1 Introduction

A long time ago, I watched the movie "21," and the scene where the team worked together to count cards during blackjack left a deep impression on me. However, at the time, I just watched it for fun and didn't delve into the deeper reasons why card counting can win.

I recently had some free time, so I reread it. This time, for some reason, I was drawn to the mathematical beauty hidden within it—those seemingly intuitive decisions actually obeyed the laws of probability.

I suddenly realized that this movie isn't just about a battle of wits in a casino, but about the struggle between reason and intuition. The moments where you think you should trust your intuition only to have it easily overturned by numbers are particularly fascinating.

To avoid spoilers, I won't go into detail about the plot. I'll just mention a few points relevant to this article: the team's core members rely not on memory but on a keen sense of probability; nearly every decision they make is based on exploiting information asymmetry. What viewers often overlook is that card counting is never about predicting the future; it's about using past information to calibrate future probabilities.

To understand the principles behind card counting, it is helpful to start withThe famous question asked by the professor in the filmSpeaking of——Monty Hall Problem(Also known as the "three doors problem"):

Its form is much simpler than card counting, but the logic behind it is surprisingly similar:We are all using new information to correct our judgment of unknown results.

This question seems to be just a light-hearted puzzle game, but it has caused countless mathematicians and audiences to engage in heated debate because it reveals a disturbing fact:Our intuition almost instinctively rejects the truth of probability.

2 Analysis of the Monty Hall Problem

2.1 What is the Monty Hall Problem?

The Monty Hall problem originated from the American television game show "Let's Make a Deal" in the last century. The host asked the contestants to choose one of three doors: one door contained a car (the grand prize); the other two doors contained goats (no prizes).

The contestant first chooses a door, and then the host (who knows what is behind each door) will choose one of the remaining two doors.Open a door with a goatAt this point, there are only two doors left: the one you originally chose, and the one that has not been opened.

The host will then ask you: "Do you want to stick with your original choice, or switch to another door?" This is the Monty Hall question:

The question may seem simple, but almost everyone who hears it for the first time willIntuitively, I think it doesn't matter whether I change it or not.——Anyway, there are two doors, and the probability of winning is 50% each, right?

However,The math answers the exact opposite.：If you choose to change, the probability of winning is 2/3If you stick to your original choice, the probability of winning is only 1/3.

It is this counter-intuitive result that has sparked countless discussions and debates over the past few decades: from TV viewers to mathematics professors, from psychologists to probabilists, everyone has tried to explain why our intuition is so stubbornly "wrong" on this issue.

To understand “why it is easier to win the lottery by changing the door”, we can look at this issue from two different perspectives, from complex to simple.

2.2 Why is the probability of winning higher after changing the door?

2.2.1 Standard mathematical solution: using conditional probability calculation

Let’s first use the most “formal” mathematical method to solve this problem, which is the “variable substitution” thinking mentioned by the professor in the movie.

Suppose there are three doors, A, B, and C. Only one of them has a car behind it. You choose A. The host knows the answer and opens another door (say, C), which must be a sheep. Now, he gives you a chance - do you choose B instead?

To determine whether it is cost-effective to change the door, you are actually comparing the probabilities of two situations:Case 1: Stick to the original door (A) → What is the probability of winning the car?Case 2: Change door (B) → What is the probability of winning the car?

Let's look at situation 1: You choose A at first. The probability that there is a car behind A is only 1/3This will not change, no matter what the host does later, your first choice has already determined the probability.

Let's look at the second case: If there is no car behind door A, then the car must be behind one of the two doors B or C (the total probability of the car being behind the two doors is 1/3+1/3=2/3). The door opened by the host must be the sheep, so the probability of winning the car is "concentrated" on the remaining unopened door. Therefore, the probability of winning the car by changing the door = 2/3.

This logic is actually the "conservation of probability" - the initial 1/3 of "good luck" remains on the original door, and the 2/3 of "bad luck" is concentrated on the other door after the host helps you filter out the wrong options. So,Changing the door is equivalent to betting all the 2/3 probability on the remaining unopened door, so the chance of winning is naturally higher.

