Korva is a scroll of the internet’s best ideas.
One deal beats the other 99
The power law math
The power law is scale invariant, so the 80-20 rule can be applied again and again. If 20 percent of deals produce 80 percent of the profits, the top 4 percent of deals produce 64 percent of them, and the top 0.8 percent produce 51.2 percent.
Because 0.8 percent is less than one deal in 100, a firm that makes 100 investments should find its single best one outearning the other 99 together. Altos Ventures has now made around 200 investments, and it expects that pattern to repeat in each hundred.
Spotting the winner
The winner should become obvious over time. It starts beating plan, rises above the noise, doesn't keep running out of cash, and doesn't keep coming back for extra money because it misplanned again. A company that great smacks you in the face as blindingly, obviously outstanding, and an investor who can't recognize it after years of watching has no place in the business.
Sutter Hill used to ask its general partners and its limited partners every year to guess which deal in the portfolio would be the best. Over multiple decades, the general partners were no better than the limited partners at picking the big winners.
Why does the 80-20 rule keep repeating?
Economist Vilfredo Pareto observed in the 1890s that about 80 percent of Italy's land was owned by about 20 percent of its people. Distributions that follow a power law look the same at every scale, so the top slice of the top slice is lopsided again. The same pattern shows up in city sizes and in how often words appear in a language. Linguist George Kingsley Zipf documented the word pattern in the 1930s.
How lopsided are venture returns in practice?
Venture returns are often described as a power law business, where a few outcomes carry nearly everything. In his book Zero to One, Peter Thiel wrote that the best investment in a successful venture fund equals or outperforms the entire rest of the fund combined. That is why venture firms judge themselves by their biggest winners and not by their average deal. A fund can lose money on most of its investments and still do well.
Why do experts struggle to pick winners early?
Early-stage companies face enormous uncertainty, and the qualities that make one succeed often look unremarkable at the start. Paul Graham wrote in his essay Black Swan Farming that the startups that turned out best at Y Combinator often seemed like bad ideas at first, Airbnb among them. Investors who reject anything that looks odd end up rejecting the outliers along with the duds. Graham counted himself among those who got it wrong.
More from Joys of Compounding
