still 47.37%–they don’t remove red slots on the wheel when you’ve won. But in this imagined card game we’ve come up with, the difference is significant after five red wins in a row. The chances of being dealt a red card on the next play have gone down to 13 divided by 33, or 39.4%. Meanwhile, the chances of being dealt a black card on the next play have gone up to 18/33, or 54.54%.
Now suppose you only bet a single dollar on the first five bets, so you’ve lost $5. But now that your odds of winning (by betting on black) are more likely than not, you raise the size of your bet to $20. Your expected value is 54.54% of that $20, or $10.90. And if you kept playing long enough, you’d continue to find opportunities to make bets with a positive expectation.
That’s a remarkably accurate description of how card counting systems work, but instead of tracking the colors of the cards, you’re tracking whether the cards are high or low. In some systems, what constitutes “high” or “low” varies based on goals having to do with accuracy and ease of play. Either way, you’ll know when the deck is in your favor, because when the deck has lots of high cards in it, you’re more likely to get a 3/2 payout on a natural 21, and you can bet more to take advantage of the situation. You’ll bet less when the deck has lots of low cards in it, because the dealer is less likely to bust, which isn’t a favorable situation for the player.
Some of the most commonly used counting systems include the following:
Each card has a value to a blackjack counter, and that value is called “the effect of removal”, or “EOR”. The EOR for the aces and fives have the largest effect on a player’s chances. If you remove all the aces from the deck, then a natural 21 is impossible, which subtracts considerably from the player’s chances of getting ahead. On the other hand, if you remove all of the fives from a deck, the EOR is so great that it would make almost any blackjack game a positive expectation game for the player.
The simplest card counting system is called an “Ace-Five Count”, and you simply add 1 to the count every time a five is dealt and subtract 1 from the count every time an ace is dealt. The higher the count is, the more you bet. If you have a wide enough betting range, say anywhere from 1 unit to 10 units, this is a remarkably easy and effective system. Most people are looking for more of an edge than this simple count offers, though, and they use slightly more complicated systems. (Something to notice about this system is that there are just as many +1 cards in the deck as there are -1 cards in the deck—this is called a “balanced” counting system, as a result. Not all systems are balanced.)
All card counting systems estimate multiple factors related to their effectiveness. The first of these is called “betting correlation”, and it’s probably the most important one. This number determines how well the count estimates the player’s edge—this is important for making decisions about betting amounts. The second of these is called “playing efficiency”, and this estimates how well the count takes into account changes to basic strategy. This factor isn’t as important as it may seem, but players who want to squeeze every percentage point of value out of their blackjack card counting system pay attention to it. The third and possibly least important factor is the “insurance correlation”, which is an estimate of how well the system predicts whether or not insurance is a good bet.
We already discussed balanced systems versus unbalanced systems, but just to reiterate, a balanced system is one in which the count evens out to 0 when you count through the entire deck. For example, the Hi-Lo system, which is quite popular, is a balanced system that counts aces and tens as -1 and counts any card ranked 2 through 6 as +1. There are 20 cards in each category, positive or negative, and when you finish counting through a deck, you’ll have a total running count of 0. In an unbalanced system, this isn’t true.
The Hi-Lo system is also an example of a single level system, because the cards are all valued at either +1 or -1. A higher level system might give different values to different cards. For example, if a system counts aces as -2 and fives as +2, and then counts tens as -1 and 2s, 4s, and 6s as +1, then the system would be a multi level technique.
Some card counting systems go even further and require you to keep a side count of the aces that have been dealt. The more complicated the rules for the system are, then the more accurate it is, but you need to remember that your system should also be practical. You’ll be counting cards in a crowded and noisy casino, so you need to be able to concentrate without looking like your concentrating. Practice is necessary, but so is a simple enough strategy.
In almost all casino games, each wager is made on an independent event, so there’s no way to overcome the game’s insurmountable mathematical edge. Blackjack is an exception, though—the odds change with every card that’s dealt in a blackjack game. For example, if you spin the wheel on a roulette game, and you wager on black, then 18 of the possible outcomes are winners. On your second spin, that same wager on black has the same number of potential winning outcome: 18. In blackjack, though, the cards that have been dealt change the number of potential outcomes. For example, if all four aces have been dealt, the probability of being dealt a natural 21 have been reduced to 0%. (You can’t get a natural blackjack without an ace in the deck.)
