TechFlow Logo
Login/ Sign up
ETH Gas
Gwei
Fear
gas
Stanford Business School Internal Guide: Putting Fame Aside, Here’s How to Precisely Identify the Real Top 5% of "Elite Venture Capital Firms"

Stanford Business School Internal Guide: Putting Fame Aside, Here’s How to Precisely Identify the Real Top 5% of "Elite Venture Capital Firms"

2026.08.25
Share
Founders shouldn't just focus on Sequoia.

Author: Ilya Strebulaev

Compiled by: TechFlow

TechFlow Digest: A Stanford Graduate School of Business professor reconstructs venture capital rankings using data from 230,000 investments, revealing that 90% of the industry's profits are captured by just 5% of VCs. For founders, choosing the wrong investor could mean returns differ by an order of magnitude; this article provides actionable criteria for selection.

This is a guest article by Ilya Strebulaev. He teaches venture capital and private equity at Stanford GSB and leads the school's venture capital program. He publishes VC research, investor rankings, and founder-focused fundraising guides in his newsletter.

Every founder who has raised money from VCs has heard the same names: Sequoia, Andreessen Horowitz, Benchmark. But there is one number that should influence your fundraising strategy more than these names. Based on my estimates, approximately 5% of venture capitalists generate about 90% of the entire industry's profits. If you are deciding where to take money, the most important question is: Is your investor among this top 5%? Brand name does not accurately reflect this.

Figure: The top 5% of VCs generate roughly 90% of industry profits. Whether your investor belongs to this group matters far more than their brand recognition.

Until now, there has been no transparent, fully data-driven method to answer this question. The Forbes Midas list, which the industry relies on, is largely a black box. When we attempted to reverse-engineer it based on its public methodology, even our best-fitting replication missed 49 of its own top 100. Among investors appearing in both our rankings and the Midas list, the correlation was only about 0.27.

Therefore, Blake Jackson and I built an alternative: the 2026 Strebulaev-Jackson VC Ranking, based on over 230,000 investments made by nearly 13,000 venture capitalists across more than 5,000 companies over 30 years. Every score can be traced back to a specific investment in a specific company on a specific date. We apply no editorial judgment, make no manual adjustments, and rely on no self-reported data from companies.

This article contains the complete top 100 list, the methodology behind it, and—because a ranking alone won't get you a good investor—what you should do once you have it.

What Exactly Does This Score Measure?

Six factors drive this score. They are worth reading closely, as they collectively reward behaviors founders want from their investors.

Dilution. A 10% stake in Round 1 is no longer 10% at exit. Both companies might sell for $1 billion, but if one went through four funding rounds along the way, early investors get diluted in each round. We track equity changes for every investment across subsequent rounds.

Net Profit. Turning $10 million into $2 billion is different from turning $1 billion into the same $2 billion. We subtract the cost of each investment, reward capital efficiency, and penalize throwing massive amounts of cash around just to manufacture a few headline victories. In our dataset, net returns are negative for roughly three-quarters of investments.

Value Add. Lead investors holding board seats contribute more than passive check-writers, so we award extra points for these roles.

Attribution Between Firms and Individuals. Investors move between firms, so when a partner lands a firm's best deal after jumping ship, both firms should get credit. We split it: 25% goes to the firm where the investment occurred, and 75% goes to the firm the partner currently works for. This reflects academic evidence that most return variation stems from individuals rather than institutions.

2026 Results: Top Firms

Sequoia tops the list with 10,158 points. Andreessen Horowitz ranks second with 8,292. Accel, DST Global, and Tiger Global round out the top five. Two names in the top 20 deserve any founder building a target list to reconsider. Parkway Venture Capital, founded in 2019, ranks 19th thanks to Figure AI. Notable ranks 20th. Neither is a household name, and both outrank firms with vastly greater brand recognition.

Figure: Sequoia leads with 10,158 points; two of the top 20 are names most founders have never encountered

Ranks, firms, headquarters, founding year, rank scores, and top deals. "Top deal" refers to the single investment that contributed the highest score for that firm according to our methodology, not necessarily its most famous or highest-valued holding.

