---
This article was aided by OpenAI Sol 5.6 and the SignalRank Agent (https://agent.signalrank.com)
date: "2026-07-10"
---
Ilya Strebulaev has done something useful for a market that often runs on anecdote. His new ranking of angel investors brings structure to a part of venture that is usually discussed through reputation, memory, and selective war stories.
That matters. There are still too few academics looking seriously at early-stage venture or indeed venture at all. The asset class needs more people willing to collect data, expose assumptions, and make rankings inspectable.
But a ranking is only as good as it is actionable. Can I make investment decisions based on it? That is important for an LP and a co-investor. And for founders, how good is my investor?
Strebulaev is clear about the limitation of his own leaderboard:
“The ranking uses lifetime US unicorn investments.”
and
“Every lifetime leaderboard has the same blind spot. The unicorns behind those totals were often founded a decade or more ago, those companies have since matured, and many of the angels who backed them have moved into formal funds, slowed down, or stopped angel investing entirely. A lifetime ranking rewards being early to the last cycle. It says much less about this one.”
The ranking uses lifetime U.S. unicorn investments. That makes it a historical contribution map. It tells us who appeared on the cap tables of companies that eventually became unicorns. It is useful for understanding the last cycle and historical merit, like a hall of fame. It is less useful for deciding who a founder should target now, who an LP should underwrite now, or who a platform team should prioritize now.
The Lifetime List Answers a Real Question
Strebulaev’s leaderboard membership is aligned to most observers intuition. Y Combinator leads with 113 unicorn investments, followed by Plug and Play at 52 and 500 Global at 41. Sand Hill Angels is the best-placed angel group at 31. It does not measure efficiency (how many failures it took to make a success).
The list also reveals something important beneath the institutional names. Past the top five, the ranking becomes mostly individuals. In Strebulaev’s count, 271 of the 304 investors in the Top 200 are individuals. That is 89%. In the Top 100, the individual share rises to 91%.
The angel market looks institutional at the very top because accelerators invest in hundreds or thousands of companies. But underneath that, early-stage venture still depends heavily on people writing personal checks, building reputations, and getting into rounds before firms are willing or able to move.But still, a lifetime ranking is a weak decision tool and Ilya acknowledges that.
The Blind Spot Is Time
Every lifetime leaderboard has the same bias. It rewards being early to companies that were often founded a decade or more ago. Some of those investors are still active. Some are not. Some have moved into formal funds. Some write fewer checks. Some were attached to a platform, geography, or network structure that no longer exists in the same form.
That does not make their contribution less real. It does make the ranking less actionable.
For a founder raising now, “who backed the most unicorns over a career?” is a different question from “who is currently effective at my stage?”
For an LP, “who was present in the last cycle’s eventual unicorns?” is a different question from “who is showing repeatable signal in the present market?”
For a venture firm, “who has lifetime prestige?” is a different question from “who should I build a sourcing relationship with this year?”
The current market needs a ranking system that is current and measures recency, only that can align to real-time decisions being made.
The SignalRank Cut Starts With Stage
SignalRank’s 10-year leaderboard dashboard takes a different approach. The stage specific scores on the tabs (shown as rankings over time) is not a lifetime honor roll. It is a current stage-specific leaderboard. The dashboard cut used here is the current-year stage specific ranking with a 2016-2025 scored period, and it includes companies, unicorns, efficiency, investor score, and rank. Each year measures only three years of prior investments at the stage.
That changes the answer immediately.
Y Combinator still ranks first in the SignalRank Pre-Seed cut, but the reasons are visible. In the current dashboard data, YC has 2,357 companies, 5 unicorns, 0.21% efficiency, and a Pre-Seed investor score of 32,128.66. That is a scale story. It is not the same story as “113 lifetime unicorn investments.”
Below YC, the list changes character quickly. Heroic Ventures ranks second with 5 companies, 1 unicorn, and 20.00% efficiency. Alameda Research ranks third with 8 companies, 1 unicorn, and 12.50% efficiency. Sequoia ranks fourth with 58 companies, 2 unicorns, and 3.45% efficiency.