As a general rule, a deck with a lot of high cards (tens and aces) left in it is better for the player, while a deck with a lot of low cards (twos, threes, fours, fives, and sixes) is better for the dealer. When the deck is rich in high cards, the player has a better chance of being dealt a natural, which pays out at 3 to 2. Perceptive readers might notice that the dealer also has a better chance of getting a blackjack, too, but the dealer doesn’t get a 3 to 2 payout—only the player does.
Some of the advantage from counting cards is had from making some simple basic strategy changes. For example, in a deck that’s rich in tens and aces, it makes sense to avoid hitting stiff hands. The dealer doesn’t get to make this decision; the dealer has to hit a certain total and stand on a certain total, regardless of whether or not it’s the right play given the composition of the deck.
Another strategy change with a deck that’s composed more than 2/3 of high cards is to take insurance. Normally that’s a sucker bet with a huge negative expectation, but not when you’re counting cards and know the score.
When the deck is rich in lower ranked cards, the dealer is less likely to bust, making it harder for the player to win that way. It’s also less likely that a player will be dealt a natural when there are still a lot of low cards in the deck.
The bulk of the player’s advantage comes from betting more when the deck is rich in high cards and betting less when the deck is rich in low cards. The 3 to 2 payout for a blackjack is where the counter gets most of his edge. Most counters range their bets from 1 to 5 units based on how favorable the count is, but some even range their bets from 1 to 10 units. The casinos watch players who ranger their bets, though, so don’t be surprised if you don’t get some “heat” from the casino. Professional blackjack players use various means of “camouflage” to disguise the fact that they’re counting, and they even sometimes work in teams. You can see examples of camouflage in action in the movie 21, which shows one way in which a team takes advantage of a hot deck.
You don’t have to memorize all the cards that have been played in order to count cards. Card counting uses a heuristic method of tracking the ratio of high cards to low cards. The Hi-Lo system is one of the most common ways of doing this. It assigns the following values to the following cards:
You start your count at 0, and then as each card is dealt, you adjust the account by either adding 1 or subtracting 1. That running total is called the “running count”, but the running count alone isn’t enough to make you profitable. Most casinos use multiple decks, so you have to convert the running count into what’s called the “true count” in order to make the correct decisions.
In order to calculate the true count, you first estimate how many decks are still left in the shoe (the device which holds the cards for the blackjack dealer). Then you divide the running count by the number of decks left. For example, if your running count is +4, and you estimate that 4 decks remain in the shoe, then the true count is +1.
The higher the count, the more you bet. Competent counters can reduce the house edge, in which the player has a disadvantage of 1% or 2% to the casino, to an edge of 0.5% to 1.5% over the casino. If you’re playing for $20 per hand on average, and you play 100 hands per hour, then you’re putting $2000 per hour into action. With a 1% advantage, that means you can expect, in the long run, to earn $20 per hour for your efforts.
Keep in mind that this number isn’t guaranteed, either. In the short run, anything can happen, and even card counters with large bankrolls can go broke quickly if luck turns against them. Besides short term fluctuations in the mathematics of the game, casinos will often refuse to let you play blackjack. In Atlantic City, it’s illegal for a casino to bar a player from blackjack, but they’ve made the rules there so tight that it’s practically impossible to get an edge over the casino.
Other card counting strategies offer varying degrees of complexity in exchange for varying degrees of accuracy. Most players look for a balance of practicality and play-ability when choosing a system. Other pages on our site look at the specifics of these other counting techniques.
The Hi Opt 1 card counting system is so named because it’s considered “highly optimum”. This is one of the oldest and most traditional card counting systems, and it’s still in common use among advantage players today. Charles Einstein formulated this system in 1968, but it was further improved upon and publicized by Lance Humble and Carl Cooper in the book The World’s Greatest Blackjack Book.
You can skip this section if you already understand the basics of counting cards, but if not, this is how it works in a nutshell. Some cards in the deck improve the player’s odds. Other cards in the deck improve the casino’s odds.
For example, players get an automatic win and a 3 to 2 payout when they’re dealt a “blackjack” (or “natural”). Since that’s the case, 10s and aces are favorable cards for the player. You could never be dealt a blackjack if all of the aces or all of the 10s were removed from the deck beforehand, so it just makes sense that if there is a higher proportion of those cards in the deck then normal, you’ve got a better chance of a big win.
Card counting enables players to estimate their relative advantage or disadvantage by keeping a running tally of high cards versus low cards. When the deck still has a lot of high cards left in it, card counters raise their bets in order to take advantage of the better odds they’re being offered.
When explained that way, card counting sounds pretty simple, doesn’t it? Just keep in mind that you have to keep up with this math at an almost lightening-pace in a loud, distracting environment. It’s not as easy as it sounds.
