The Power Law: Why Venture Capital Math Defies Common Sense

Series: The Venture Capital Reality  |  Article 1 of 4

Prepared by Richstorm.co

Most people assume that smart investors succeed by making smart bets — spreading risk, picking quality companies, avoiding obvious mistakes. Venture capital works on a completely different logic. In VC, the goal is not to avoid failure. The goal is to find the one company that returns the entire fund on its own.

This is called the power law — and once you understand it, the seemingly irrational behavior of venture capitalists begins to make perfect sense.

The future cannot be predicted. It can only be discovered. And most attempts at discovery fail.

That quote, from a legendary venture capitalist interviewed in Sebastian Mallaby's book The Power Law, is the most honest summary of how this industry actually operates. It is also the key to understanding why VC math looks nothing like conventional investing.

What Is the Power Law?

A power law distribution is a statistical pattern where a small number of observations account for a disproportionately large share of the total outcome. It is the opposite of a normal distribution — where most observations cluster around an average — and it appears in surprisingly many places: the distribution of city sizes, earthquake magnitudes, wealth across a population, and the returns generated by startup investments.

In venture capital, the power law works like this: across a portfolio of investments, the vast majority return nothing. A smaller group return modest multiples. And a tiny fraction — sometimes a single company — generate returns so large they dwarf everything else combined.

This is not a quirk of bad fund management. It is the fundamental structure of how startup returns are distributed, confirmed across multiple large datasets.

The Empirical Distribution

Correlation Ventures analyzed over 21,000 startup financings from 2004 to 2013. The results are striking:

Source: Correlation Ventures (21,000+ financings, 2004–2013); Horsley Bridge LP data (7,000 investments, 1975–2014).

Read that last row carefully. Roughly 1.5% of investments — in a portfolio of 30 companies, that is less than one company — generate approximately 45% of the total fund's value. Everything else combined generates the remaining 55%, with the bottom 65% contributing essentially nothing.

The Math Behind Why This Works

The power law creates a counterintuitive investment strategy. To understand why, consider a simple model.

A $100M fund invests in 20 companies, deploying roughly $4M per company after fees. Suppose 65% fail completely, 25% return modest amounts, and 10% — two companies — return 30x each. Those two winners generate $240M between them. The fund returns $248M gross on $100M invested. LPs receive roughly 2.5x their money.

Now suppose the fund manager, worried about losses, passes on both of those 30x companies because they seemed too risky. The remaining portfolio — the safer bets — produces $42M. The fund returns 0.42x. LPs lose more than half their money. The cautious strategy performed catastrophically worse.

In venture capital, playing it safe is the riskiest strategy of all.

This is why experienced VCs say they only invest in companies that could return the entire fund. A company with a realistic ceiling of 5x is not interesting regardless of how good it looks — because even if it succeeds perfectly, it barely moves the needle on a portfolio that needs one 50x winner to justify its existence.

The Break-Even Reality

The minimum winner multiple required to break even depends on three variables: how many companies succeed, what those companies return, and how much the fund loses to fees before a dollar is deployed.

Assumes $100M fund, 20 companies, 2% management fee, 10-year life, 65% loss rate, 10¢ recovery on $1 losses. MOIC: Multiple on Invested Capital; It's the simplest measure of investment return — how many dollars you got back for every dollar you put in. 1.0x means breakeven

The table illustrates a brutal truth: with realistic loss rates, the required winner multiple is enormous. Anything below 20–30x from a meaningful fraction of the portfolio is unlikely to return capital to LPs after fees.

Why Most VCs Still Fail

If the power law is well understood, why do most VC funds underperform? The answer is that understanding the theory and executing on it are very different problems.

Problem 1: Fees erode before deployment

A standard 2% annual management fee on a $100M fund over 10 years extracts $20M before a single investment is made. The fund effectively deploys $80M while needing to return $100M plus profit to LPs. That 25% headstart for the GP comes entirely at LP expense.

Problem 2: Median VC underperforms public markets

Cambridge Associates and Kauffman Foundation data consistently show that the median VC fund — after fees — returns less than a simple S&P 500 index investment over the same period. The asset class looks attractive in aggregate because a handful of top funds pull the average up dramatically. Strip out Sequoia's early Google bet and Benchmark's Uber investment, and the numbers look considerably less impressive.

Problem 3: The power law applies to funds too

The same concentration dynamic that governs startup returns also governs VC fund returns. A small number of firms — Sequoia, Andreessen Horowitz, Benchmark, Accel — capture a disproportionate share of the best deals precisely because they have the brand and network to access them. Other funds are left competing for the remaining deal flow, which by definition excludes the most likely power law outcomes.

Problem 4: Missing the one company is catastrophic

Venture investor Seth Levine modeled a hypothetical $100M fund with 20 investments and found that the single company returning 10x or more produced almost $100M in proceeds on its own. If the fund had missed that one investment, it would have failed to return capital after fees. The difference between a good fund and a failed fund often comes down to a single allocation decision.

What This Means for Different Investors

For LPs investing in VC funds

If you do not have access to top-quartile funds, you are statistically better off in a diversified index. The VC asset class produces strong returns only in its upper quartile — and those funds are typically oversubscribed and closed to new LPs. Median VC is not a good trade for the illiquidity premium it demands.

For founders seeking VC

Venture capital is only the right funding source for a narrow category of businesses: those with realistic potential for 50x+ returns, meaning massive addressable markets and winner-take-most dynamics. A profitable, growing business with a $50M revenue ceiling is an excellent company but the wrong fit for VC — not because it is a failure, but because it cannot produce the return multiples the math requires.

For investors evaluating VC as an asset class

The power law explains why venture capital simultaneously drives enormous value creation for society and produces disappointing returns for most participants. The innovation is real. Google, Airbnb, Stripe, and thousands of other transformative companies were built with venture capital. But the financial returns from that innovation concentrate in a very small number of funds, founders, and early employees — while the majority of the capital invested in the industry earns below-market returns.

The Bottom Line

The power law is not a flaw in venture capital. It is the defining feature of the asset class. It explains why VCs ignore safe bets, why they concentrate in a handful of sectors and geographies, why they double down on winners, and why the majority of funds underperform despite being populated by intelligent, hardworking people.

Understanding the power law is the prerequisite for understanding everything else about venture capital: how GPs and LPs relate to each other, why rejection rates for founders are so brutal, and why the industry's own narrative is built almost entirely on survivors.

Those topics are the subject of Articles 2, 3, and 4 in this series.

Series: The Venture Capital Reality

Article 1 — The Power Law: Why VC Math Defies Common Sense

Article 2 — The LP/GP Relationship: Who Really Bears the Risk  (coming next)

Article 3 — The Rejection Funnel: What Getting VC Funded Actually Looks Like  (coming)

Article 4 — The Survivor’s History: What The Power Law Book Gets Wrong  (coming)

RichStorm LLC is a science-first investment analysis publication. All content is for informational purposes only and does not constitute investment advice. See richstorm.co/disclaimer for full disclosures.