The point is not that SignalRank has found a single perfect ordering. The point is that the ranking exposes the tradeoff. Scale, efficiency, and stage fit are all visible.
Stage Specificity Changes the Target List
A founder does not raise “venture capital” in the abstract. A founder raises a Pre-Seed, Seed, Series A, or Series B round. The investor set changes at each step. So should the ranking.
In the SignalRank data, the top of Pre-Seed is not the same as the top of Seed or Series A.
At Seed, Andreessen Horowitz ranks first in the current cut, with 202 companies, 17 unicorns, and an 8.42% efficiency rate. Lightspeed ranks second. SV Angel ranks third. Lux ranks fourth. Sequoia ranks fifth.
At Series A, Andreessen Horowitz again ranks first, but the underlying profile changes: 252 companies, 37 unicorns, and 14.68% efficiency. Sequoia is second with 145 companies, 29 unicorns, and 20.00% efficiency. General Catalyst, Lightspeed, SV Angel, Thrive, Founders Fund, Redpoint, Lux, and Benchmark fill out the top ten.
Those are not small differences. They are the differences a founder, LP, or business-development team actually needs.
The current leaderboard is computed daily against three years of stage specific investments.
Recency Makes the Ranking Usable
The other upgrade is time range.
Historically elongated rankings are good at assigning credit. They are bad at measuring current opportunity.
Dan Gray made a similar point in his recent X post:
The simple explanation is that any ranking based on a cumulative score, rather than a normalised rate or efficiency metric, will reward scale over quality and favour larger firms with more activity.
As he says, if the objective of the ranking is to reflect impact or relevance that's fine, but it should not be perceived as a ranking of performance.
Venture changes too quickly for career-to-date counts to be the primary decision signal. Partners move. Funds change strategy. Platforms scale up or down. Some angels become institutional managers. Some stop writing checks. Some firms remain active but shift stage.
That is why a limited current window is more useful. It narrows the question from “who has ever been good?” to “who has been effective in the relevant recent market?”
The SignalRank dashboard does not just count outcomes. It places the investor in a stage-specific frame and shows the ingredients of the rank. At Pre-Seed and Seed, efficiency matters because the stage is noisy and sample sizes vary wildly. Later stages can lean more heavily on MOIC and realized company quality because there is more information in the round itself.
This is the practical difference:
A lifetime ranking is a reputation archive.
A stage-specific recent ranking is a targeting tool.
A transparent scoring system is a way to improve both.
How to Upgrade Ilya’s Ranking
The right response is to build on Ilya’s work.
A more actionable ranking would add five layers:
Stage: separate Pre-Seed, Seed, Series A, and later participation.
Time window: show lifetime, 10-year, 5-year, and post-2020 rankings separately.
Activity status: identify who is still writing checks, who has moved into a fund, and who appears inactive.
Investor type: separate individuals, angel groups, accelerators, rolling funds, micro-funds, and formal venture firms.
Efficiency and denominator: show not just unicorn counts, but companies backed, hit rate, and minimum sample thresholds.
Strebulaev already points toward this in his own discussion of the post-2020 list. The line between individual angel and micro-VC is blurring. Todd and Rahul’s Angel Fund appears next to Todd Goldberg and Rahul Vohra individually. Operators are pooling capital into named vehicles. The best early check often comes attached to a small group rather than one person.
That observation is important. It means the ranking system has to evolve with the market structure. SignalRank’s threshold system is one way to do so.
The Takeaway
Strebulaev’s ranking is valuable as a historical map. It shows who accumulated exposure to U.S. unicorns over time, and it gives the angel market more empirical structure than it usually gets.
But historical visibility is not the same as current usefulness.
For decisions, rankings need stage, recency, activity, investor type, and efficiency. Without those, the list rewards the last cycle. With them, it can help founders find the right investor, help LPs understand repeatable edge, and help the market distinguish reputation from current signal.