Different card counting methods apply different values to the cards.
This is a traditional single level, balanced counting system. That means that you only have to add 1 or subtract 1 to the count for each card, and if you count through the entire deck, you’ll wind up with 0 at the end of your count. (There are as many +1 cards in a deck as there are -1 cards in a deck.)
For purposes of this count, you’ll add +1 to the count every time you see a 3, 4, 5, or 6, and you’ll subtract 1 from the count every time you see a card with a value of 10. All other cards count as 0 for purposes of keeping up with the count.
You should practice your counting skills at home before trying this in a casino. With a balanced system like the Hi Opt I, it’s easy to see how accurate you are, because if you count through a deck and wind up with a total other than 0, you know you’ve still got some work to do.
As with other counting systems, you raise your bets when the count increases, and you lower your bet when the count is low. This is where the bulk of your edge from a card counting system comes from.
The count can also affect your strategy decisions, such as whether to hit, whether or not to double down, and whether or not to take insurance. You can get an edge over the casino by just sticking with basic strategy and raising and lowering your bets based on the count, but you increase your edge by making the appropriate strategy adjustments.
One aspect of the Hi Opt I system is that it requires a conversion from the running count to the true count in order to remain accurate. This takes into account the number of decks you’re playing with.
The reason for this should be obvious, but a simple illustration should clarify the concept. If a single deck has 4 aces in it, and 3 of those aces have been dealt, then 75% of the aces are gone. But if you’re playing with a shoe with 8 decks in it, then you started with 32 aces, so you have 29 aces left. Only about 10% of the aces have been dealt.
To adjust for that, you divide the running count by the number of decks you estimate are still left in the shoe.
Most of the players at the blackjack table are playing at a disadvantage to the house of between 2% and 4%. That means, over time, that they will lose $2 to $4 of every $100 they wager. On the other hand, if you’re using the Hi Opt I system and basic strategy, you should be able to maintain an advantage over the house of around 1%, which means you’ll win, on average (eventually), $1 for every $100 you wager.
Casinos ban card counters, so you want to disguise the fact that you’re counting cards as much as you can. One way to disguise your skill level is to occasionally make basic strategy mistakes. In other words, you’ll play a hand wrong some of the time.
Which hands do you play incorrectly, though? Here’s a simple hint—use correct basic strategy when you’ve raised your bets. Use correct basic strategy most of the time when you’re betting your minimum. Only once in a while should you play incorrectly, but that should always be when you have a lower amount of money on the line.
In fact, it’s not a bad idea to keep a basic strategy chart with you at the table. This sends an unspoken message to the casino that you’re a novice. After all, a card counter would have memorized basic strategy and wouldn’t need a strategy chart, right?
The Hi Opt I provides an excellent balance between being easy to learn and effective at the tables. It’s worth learning, and if you’re interested in becoming an expert in this method, The World’s Greatest Blackjack Book is indispensable.
The KO Count is one of my favorite blackjack card counting systems. KO, of course, stands for “knockout”, and the system is also sometimes called the “knockout blackjack system”. The reason I like the KO count so much is that it’s easy to use, and it eliminates the need for converting the running count into a true count.
The KO Count works in a similar manner to all other card counting systems. Counting cards doesn’t require memorizing specific cards that come out of the deck at all. Instead, it uses a heuristic system to estimate the proportion of high cards to low cards left in the deck. When a deck has a proportionately higher number of 10s and aces in it, it’s more profitable for the player.
The reason for this should be obvious, but if it’s not, think about this. One hand in blackjack results in a higher payout than any other hand. That’s a “natural” 21, which is also called a “blackjack”. Only 2 values of cards can combine to form such a hand. One of those is the ace, and the other is then cards with a value of 10.
If you removed all of the aces from the deck, your chances of getting dealt a natural would become 0, right? So obviously, if the deck has fewer low cards and more high cards, you’re more likely to get dealt a natural.
And when you ARE dealt a natural, you get paid off at 3 to 2. So if you increase the size of your bets when you’re likely to get a 3 to 2 payout, it stands to reason that you’d have a better chance of winning more money, right?
In the knockout system, every card with a value of 2, 3, 4, 5, 6, or 7 is going to count as +1. All the aces and 10s count as -1. There are 24 of the former and 20 of the latter, so if you counted through an entire deck using this system, you’d have a total of +4 when you finished. (This is called an “unbalanced” system as a result.)
