Zero to One: Notes on Startups, or How to Build the Future · chapter 7 · member edition

Follow the Money

So what

Chapter 7 is the book’s strongest from an evidence standpoint and the one where Thiel is most qualified to speak from direct experience. The power law in VC returns is one of the most empirically validated phenomena in startup finance, confirmed by Correlation Ventures (21,640 financings showing 65% loss rate and 4% returning 10x+), Horsley Bridge Partners (decades of fund-of-funds data), AngelList (empirical power-law with alpha=2.42), and the Kauffman Foundation (most VC funds underperform public markets). Y Combinator’s portfolio confirms the pattern: fewer than 20 of 4,000+ companies represent over 75% of total value. Thiel’s specific examples (Founders Fund’s Facebook return, Benchmark’s eBay/Uber) all check out. The critique of Modern Portfolio Theory for VC is technically sound – MPT assumes normal distributions with finite variance, which AngelList data proves false for startup returns. The only BS is in overextension: Thiel implies the power law should govern all life decisions (career, relationships), not just VC portfolios. And his anti-diversification prescription is contradicted by IMD data showing diversified VC funds outperform concentrated ones. The power law describes returns accurately but does not prescribe concentration – you need MORE bets, not fewer, to catch the outliers.

Verdict

MostlyAccurate

Claims checked in this chapter (9)

holds⚠ re-audit pending
Venture capital returns follow a power law distribution where a small number of investments generate nearly all returns.
needs context⚠ re-audit pending
The best investment in a successful fund should equal or exceed the rest of the fund combined.
holds⚠ re-audit pending
Y Combinator's portfolio demonstrates power law dynamics where a tiny fraction of companies account for the vast majority of total value.
holds⚠ re-audit pending
The power law is unintuitive because humans naturally think in normal distributions. Cognitive biases make people systematically underestimate the likelihood and magnitude of extreme outcomes.
contested⚠ re-audit pending
Diversification is wrong for venture capital because it contradicts power law logic. VCs should concentrate bets on potential winners.
needs context⚠ re-audit pending
Career decisions follow a power law -- choosing the right company or field matters exponentially more than anything else you do. 'Life is not a portfolio.'
holds⚠ re-audit pending
Most businesses never interact with venture capital -- VC finances only about 1% of new firms but VC-backed companies represent 77% of US public market capitalization.
holds⚠ re-audit pending
A single VC investment must have the potential to return the entire fund to be worth making. This is the fund-returning threshold that should guide all investment decisions.
holds⚠ re-audit pending
Modern Portfolio Theory (Markowitz) is wrong for startups because MPT assumes normally distributed returns with finite variance, but VC returns follow a power law with unbounded mean.