Why the World Isn’t Average: A Guide to Power Laws

From a young age, we are taught to think in terms of averages. We learn about the average grade, the average height, and the average salary. We are introduced to the gentle, symmetrical slope of the bell curve, a model that describes so many phenomena in the natural world. It’s a comforting, predictable framework where most results cluster around the middle, and extreme outliers are rare. There's just one problem: many of the most important systems that shape our modern world—from our economy to the internet—do not follow this tidy pattern. They are wild, unequal, and defined by extremes. These systems operate under a different and far more dramatic logic: the power law. Understanding this concept is not just an academic exercise; it's an essential toolkit for making sense of a world that is anything but average.

The Familiar Comfort of the Bell Curve

Before we venture into the wild territory of power laws, let's revisit the familiar ground of the normal distribution, or the "bell curve." Imagine we measure the height of 10,000 adult men in the United States. When we plot the results on a graph, we will almost certainly get a beautiful, symmetrical bell shape.

A few individuals will be exceptionally tall, and a few will be exceptionally short. These are the outliers at the "tails" of the curve. However, the vast majority will cluster around the middle. The most common height (the mode), the middle value if you lined everyone up (the median), and the mathematical average (the mean) will all be nearly identical. In this world, the average is a genuinely useful and representative number. If you know the average height is around 5 feet 9 inches, you have a very good sense of what a typical person looks like.

This "Mediocristan," as statistician Nassim Nicholas Taleb calls it, is the realm of the predictable. An outlier, like a person who is 7 feet 6 inches tall, is rare but not world-changing. They are still just a few standard deviations from the mean. No single observation can dramatically skew the entire picture. This is the logic that governs many physical traits, test scores, and measurement errors. It's a world where things tend to even out.

Welcome to Extremistan: The Power Law

Now, let's leave the world of human height and enter a different domain: personal wealth. If we plot the net worth of 10,000 Americans, the graph will look nothing like a bell curve. Instead, we'll see a graph with a huge spike at the low end and a very, very long tail stretching out to the right.

The vast majority of people will have a modest net worth, clustered together on the left side of the graph. Then, as we move right, the numbers drop off steeply. But the tail doesn't just end; it keeps going. Out on this long tail, we find a small number of individuals whose wealth is not just a little more, but orders of magnitude greater than everyone else. The presence of a single billionaire in this sample would completely warp the "average."

This is a power-law distribution. It's a relationship where a change in one quantity triggers a proportional change in another. They are often called "heavy-tailed" or "long-tailed" distributions because the tail, which represents rare and extreme events, is much "heavier" and longer than in a normal distribution. It means that extreme outcomes are far more likely and have a much greater impact than a bell curve would ever predict.

This pattern is not an anomaly; it is everywhere:

• City Populations: A few megacities (New York, Los Angeles) house millions, followed by many smaller cities and a vast number of small towns.
• Website Traffic: A handful of sites like Google, YouTube, and Amazon receive a colossal share of all internet traffic, while millions of smaller blogs and websites get a trickle.
• Book Sales: A few bestsellers dominate the market, selling millions of copies, while the vast majority of books sell only a few hundred or thousand.
• Venture Capital Returns: A single successful investment in a portfolio (the next Google or Amazon) can generate more returns than all the other failed or mediocre investments combined.
• Social Media: A tiny fraction of videos go viral and get billions of views, while the average video gets a handful.

In this world of "Extremistan," the outlier is not an exception; the outlier is everything.

Why 'Average' Is a Dangerous Word in a Power-Law World

The most critical mistake we can make when dealing with a power-law system is to use the average (the mean) as a summary statistic. The mean is extremely sensitive to outliers.

Let's use a simple thought experiment. Ten people are in a coffee shop, each with a net worth of $50,000. The mean net worth is $50,000. The median net worth is also $50,000. The average here is a perfectly reasonable description of the group.

Now, a tech CEO with a net worth of $1 billion walks in. There are now eleven people. The total net worth is $1,000,500,000. The new mean net worth is over $90 million per person. Suddenly, on average, the coffee shop is filled with multimillionaires. This statistic, while mathematically correct, is profoundly misleading. It tells us absolutely nothing useful about the financial situation of the original ten people.