When the count is positive, you’ll increase the size of your bets. When it’s negative, you’ll lower your bet to the minimum. The higher the count, the more you bet.
It’s almost just that simple.
There’s one additional wrinkle though. You have to take into account how many decks are in use.
In most card counting systems, a player has to convert the running count into a true count in order to adjust for the additional decks in play. For example, if you’re playing in a single deck game, and all of the aces have been dealt, then you now have a 0% chance of being dealt a blackjack. But in an 8 deck game, when four aces have been dealt, you’ll still have 28 aces left in the deck.
The effect of each individual card is diluted by the large number of decks in the shoe. The conversion from the running count to the true count requires some division and some estimating. The formula is simple enough—you just divide the count by the number of decks you estimate are left in the shoe.
In the KO system, you can skip the division. This is one of the reasons that the system isn’t balanced. The other quirk about the KO system is that you don’t always start your count at 0, as you would in other systems.
Your starting count in the KO system is determined by the number of decks in the shoe. If you’re playing in a single deck game, your initial count will be 0, but if you’re playing against 2 decks, you start your count at -4. With 6 decks, you start your count at -20, and with 8 decks, you start your count at -28.
You’ll need to decide on a bankroll and a betting spread before you play. A common betting spread is 1 to 5 units. So you might start with a bankroll of $10,000 and have a betting spread of $100 to $500.
On hands where the count is negative, you’ll stick with your $100 bets. As long as you’re using basic strategy, you’ll only be playing at a disadvantage of 1% or so during those hands.
On hands where the count is positive, you’ll raise your bet according to how high the count is. You might bet $200 when the count is +2, $300 when the count is +4, and so on. On these hands, your advantage might be as much as 1% to 4%, so you’ll make up for the negative expectation on the other hands while maximizing your potential winnings.
You can adjust your strategy for playing each hand according to the count, and you’ll increase your edge slightly by doing so. But it’s not necessary to make strategy adjustments to be a profitable card counter.
Between 70% and 90% of the edge you get from counting cards comes from just raising and lowering your bets at the appropriate times. If you’re truly dedicated to milking every last tenth of a percentage from your edge, then you can take the time to learn the basic strategy adjustments based on the count, but if you’re that type of person, you’ll probably also want to learn a more advanced counting system than the KO.
The Knockout Blackjack system was designed to be easy to use and effective at the same time. Anything that’s meant to be easy is going to sacrifice a certain amount of accuracy. So the Knockout System is ideal for novices and beginners, but experts who want to take it to the next level will probably want to start experimenting with some of the more involved systems like the Omega II or the Hi Opt I counts.
The Omega 2 (or Omega II) card counting system is a relatively advanced blackjack card counting system (created by Bryce Carlson) that was more popular in the 1990s than it is now. It’s most effective when used with an additional side count of aces. Like most of the more advanced card counting systems, the Omega 2 is a multi-level system. Some cards are worth 2 points, while others are worth 1.
The difficulty of the system might be well worth it, because it’s one of the most accurate and effective card counting systems available. The betting correlation is 0.92 without the optional side count, but if you can keep up with the aces, too, then you can achieve a betting correlation of 0.99. The trick is being able to manage the count.
The most detailed explanation of the Omega 2 can be found in Carlson’s 2001 book, Blackjack for Blood. This page highlights most of the details of the system, but we recommend buying a copy of the book if you’re serious about putting this card counting strategy into action for yourself.
Half of the low cards are worth +1, while the other half of the low cards are each worth +2. The 2, 3, and the 7 are worth +1. The 4, 5, and the 6 are each worth +2. 9s are worth -1 and 10s are worth -2.
These values should make intuitive sense. Obviously the 10s are more favorable to the player than the 9s, because a 10 can make a natural; a 9 can’t. And everyone knows that the 4s, 5s, and 6s are the worst cards in the deck for the casino. If you made no other change to a deck than to remove all of the 5s, you would change the game of blackjack from a negative expectation game for the player to a positive expectation game. (I learned this from Lance Humble’s The World’s Greatest Blackjack Book. He claims that Lawrence Revere would filch 5s from the deck and throw them away in some of the underground cards games in which they used to play. It’s an entertaining anecdote. I don’t know if it’s true or not, though.)
You’ll notice that this is a balanced card counting system. If you count through a single deck and make no mistakes, you’ll have a total of 0 when you reach the end of the deck.
Like other card counting systems, you want to use these values to keep a running count, which, when converted to the true count, will serve as an estimation of how favorable the deck is to the player. The higher the positive count is, the more you bet.
