A Scrolly Explainer

Why the Most Talented People Aren't the Most Successful

A simulation of 1,000 careers over 40 years reveals that luck — not talent — is the hidden engine of extreme wealth.

1,000
Simulated careers
40
Years simulated
1 finding
That changes everything
Talent Distribution Average Low High

Most people cluster around the mean

?
Wealth Distribution Top 20% own 80% Most people Richest

A tiny few hold almost everything

In 2017, eight men held as much wealth as the poorest 3.6 billion people on Earth. If talent is normally distributed but wealth is wildly skewed, something else must be at work.

The Meritocracy Assumption

The standard explanation is meritocracy: people who work harder, think smarter, and take bigger risks earn more. Over time, small differences in talent compound into large differences in outcomes.

It's a comforting story. It justifies inequality as earned. It shapes how we fund research, hire employees, promote leaders, and award grants.

But here's the problem: nobody is a billion times more talented than anyone else. Nobody works a billion times more hours. Yet some people have billions of times more wealth.

Average IQ is 100, but nobody has an IQ of 1,000 or 10,000. The same holds for effort. Yet when it comes to rewards, some people do have billions of times more than others.

The Italian physicists Alessandro Pluchino, Alessio Biondo, and Andrea Rapisarda at the University of Catania decided to build a simulation to test what's really going on.

Building the Simulation

The "Talent vs. Luck" model is elegantly simple. Three ingredients:

1,000 People with Normally Distributed Talent

Place 1,000 agents on a grid. Each gets a talent score drawn from a bell curveA Gaussian/normal distribution with mean 0.6 and standard deviation 0.1. About 68% of agents fall between 0.5 and 0.7 talent. (mean = 0.6, standard deviation = 0.1).

Some are more talented, some less — but nobody is dramatically different. Everyone starts with 10 units of capital.

Less talented More talented

Random Lucky and Unlucky Events

Scatter 500 event-points across the grid — half lucky (green), half unlucky (red). These events move randomly, like opportunities and setbacks drifting through a person's life.

Lucky event Unlucky event

40 Years of Career Simulation

Run the clock for 40 years (80 six-month intervals). The rules:

  • Lucky event hits you: Double your capital — but only with probability proportional to your talent.
  • Unlucky event hits you: Lose half your capital, regardless of talent.

Talent helps you exploit good luck. It can't protect you from bad luck.

Lucky event Capital × 2 Probability = talent score (0–1) Unlucky event Capital ÷ 2 Probability = 100% (no protection)

The Results: A Power Law Emerges

After 40 simulated years, despite everyone starting with equal capital and talent following a bell curve, the final wealth distribution is a power lawA mathematical relationship where a small number of observations have extreme values, creating a "long tail." Unlike a bell curve, power laws have no meaningful average..

0
of capital held by top 20%
0
out of 1,000 with 500+ units
0
ended below starting point
Final Wealth Distribution (log-log scale) Number of agents Capital / Success 1 10 100 500+ 0.01 0.1 1 10 100 1000+ Power law slope ≈ −1.3 Top 20 hold 44% of all capital

The simulation reproduces real-world wealth inequality from nothing more than talent + randomness.

The Bombshell: Success ≠ Talent

The most successful agent — maximum wealth of 2,560 units — had a talent score of just 0.61. Barely above average.

Meanwhile, the most talented agent (talent = 0.89) ended with less than 1 unit of capital — a 90% decline from where they started.

Talent vs. Final Capital Talent score Final capital (log) 0.3 0.4 0.5 0.6 0.7 0.8 0.9 0.01 0.1 10 100 2500+ mean Most successful: T = 0.61 Capital: 2,560 (barely above average talent) Most talented: T = 0.89 Capital: <1 unit (90% decline)
The maximum success never coincides with the maximum talent, and vice versa. — Pluchino, Biondo & Rapisarda (2018)

The Real Predictor: Luck

If talent doesn't predict success, what does? The answer is unambiguous: the number of lucky events experienced. The correlation between success and luck is overwhelming — far stronger than between success and talent.

Most Successful Agent

Talent: 0.61 | Final: 2,560

Year 0 Year 20 Year 40 Burst of luck, ages 35–40

Least Successful Agent

Talent: 0.74 | Final: 0.0006

Year 0 Year 20 Year 40 Run of bad luck erases all

Same simulation. One agent's talent was average but caught a wave of lucky breaks. The other had well-above-average talent but was crushed by a run of bad luck. The lucky agent ended with 4 million times more wealth.

So What Should We Do About It?

The researchers tested four strategies for distributing resources — modeling research funding, but applicable to any investment context.

Funding Strategy Efficiency (normalized)

0 0.25 0.50 0.75 1.0 Normalized efficiency index Equal (1u each) 1.00 ★ BEST Random selection 0.70 Mixed (premium + baseline) 0.50 Fund top 25% 0.30 Fund top 10% 0.15 ← WORST

The winner, by a wide margin: equal distribution. Giving 1 unit every 5 years to all 1,000 agents doubled the percentage of talented people who achieved success (from 32% to 69%), at the lowest total cost.

The "meritocratic" approach — funding past winners — ranked dead last. Randomly distributing funds outperformed backing winners by nearly 5×.

The Micro vs. Macro Paradox

The researchers surface a crucial distinction:

Micro View: Individual Level

👤 Talent = hit rate

A more talented person has a higher probability of exploiting any given opportunity. Talent sets the hit rate.

Macro View: Population Level

Many average-talent agents Moderate talent Above average Top performers Mostly average talent!

Moderately talented people dominate the top — simply because there are far more of them. Luck sets the at-bats.

Talent sets the hit rate. Luck sets the at-bats.

The aggregate result: success looks like it rewards talent, but it mostly rewards exposure to luck.

What This Means For You

This isn't an argument against working hard or developing skills. Talent is necessary — the simulation confirms that. But it's not sufficient, and it's rarely the deciding factor at the extremes of success.

For Individuals

Maximize surface area for luck. Diversify your bets, meet more people, try more things. You can't control lucky events, but you can increase your exposure to them.

For Managers & Funders

Stop over-indexing on past performance. It's contaminated by luck. Spread smaller bets across more people rather than concentrating resources on proven "winners."

For Policymakers

Universal baseline support (education, healthcare, seed funding) is more efficient than elite rewards at maximizing societal talent utilization.

For Self-Awareness

If you're successful, be honest about the role of luck. If you're struggling despite talent and effort, the simulation says: that's statistically normal, not a personal failure.

Talent opens doors.
Luck decides which ones you walk through.

Based on "Talent vs Luck: The Role of Randomness in Success and Failure" by Pluchino, Biondo & Rapisarda (2018). The research won the Ig Nobel Prize in Economics — an award for research that "first makes you laugh, then makes you think."

Luck dominates at extremes Equal funding beats elite funding Surface area for luck matters

Read the original paper (arXiv)

Made with scrolly.toReport