To be more intuitive, you can also list all the situations:

First selection	Car location	The host opened the door	Door change results	Win or not
A	A	C	Change → B	lose
A	B	C	Change → B	win
A	C	B	Change → C	win

You'll find that in three equally probable situations, switching the door will win twice (2/3), while not switching will only win once (1/3). If you understand this logic from this point on, congratulations—you've mastered one of the most important ideas in probability theory:Probabilities can be revised as new information becomes available.

2.2.2 The simplest heuristic: choosing one door is now equivalent to choosing two doors

In fact, this problem can be understood more simply -“"Changing doors" is actually equivalent to the host giving you a new choice: choose one door or two doors.

Think about it: If you don't change the door, it's like sticking to the original one. The probability of winning is always 1/3But if you choose to change the door, since the host has helped you eliminate the door with the sheep in it, the remaining door will inherit the "total" winning probability of the other two doors.

In other words, although on the surface you just changed from one door to another, in reality you expanded your choice from "one door" to "two doors". 1/3 Change the door: You have chosen "the whole of two doors", and the probability of winning the car becomes 2/3.

From this perspective, the so-called "changing doors" does not mean that luck has improved, butYou used the new information to redistribute the original 2/3 of the error probability to your side.This is the simplest and most essential way of thinking about the Monty Hall problem:It's not probability magic, but information redistribution.

Note: Many people like to use Bayes' formula to prove this problem. In fact, it is correct, but the key is to understand how information changes probability. The formula only helps you confirm the answer.

3 Analysis of the card counting logic in the movie "21"

3.1 Overview of Card Counting in Movies

In 21, the card counting team's strategy is a textbook example of probability application. On the surface, they are just "better at counting" than others, but the real core lies in -They are dynamically using information to update their judgments.

At a casino blackjack table, the dealer will mix and match several decks of cards. As the cards are revealed, the composition of the remaining cards gradually becomes unbalanced: a high number of low cards dealt initially means a higher proportion of high cards remaining, and vice versa. Card counting is the process of tracking this imbalance and adjusting betting strategies accordingly.

They used a method called “Hi-Lo Count” In simple terms,By observing the cards that have appeared, estimate the ratio of high cards to low cards in the remaining deckThe higher the count, the more big cards are left, and the player's chances of winning increase accordingly.

Through this information update, they can increase their bets during favorable periods and withdraw them during unfavorable periods, ensuring that funds are invested only in the most reliable opportunities. In other words, what they do is not prediction, butDynamically revise your judgment of future outcomes based on new information you continually obtain——This is exactly the same as the core idea of the Monty Hall problem: information update → probability redistribution.

3.2 The Mathematical Logic of Card Counting: Dynamic Correction of Probability

The core of card counting is not "remembering what each card is", butAfter each card is turned over, use this new information to correct the probability judgment of the remaining handsThis is consistent with the essence of what we learned in Chapter 2 using the Monty Hall problem: new information emerges → we should reallocate probabilities for future events → adjust our strategy based on the new probabilities.

Let’s break this process down into several more specific steps – explaining both “what it is” and “how to use it.”

1) Counting: Compressing complex deck information into a single “signal”

When one or more decks of cards are mixed together in a casino, the distribution of the cards in the deck is initially known (for example, how many 10s there are, how many low cards there are). As cards are dealt, the composition of the remaining deck changes.

Card counting is to "compress" the information of the cards that have appeared into a simple value (calledrunning count), a common practice is high-low counting (Hi-Lo):

Cards	Count value	illustrate
2 ~ 6	+1	The more small cards appear, the more favorable the remaining deck is to the player.
7 ~ 9	0	Neutral, no scoring
10. J, Q, K, A	−1	The more big cards appear, the more disadvantageous the remaining deck is to the player.

By adding up the dealt cards one by one, a running count (which may be positive or negative) is obtained. This number itself is a compressed description of the "bias of the remaining deck": the higher the count, the higher the proportion of high cards remaining, and the better the player's chances of winning.