Unlike in some of the other, easier card counting systems, aces have a value of 0 in the running count using Omega 2. But Carlson recommends keeping a separate count of how many aces are still in play. When the deck is still rich in aces, you have a better chance of being dealt a blackjack, so you should increase your bets correspondingly. Since a blackjack (or natural) pays out at 3 to 2, it’s important that you get as much money into action as possible during those situations.
If everything on the page above seems like gibberish to you, then read through the short glossary below and give it another try. The page was written with the assumption that the reader understands the basics of card counting, but if that isn’t the case for you, then the following definitions should help you get up to speed quickly.
Balanced vs. Unbalanced – In a balanced card counting strategy, the positive and negative values balance out. This means that a running count through a single deck will result in a total of 0 if you correctly counted the cards. Most systems are balanced, but some unbalanced systems exist. The goal of most unbalanced systems is to eliminate the need to convert the running count into a true count, but they do so by sacrificing a certain amount of accuracy.
Betting Correlation – The betting correlation is a measure of how well a card counting system reflects the advantage or disadvantage to the player. Remember that all card counting systems simply ESTIMATE the player’s advantage over the house. It’s not 100% correct. The higher the betting correlation, the more accurate the system.
Levels – Card counting systems are categorized into multi-level and single-level systems. That’s just an indication of how much the cards are worth. In a single level system, the cards are worth +1 or -1, with no variations. In a multi-level system, some cards are worth more than +1 or -1. In the Omega 2, cards can be worth + or – 1 or + or – 2. Some systems have even more values. Multi-level systems are more accurate, but they’re also harder to learn and master.
Running Count vs. True Count – The running count is the total that you’ve counted. But when you’re dealing with multiple decks in a dealer’s shoe, the effect of each card is diluted by the large number of cards you’re dealing with. Most systems require that the counter convert the running count into a true count. This is done by dividing the count by the estimated number of decks still in the shoe.
Side Count – A side count is a count that you’re keeping up with separately from the running count. In most card counting systems, it’s a count of aces. Why aces are important should be obvious—you can’t be dealt a blackjack (with the 3 to 2 payout) without any aces in the deck.
The Red Seven Count is a blackjack card counting system devised and popularized by Arnold Snyder in his books Blackbelt in Blackjack and The Big Book of Blackjack. The purpose of Snyder’s system is the same as that of all blackjack counting systems—to track the number of high cards versus low cards in the deck. Since a deck that has lots of high cards in it is favorable to the player, a skilled card counter can raise his bets when that situation comes up and actually play a casino game where he has the advantage over the casino.
Snyder recommends that Red Seven Count users memorize basic strategy, and he is correct. Basic strategy is a must for all players, especially card counters, because it’s like the home base of correct strategy. In fact, some card counters rely solely on changing their bet sizes in order to take advantage of the deck’s composition. In other words, they never both learning how to change their decisions based on the count. Even if you’re willing to learn the rules for changing basic strategy decisions based on a card counting system, you have to memorize it before you can deviate from it.
Snyder claims that the Red Seven Count is 80% as effective as the Hi Lo Count, and he also says that this system is one that can be used easily and as a professional level strategy. The advantage that the Red Seven Count has over other systems is that it eliminates the need to convert the running count into a true count. The “running count” is the actual number that you’ve come up with while tracking the cards; the “true count” is that number divided by the estimated number of decks still left in the shoe. Many players have trouble with this aspect of counting, so the Red Seven Count eliminates it entirely.
In any card counting system, a heuristic strategy is used to keep up with the ratio of high cards to low cards in the deck. You don’t have to memorize every card that’s been played, and in fact, you probably couldn’t do so unless you’ve been practicing memory strategies for quite a while. All you have to do is add 1 and subtract 1.
In this particular system, aces and tens are worth -1. Every time you see an ace or a seven, you subtract 1 from your running total. Low cards (2, 3, 4, 5, and 6) are counted as +1. When your count becomes positive, you raise your bets, but when it’s low or negative, you bet the minimum while you wait to get the advantage again.
So far, this sounds just like the Hi-Lo system, but here’s the wrinkle, and it’s also the reason for this system’s name: RED 7s also count as +1. That makes this an unbalanced count, which is unusual—most systems are balanced. That means the total when you’ve counted through an entire deck will eventually land on 0. In this system, that’s not true. If you count through a single deck all the way through, you’ll end with a total of +2.