The median, however, tells a very different story. If we line up all eleven people by net worth, the person in the middle (the sixth person) still has a net worth of $50,000. The median remains unchanged and provides a far more robust and truthful picture of the typical person in the room.

In a power-law domain, the mean is pulled dramatically by the extremes in the long tail. When you hear the "average" salary in a profession that includes a few superstars, the "average" return on a venture fund, or the "average" number of followers for a social media user, you should be highly skeptical. The average is a lie that hides the winner-take-all reality of the system.

A Practical Toolkit for Navigating Extremes

Recognizing that we live in a world governed by both bell curves and power laws is the first step. The next is to equip ourselves with the right mental tools to navigate it.

Embrace the Median, Question the Mean

This is the most direct takeaway. Whenever you encounter data that you suspect might follow a power law—like income, wealth, company sizes, or creative success—immediately look for the median, not the mean. The median answers the question: "What is the experience of the typical person or entity?" The mean, distorted by a few massive outliers, often answers the question: "If the total pie were divided equally, what would everyone get?"—a scenario that never happens in a power-law system.

Think in Orders of Magnitude (Use Log Scales)

Our brains are not wired to intuitively grasp the vast differences in a power-law distribution. The difference between a net worth of $100,000 and $1 million is huge, but the difference between $1 million and $1 billion is of a completely different nature.

One powerful tool for visualizing and understanding these systems is a logarithmic scale. On a standard linear scale, you might plot 1, 2, 3, 4. On a log scale, you plot powers of 10: 1, 10, 100, 1000. This compresses the extreme values, making the underlying pattern clearer. When you plot a power-law distribution on a log-log graph, it magically transforms into a straight line. This helps us see the relationship more clearly and appreciate that a move from 10 to 100 is, in a systemic sense, just as significant as a move from 100,000 to 1,000,000. It trains our minds to think in terms of multiples (10x, 100x) rather than simple addition.

The 80/20 Rule Is Everywhere

You have likely heard of the Pareto Principle, or the 80/20 rule. It states that, for many events, roughly 80% of the effects come from 20% of the causes. For example:

• 80% of a company's revenue comes from 20% of its customers.
• 80% of software crashes are caused by 20% of the bugs.
• 80% of your results come from 20% of your efforts.

The 80/20 numbers are not a magical constant, but the underlying principle is a folksy expression of a power-law dynamic. It highlights a severe imbalance between inputs and outputs. In any system where you suspect a power law is at play, ask yourself: "Where is the critical 20%?" Identifying and focusing on that small segment of causes can lead to disproportionately large results. It's a strategy for applying your effort where it will have the most impact, whether in business, productivity, or learning.

Acknowledge Tail Risk and Black Swans

In a bell curve world, extreme events are so rare as to be practically impossible. The probability of an American man being 15 feet tall is effectively zero. But in a power-law world, the "impossible" happens. These are the "black swan" events: high-impact, hard-to-predict, and rare events that are beyond the realm of normal expectations.

The 2008 financial crisis, the explosive growth of the internet, or a single viral video changing a person's life overnight are all examples of tail events. In a power-law distribution, the tail is fat, meaning these extreme outcomes, while individually rare, are collectively significant. You cannot ignore them. This means that in finance, you must account for market crashes that seem impossible. In business, you must recognize that one competitor could emerge and change the entire industry. In a creative career, you must understand that success is often not a steady climb but a series of small bets with the potential for one massive, breakout success.

Living in a World of Extremes

We are conditioned to seek the safety of the average. But in more and more areas of our lives, the average is an illusion. The forces of technology, globalization, and network effects have supercharged power-law dynamics, creating winner-take-all markets and a world of extreme outcomes.

By abandoning a one-size-fits-all reliance on bell-curve thinking, we can see the world more clearly. We can learn to value the median over the mean, to think in multiples instead of increments, to focus our efforts on the vital few, and to prepare for the radical impact of the outlier. Understanding the difference between Mediocristan and Extremistan is more than a statistical curiosity—it is a fundamental upgrade to our mental software, allowing us to make better judgments and more strategic decisions in the unequal and unpredictable world we truly inhabit.