Figure: All 100 institutions with headquarters, founding year, score, and top deal; 62 based in California, 19 in New York, 6 in Massachusetts and Texas each

Scores are rounded to the nearest integer. Geographically: 62 of the top 100 firms are headquartered in California, 19 in New York, 6 in Massachusetts, and 6 in Texas. Regardless of scaling happening at the seed and angel stages, the institutional core of U.S. VC has not left its historical heartland.

Four patterns in this table matter more to founders than the ranking order itself.

The Gap Between Good and Famous Is an Order of Magnitude

By rank 10, scores drop to around 3,000, less than a third of Sequoia's. By rank 100, the score is 245. The number-one firm scores roughly 41 times more than the 100th-place firm.

Figure: By rank 10, scores fall below a third of Sequoia's; by rank 100, the score is 245, making the top firm's score 41 times that of the hundredth

The VC power law is usually described at the individual deal level, but it applies just as strongly to the firms themselves. For founders, this means the gap between a top-decile investor and merely a well-known one spans an order of magnitude—a difference worthy of serious pursuit.

Counting Unicorns Almost Doesn't Matter

This ranking is not a contest for unicorn counts. SV Angel invested in roughly 139 unicorns yet ranks 31st. Insight Partners invested in about 124, ranking 28th. Felicis holds 58 unicorns, ranking 74th. Meanwhile, DST Global has 62 unicorns, ranking 4th. Thrive has 47, ranking 8th.

Figure: SV Angel invested in 139 unicorns but ranks 31st, Thrive invested in 47 and ranks 8th; the methodology rewards actual value retained post-dilution, not logo count

Why do firms with far fewer unicorns rank much higher? Because the methodology rewards realized value, not the sheer number of standout deals. An institution that cuts a small, heavily diluted check for a later-stage unicorn scores low: dilution adjustments shrink the stake, profit adjustments deduct costs, and in crowded cap tables, tiny early positions may be worth little at exit. An institution holding large, concentrated, board-level stakes in a handful of winners scores highly. Per-unicorn, the most diversified firms score around 6 points, while the most concentrated exceed 80.

For founders, this gap reveals which investors are committed and actively involved, versus those who cut many small checks and hope one or two hit.

Your Best Investor Might Be One You've Never Pitched To

Bessemer, the oldest among the top 100, traces its VC roots to the 1970s. Inflection Ventures, the youngest, was founded in 2022. Eight of the top 20 were established before 2000, proving how enduring true VC brands can be. Yet 21 firms in the top 100 were founded in 2015 or later. Several rose quickly on recent, rapidly appreciating bets.

Figure: 21 of the top 100 were founded in 2015 or later; performance records compound over time, and a decade is enough to build one. Source: The VC Corner, 2026

A cohort of life sciences firms makes the list based on therapeutic bets rather than software: OrbiMed (27), Atlas Venture (38), ARCH (49), Versant (61), and Sofinnova (75), all leveraging concentrated, capital-efficient plays. Crypto-native institutions like Paradigm (34), Pantera (76), Multicoin (84), and Polychain (94) rise via another opportunity set. The methodology is sector-neutral; it measures value created minus costs and decay, wherever it occurs.

If you are founding a company in a specialized field, your ideal investor may be a domain expert. And they will never appear at the top of generalist media lists.

AI Is Already Reshaping the Top of the List

Among the top 100, 23 firms have their highest-scoring single investment in a frontier AI or AI infrastructure company. This accounts for nearly a quarter of the list. These anchor a cohort of companies that either did not exist or were very small just five years ago.

Figure: Nearly a quarter of the top 100 list frontier AI or AI infrastructure companies as top deals; these companies mostly did not exist or were very small five years ago

The decay factor means this reshaping happens in real time. Institutions that concentrated early bets on leading AI firms—like Thrive on OpenAI, Menlo on Anthropic, Lightspeed on Mistral—get immediate returns instead of waiting years for redemption.

What a Ranking Cannot Tell You

Some founder somewhere is already pasting this table into a spreadsheet, sorting by rank, and preparing to email the top 1 to 100 firms in order. I understand the impulse. Rankings are tempting precisely because they pretend to think for you.

But if there is one thing you should not do, it is use this list that way. Nor should you use any list published by Ruben, myself, or anyone else. A ranking is an argument for what constitutes "good," compressed into a single number. The order is the least interesting part. What matters is the reasoning underneath. You need to check whether that reasoning aligns with what you actually want to accomplish, and what your own research uncovers when applied to your situation. Read the methodology, challenge parts you disagree with, then do the homework yourself.