2) What is really useful is not the running count, but the "true count"“

The running count is affected by how many decks are currently left - the same running count means different things with one deck than with six. So divide the running count byEstimated number of decks remaining,get True Count:

For example, suppose there are 6 decks (312 cards) and the current running count is +6, but there are approximately 3 decks left on the table. Then the true count is ≈ +6 / 3 = +2. This +2 means that at the level of each deck, the remaining deck has a "+2 bias signal" for the player.

3) Mapping true count to “player advantage” and then to bet size

Empirically (there are slight differences in different literatures),Each +1 in the True Count roughly corresponds to a small percentage increase in the player's expected value.The common experience value is: for every 1 point of true count, the player's expected value rises by about 0.5%(This value is an empirical approximation and actually depends on the rules and specific implementation).

Continuing with the above example, if the true count = +2, the player's expected value against the dealer could increase by approximately 1% (2 × 0.5%). The "real players" on the card counting team would then increase their bets accordingly: betting small when the advantage is zero or negative, and betting large when the advantage is positive. In the long run, this selective betting strategy will generate positive returns.

Note: Card counting does not allow you to win every hand, but allows you to increase your bets in situations where you are sure and reduce them in situations where you are not sure, thereby turning the long-term compound expected value into a positive number.

4) Risk Management (Why Winning Isn’t a Guarantee)

Even if a true count of +2 means an expected value increase of 1%, this is only the expected value (long-term average), and individual games still have high variance. Therefore, the team must also consider:

• Bet spread: What chip multiplier should be used to switch between high and low true counts?
• Volatility and bankruptcy risk: A positive expectation does not mean that you will not go bankrupt in the short term. Appropriate money management (such as the inspiration of Kelly's principle) is needed to determine the betting ratio.
• The risks of casino intervention: Large bets, frequent entries and exits, or suspicious behavior will attract the attention of the casino (we will discuss this in detail in the reality section later).

5) The relationship between card counting and Monty Hall

Putting the above steps together you will find: Monty Hall Problem:

• The host opens the door and provides new information → we redistribute the probabilities accordingly → finally make a decision to “switch the door” or “not switch the door”;
• Card Counting (Blackjack): Every time a card is turned over, a piece of information is revealed → Running Count + True Count compresses the amount of information into an actionable signal → We adjust our bets (decisions) based on the True Count.

The core idea shared by both isBayesian information updating: It is not about intuitively "feeling it should be", but about systematically revising the probability when new information arrives, so that the long-term expectations shift in favor.

3.3 Card Counting Team Strategy and Psychological Warfare

In 21, card counting isn't a solo endeavor, but a carefully planned team effort. Through clear division of labor and tacit cooperation, team members transform mathematical probability into practical casino operations.

Teams are typically divided into two core roles:

• SpotterThey lurk at various gaming tables, observing the action with low stakes. While seemingly playing casually, they are actually mentally calculating the points and tracking the ratio of high to low cards in the remaining deck.
• Big PlayerWhen a wind scout determines that a table has a clear advantage, they will use a secret signal to summon real players. These real players appear to be wealthy gamblers passing by, but in reality they are placing bets on tables with calculated odds, often investing large sums of chips.

Cooperation process:

1. Detecting Advantage Tables: The Wind Seeker places low stakes bets at each table while tracking the Hi-Lo count in real time, waiting for the deck to turn in the player's favor.

2. Sending signals: Once the count shows that the proportion of remaining high cards is high, the wind detector uses a predetermined code to notify the real players.

3. Big bets: Real players come to the table to place bets and take full advantage of the game.

4. Withdraw: When the count falls or the advantage disappears, the wind explorer sends an "evacuation" signal, and the real players leave the market in time to ensure that funds are only invested at the most reliable stage.

Psychological warfare and casino games:

The team relies not only on mathematical judgment but also on psychological strategies. Casinos closely monitor player behavior: unusual bets, emotional reactions, or the rhythm of raising bets can all raise eyebrows.