The pivot is the running count total that indicates you now have a 1% or greater advantage over the casino. It varies based on how many decks are in play, but your use of the pivot along with the unbalanced nature of the card counting system make it unnecessary to convert the running count into a true count. So instead of starting your count at 0, you’ll start your count as the number of decks multiplied by -2. So if you’re playing in a single deck game, you’ll start at -2. In a two deck game, you’ll start at -4. In an 8 deck game, you’ll start at -16.
The idea is to raise the size of your bet based on the count. In this system, if the running count is negative, you only bet one unit. If the count is 0, you bet two units. If you’re playing in a single deck game, you bet four units any time the running count is over +2. If you’re playing in a shoe game, with a lot of decks, you’ll need to raise your wagers even more to make up for the casino’s advantage. You’ll wait until you have a running count of +6 or higher to raise your bet to three units, and then you’ll move up to four units at +8, six units at +12, and 8 units at +16.
You’ll notice, if you ever try using this system, that many—in fact, most—of your bets are for one or two units. According to Snyder, these bets are called “waiting bets”, because you’re just waiting for the deck to get better in order to put more money into action. The difference between your “waiting bets” and your “highest bet” is called your betting spread. The higher your spread is, the more likely it is that you’ll attract heat from the casino.
You can use this strategy without deviating from basic strategy, but if you learn the deviations, you’ll be able to increase your advantage over the casino even further. Insurance is the first and easiest decision to make. When using basic strategy, you never take insurance, but when using this count, you take insurance any time the count is 0 or higher in a single deck game and any time the count is +2 or higher in a multiple deck game.
Normally you’d take a hit if you had a 16 versus a dealer 10, and you’d also take a hit if you had a 12 versus a dealer 3. Any time you have a running count of 0 or higher, you should stand instead.
Those are the two easiest strategy changes to memorize, and they’re also the ones that have the biggest effect on your advantage over the casino. Doubling down is another easy tactical deviation that’s easy to remember—any time the count is +2 or higher, double down any time you have a total of 10, no matter what the dealer is showing. You should also stand on a 12 versus a dealer 2 and stand on a 15 versus a dealer 10.
Snyder also recommends a few other counting systems in his books. These include the Hi-Lo Lite Count and the Zen Count. He also presents a more advanced version of his Red Seven Count in his books.
Zen is a school of Buddhism. It doesn’t really have anything to do with card counting, but it’s popular to name certain strategies and tactics “zen” in order to give them a certain Eastern mystical air about them.
Arnold Snyder introduced the Zen card counting system in his excellent book Blackbelt in Blackjack: Playing 21 as a Martial Art. It’s a more complicated system than many blackjack card counting systems, largely because it’s an unbalanced system, and also because it’s a two level system.
Card counting systems enable advantage players to estimate the proportion of high cards to low cards in a blackjack deck. This, in turn, enables them to raise and lower their bets based on this ratio. When a blackjack deck has a high ratio of high cards to low cards, the player has an advantage, and vice versa.
For example, if you had a deck of cards in which all the cards except the 10s and aces had already been dealt, your chances of receiving a “natural” (or a “blackjack”) would be significantly higher than it would if those low cards were still in the deck. Since a blackjack pays off at 3 to 2 instead of even odds, you’d want to bet more in that situation so you could win more.
Card counting systems assign values to the cards. These are usually +1, -1, or 0, but in multi-level systems, some cards might count as +2, -2, or even +3 or -3. This number represents how favorable or unfavorable the card is to the player’s odds. When this running count gets high enough, counters raise their bets to take advantage of those better odds.
Most card counting systems have an equal number of high values and low values. For example, a really simple system might value all of the 5s at +1 and all of the aces at -1. Since there are four of each card, the final count once all the cards are dealt will be 0. That’s a “balanced” card counting system.
Balanced, single-level systems are the easiest card counting systems to use, but they sacrifice a certain amount of accuracy in exchange for ease of use.
In the unbalanced Zen count, the card values are as follows:
Card counters measure the accuracy of how well a system estimates the player’s advantage or disadvantage using a number called the “betting correlation”. This is a decimal or a percentage. The closer to 100%, the more accurate the system is. The Zen Count has a high betting correlation of 96%.
The running count doesn’t take into account how many decks are in use. You divide the running count by the number of decks in the shoe in order to obtain the true count. This is the number which determines how much of an edge you have over the casino. It’s also the number that determines when and by how much you should raise your bets.
Most blackjack games, played with perfect basic strategy, offer the player a 1% disadvantage. Every +1 on the true count provides the player with an additional 0.25% edge over the casino, so once the true count reaches +4, the game is a break-even proposition. If the true count reaches +8, then the game is now offering a 1% edge for the player.