Rankings can help narrow your list, but the final choice should be driven by three things. No score can fully capture any one of them.

Consider fit, not just rank. This ranking measures diverse playstyles with a single ruler. Cross-over and hedge-fund-style firms like Tiger Global (5), DST Global (4), Dragoneer (23), Altimeter (24), Coatue (29), and Greenoaks (44) typically take large minority stakes. They rarely or never take board seats. Sequoia and Benchmark operate a deeply involved, heavy-board model. Sutter Hill (26) essentially incubates from zero. All these models score highly because they create value. But from a founder's vantage point, they feel completely different. Founders seeking deep partners should pick different firms from the same top 100 than those seeking capital plus autonomy. High rank cannot tell you which model you are facing.

Due diligence on partners, not brands. A striking fact in our data is that half of the top 100 institutions have absolutely no partners in the personal top 100. Institutional strength and individual strength are worlds apart. This is precisely why our methodology splits the scores. Sitting on your board is not a brand; it is people. Ask explicitly who will work with you, what else they are juggling, and how long they have been at the firm. Then conduct background checks with founders they have backed, including those whose companies failed. How an investor performs during down rounds is the information you need most. No ranking, including ours, provides that.

Remember, this is closer to marriage than a transaction. Investors will be your partners for seven, ten, sometimes fifteen years. That is longer than many marriages and far harder to exit. You can sell a house, pivot your product, or replace a team. But you cannot easily remove an investor from your cap table or board. This asymmetry should make you slow down, especially in that moment when a competitive funding round pushes you to accelerate.

The value-add factor exists in our ranking because deeply involved investors clearly impact outcomes. But the same board seat can open doors or block sales, overturn strategy, or fire CEOs. So do not only ask whether an investor will help you win; also ask whether you want that person present on your worst day. An investor slightly lower ranked but truly aligned with your interests beats a higher-ranked one who is not.

How to Use This List

Screen by performance. That top 5% generates 90% of the profits and is worth reaching out to genuinely. The table above shows where they are.

Filter by fit. Stage, sector, ticket size, and the level of involvement you want.

Go deeper to look at individuals. A firm gets you the meeting, but for years afterward, you work with one person.

Background-check downside risk. Talk to founders whose companies struggled; do not just look at star cases on an institution's website.

What Comes Next

We will expand the same six factors to individual investors, not just firms. The contrast with conventional wisdom becomes even sharper: more than half of our top 100 individual VCs do not appear on the 2026 Forbes Midas list at all. We will also expand the rankings internationally, releasing sector rankings starting with biotech and AI. This will incorporate verified data from both firms and investors, publishing roughly 25 years of historical series to make the rise and fall of institutions directly visible.

Q: Which VC firm performed best?

A: Sequoia Capital leads the 2026 Strebulaev-Jackson VC Ranking with 10,158 points, ahead of Andreessen Horowitz's 8,292 and Accel's 4,576. The score measures realized value after accounting for dilution, costs, and temporal decay.

Q: How much better are top VC firms compared to average ones?

A: The number-one firm scores roughly 41 times that of the 100th. By rank 10, scores have fallen below a third of Sequoia's. This indicates that the VC power law applies not only to individual deals but to the firms themselves.

Q: Does investing in more unicorns mean a VC performs better?

A: No. SV Angel invested in roughly 139 unicorns, ranking 31st. Thrive invested in 47, ranking 8th. The ranking rewards realized value post-dilution, not the count of nine-figure logos.

Q: How concentrated are VC returns?

A: The top 5% of VC firms generate roughly 90% of the industry's profits. Whether your investor falls into this top 5% is therefore decisive, not marginal.

Q: Does becoming a top VC require a decades-old institution?

A: No. 21 of the top 100 were founded in 2015 or later. Thus, ten years of concentrated, well-timed holdings are enough to compete alongside institutions with 50-year pedigrees.

Q: Which AI companies do top VCs cite as their best investments?

A: 23 of the top 100 list a frontier AI or AI infrastructure company as their highest-scoring single deal. OpenAI was nominated by four firms, xAI by three, and Anthropic and Perplexity by two each.

Q: Where are the best VC firms located?

A: Of the top 100, 62 are headquartered in California, 19 in New York, 6 in Massachusetts, and 6 in Texas.