Card counting teams must conceal their advantages: occasionally deliberately losing money, pretending to be emotional, or changing the rhythm of their betting to maintain the appearance of ordinary players. At the same time, they also need to observe the reactions of dealers, other players and casino security, and adjust their behavior strategies at any time.

The success of the card counting team depends not only on mathematical calculations, but also onUnderstanding and controlling human natureThey use probability to determine the bet, but use psychological strategies to control the risk, making the whole gambling aThe dual game between rationality and humanity, mathematics and psychology.

4 Off-topic: How do real casinos counter probability players?

The card counting team in the movie seems to be invincible, but the real casino environment is much more complicated than the movie.Keep an eye on suspicious players, observe their betting patterns and behaviors, which is actually a common countermeasure in real casinos.

To summarize the casino’s basic response measures:

1. Frequent shuffling and mixing of multiple decks

Real-life casinos often deal multiple decks of cards and shuffle them quickly. This quickly disrupts the remaining decks, rendering Hi-Lo calculations ineffective. Even the most proficient in probability calculations cannot accurately predict the next hand's advantage.

2. Betting limits and mid-term regulation

Casinos not only limit the maximum bet per table but also strictly control when players can join. If someone joins mid-game and attempts to place a large bet, the casino becomes highly alert and may even ban them from continuing. This directly undermines the movie's strategy of "real players joining the table at any time to place large bets," limiting the maneuvering space for card counting teams.

3. Dedicated monitoring and technical means

The movie-like practice of "catching card counters" is also a reality: casinos assign experienced dealers, supervisors, and security guards to observe players' betting patterns, movements, and even facial expressions. Modern casinos are even equipped with cameras and behavioral analysis systems to analyze player behavior in real time and take immediate action if any anomalies are detected.

4. Psychological and strategic interference

To disrupt the rhythm of probabilistic players, casinos may manipulate the rhythm of card dealing, dealer changes, and table rotation. Frequent table changes or shuffling make it difficult for players to establish a sustained advantage and put psychological pressure on them.

The card counting team in the movie is able to win under the idealized assumption that the casino uses fixed decks, shuffles infrequently, and that teamwork and signal transmission are perfectly executed. However, based on the above casino countermeasures, these conditions are rarely fully met in reality:

• Multiple decks and frequent shuffling reduce the accuracy of Hi-Lo counting and shorten the advantage window.
• Dedicated monitoring, betting limits and psychological interference make team collaboration difficult to implement covertly.
• High-stakes betting can easily attract attention, and the benefits are not proportional to the risks.
• The rule prohibiting joining midway directly undermines the operating mode of "real players can come to the table and place bets at any time".

in other words,Real casinos minimize the probability advantage, making it difficult for any attempt to exploit probability to generate long-term stable profits..

5. Afterword

The movie "21" is so memorable not only because the students won the casino, but also because it reveals a more universal reality:Human intuition is not always suitable for dealing with probability problems.

In the film, the protagonists, through training, free themselves from relying on intuition and instead rely on mathematical models to place bets—this is the key to their victory over the casinos. But at the same time, their success also seems extremely "inhuman": there is no passionate impulse, no intuitive gamble; everything is based on rational calculation.

In real life, it's almost impossible for us to act like them. Even if we know that card counting is a reasonable probabilistic strategy, we can still be swayed by emotions, luck, or short-term gains and losses. We want to "win," but often resent the rational rules that make winning boring.

This is the true paradox of probabilistic thinking:To master it means giving up a certain kind of "humanity".

In other words, mathematics teaches us to remain calm, while human nature makes us volatile. Casinos, the stock market, and many decisions in life are actually like a game of blackjack—you can't control the cards you're dealt, but you can choose how to bet. Therefore, card counting is just a superficial phenomenon.

The deeper revelation is:Can we maintain rationality and update our judgment in the face of uncertainty?Can we, like the card-counting team, constantly adjust our strategies based on new information, rather than dwelling on the ups and downs of our emotions?

Perhaps, "21" is not about how to win the casino, but rather——How to coexist with probability in an uncertain world.