Of course, this edge is going to go up and down as you play, but your overall goal is to have an advantage (net) of between 0.5% and 1% over the casino. Most of the time you’ll be playing at a disadvantage, but you’ll be putting more money into action when you do have an edge, and that will compensate for this.
You can also make strategy adjustments based on the true count, but most (70% to 90%) of your advantage when counting comes from ranging your bets. Only dedicated experts focus on strategy adjustments on top of betting adjustments.
The guidelines in Blackbelt in Blackjack suggest that you bet as many units as your count, with a minimum bet of one unit. So if the count is negative, zero, or +1, you’ll only bet one unit per hand. If the count is +2, you’ll bet 2 units. If the count is +3, you’ll bet 3 units. And so on.
Keep in mind, though, that the casinos watch players who are ranging their bet sizes, because that’s a clue that you might be counting cards. If you want to avoid “heat” from the casino, you should find methods to camouflage the fact that you’re a card counter. You can do this by pretending to be drunk, pretending to not be paying attention, or by pretending to be a novice. One of my favorite “camouflage” tactics is to use a strategy card and still make an occasional strategy decision make, especially when my bet size is low. It’s hard to reconcile the image of a player making occasional wrong decisions while referring to a strategy card with the image of a professional card counting expert.
The unbalanced Zen count is a good system that isn’t especially difficult to learn, but it isn’t for novices, either. If you’re new to counting cards, you might be better served by learning the Hi Lo system first.
The Wong Halves Count is one of the most advanced card counting systems you’ll find. Few people actually use this strategy, as it’s pretty complex, but it’s still an interesting counting system to analyze and discuss. Many of the principles involved in card counting in general are well-illustrated by this system.
The goal when counting cards in blackjack isn’t to memorize which cards have been dealt so that you’ll know which cards are still in the deck. That kind of savant-style behavior looks great in movies like Rain Man, but in reality, counting cards is much simpler. It’s just a way of tracking the ratio of high value cards versus low value cards in the deck.
Blackjack is unlike other casino games, because it has a memory. When you spin a roulette wheel, the odds of landing on a specific number are 37 to 1 every time. The wheel doesn’t remember what happened on previous spins.
But suppose you had a roulette wheel where a specific number was removed after being landed on? Your odds of hitting another number would improve to 36 to 1.
Those odds would continue to increase until you hit your number.
There is no roulette table that works in this way, but blackjack works in a similar way. Once a specific card has been dealt, it can’t be dealt again until the deck is re-shuffled.
Since some of the cards in a blackjack deck are better for the dealer and some of the cards are better for the player, it’s a relatively easy matter to guess at which cards make a difference to a blackjack player.
Since blackjack pays off at 3 to 2 when you’re dealt a natural, the cards that can form a natural are obviously favorable to the player. The only cards which can result in a blackjack (natural) are the tens and the aces, so you want to bet more when the deck has more of those cards compared to lower cards like deuces or fives.
All counting systems track this ratio by assigning a heuristic value to each card type and then keeping an ongoing count of that total. Usually these values are easy to keep up with—you just add 1 or subtract 1 from the count, according to the rules of the card counting system that you’re using.
Of course, this only provides an ESTIMATE of your advantage.
More complicated systems use different values for different cards, usually between 1 and 3, either positive or negative.
The Wong Halves System is different because it uses fractions. The following values are used in the Wong Halves System:
If that seems complicated, it’s because it IS complicated, which is why this system isn’t commonly used by card counters today. This system does, however, provide a very accurate estimate of how good the player’s advantage (or lack of an advantage) is.
Card counters evaluate card counting systems in part based on a number called “betting correlation”. This is a percentage that rates how accurately the system estimates the player’s advantage. The Wong Halves System has a 99% betting correlation, which is one of the best in the business.
Of course, everything is relative. The Hi Lo System, for example, has a betting correlation of 97%, which makes it only slightly less accurate. It’s also far easier to implement. In the Hi Lo System, aces and tens are still worth -1, but 2s, 3s, 4s, 5s, and 6s are all worth +1 each. That’s a lot easier to remember and use.
Almost all card counting systems, including the Wong Halves Count, requires a conversion from the running count to the true count. The running count is the raw number that you come up with while you keep count of the card values in your head. The true count is that number divided by the number of decks that still haven’t been dealt.
Converting the running count to a true count compensates for the dilution effect of having so many cards in the deck. It might seem like it wouldn’t matter, but the effect of having 1 card out of 52 dealt is significantly greater than having 1 card out of 416 dealt.
Either way, the main goal of the card counter is to bet more when the count is high. This coincides with the counter’s secondary goal of betting less when the count is low. By ranging their bets in this way, users of the Wong Halves Count put more money into action when they have an edge over the casino.
Some counters also use the count to change their basic strategy decisions to more accurately reflect what’s left in the deck. This can add a tenth of a percent (or two) to the player’s expectation, but it’s not the main source of a card counter’s profits. The main source of those profits is the additional money that’s bet when the player is more likely to be dealt a “natural”.
The Wong Halves Card Counting System is an interesting curiosity, but it’s not a commonly used system. Beginners should learn something simpler, like the Hi-Lo System, and even experts will probably be better served by using a system that’s simpler but similar in terms of accuracy.
Like the Red Seven Count, the Zen Count is a card counting system created and explained by Arnold Snyder in his books Blackbelt in Blackjack and The Big Book of Blackjack. Unlike the Red Seven Count, the Zen Count is a balanced system, and it’s also a multi-level system. This page explains both of those ideas along with what a player needs to know in order to use the Zen Count to get an edge over the house in the casinos.
All card counting systems, including the Zen Count, gain an advantage for a player via two methods:
The first of those two is the most important, and it’s possible to count cards without deviating from basic strategy and still gain an advantage over the house. To get the full benefit from any system, though, memorizing a few tactical deviations in certain situations is necessary.
So the first thing that an aspiring card counter needs to learn is basic strategy. You can’t deviate from a strategy until you know it backwards and forwards. Luckily, blackjack basic strategy is relatively simple to learn. Most people can memorize all the rules via a chart in a couple of hours at most.
The reasoning behind card counting works like this—a deck of cards with a lot of tens and aces in it as compared to lower cards is more favorable toward the player. The reason for that isn’t hard to figure out. Blackjack players get paid 3 units to 2 for a natural, and you can only be dealt a natural if there are tens and aces in the deck.
You don’t have to memorize which cards have been played. You just use a point system to keep up with the ratio. When you see high cards come out of the deck, you subtract from your running total (called the “running count”), and when you see low cards come out of the deck, you add to your running total. The only difference between most card counting methods is how much you add or subtract for certain numbers.
In a balanced card counting system like the Zen Count, you’ll have an equal number of positive and negative points in a deck of cards. This ensures that if you’ve counted through the deck correctly, you’ll wind up with a total of 0 when you get to the end. The Red Seven count is an unbalanced system, but the Zen count is a balanced system.
In a single level counting system, you only add or subtract 1, based on which card you see. In a multi level counting system like the Zen Count, the amount you add or subtract depends on the rank of the card you see. The values assigned to the cards are listed below, from smallest to highest, in order of rank:
The Zen Count has some similarities to the Hi-Lo system, but it’s a little more complicated than Hi-Lo. They provide similar betting correlations, in fact, but the Zen Count provides a better estimate of changes to basic strategy, especially as it relates to whether or not to take insurance. (See https://www.qfit.com/card-counting.htm for a comparison of the various systems along with the corresponding data about betting correlation, playing efficiency, and insurance correlation.
Betting correlation (BC) is an estimate of how accurate the count is in terms of sizing your bets. The Hi-Lo System has a BC of 0.97, which is the same as that of the Zen Count. Playing efficiency (PE) is an estimate of how accurately the count adjusts for deviations from basic strategy. The Hi-Lo System has a PE of 0.51, which is significantly lower than the Zen Count, which has a PE of 0.62. The insurance correlation (IC) estimates how well the system estimates decisions regarding insurance. The Hi-Lo count has an IC of 0.76, but the Zen Count has a 0.84.
Since playing strategy changes are more important in single deck and two deck games, the Zen Count is especially effective in those types of games.
The first step in deciding how to size your bets using the Zen Count System is to convert the running count into a true count. This conversion takes into account how many decks are in play. When you have multiple decks, the effect of any one change in that deck’s composition is less pronounced than when you’re playing with a single deck.
The calculation is simple enough if you’re good at division, but you also have to be able to estimate how many decks are left in the shoe. You divide your running count by the number of decks in the shoe to determine the true count. Then you size your best based on the true count.
You wager only a single unit if the true count is 1 or less. For every point beyond that, you bet an amount equal to the true count. So if you have a count of +2, you’ll bet 2 units, and if you have a count of +3, you’ll bet 3 units, and so on.
For information about how to adjust your playing strategy, including when to take insurance using this count, see Snyder’s book, “Blackbelt in Blackjack”.