Study Reveals Overemphasis on Founder Education Leads to Poor VC Returns

Author: Odin

Compiled by Deep潮 TechFlow

Shenchao Overview: While global VCs are all saying "invest in people, not projects," data from the University of Chicago reveals a harsh truth: investors make their worst decisions when they overly rely on founders' educational backgrounds.

This obsession with academic credentials costs the industry hundreds of millions of dollars each year. More ironically, true investors like Thiel and YC don’t look at resumes at all—they evaluate the complex whole formed by the founder and their idea. For crypto investors, this serves as a reminder to be wary of institutions that merely pattern-match for prestigious school backgrounds.

Long ago, eight researchers from Shockley Semiconductor walked into the office of a young banker in San Francisco named Arthur Rock. The “Traitorous Eight” proposed an idea: they wanted to start a competing company. Rock saw in them perhaps a special kind of frustration—brilliance with nowhere to express it—and set out to help them raise funds, founding Fairchild Semiconductor, widely regarded as the seed of Silicon Valley. This is the story of how Rock, the first believer in the team, became the first modern venture capitalist.

Rock has held for decades the belief that supporting talent is at the core of venture capital. He often says that a strong management team can find great opportunities even if it means moving beyond its current market.

His peers held different views. Tom Perkins of Kleiner Perkins focused on technology, asking whether it was proprietary and clearly superior to alternatives. Don Valentine, who founded Sequoia after working in marketing at Fairchild, was obsessed with the market. In the mid-1980s, when Sequoia considered an early investment in Cisco, most peers declined—the founding team was seen as weak. But Valentine still invested, reasoning that the networking market was so vast that even an average team could sell massive quantities of equipment.

These three individuals gave rise to three distinct philosophies of American venture capital; but Rock emerged victorious in the cultural battle. “Venture capital is a people business” is not only a brilliant slogan—it places founders at the center of the story. And if you’re selling capital to founders, this is exactly what they want to hear.

But is it really that simple? What does a "human business" actually look like?

Today, nearly every venture capital firm claims to prioritize founders.

In 2016, four economists—Paul Gompers, William Gornall, Steven Kaplan, and Ilya Strebulaev—surveyed 885 venture capitalists at 681 companies to understand how they make decisions. This study represented the most comprehensive analysis of industry decision-making and appeared to put an end to the philosophy of Perkins and Valentine.

Approximately 53% of early respondents identified the founder as the single most important factor in their investment decision. Business model and product (Perkins’s traditional focus) were chosen by about 10%. Market and industry (Valentine’s focus) were selected by about 6%. The remainder were distributed among valuation, fit with the fund, and the investor’s own value-add capabilities.

96% (92%) of venture capital firms consider the team an important factor, and 56% (55%) consider it the most critical factor for success (failure). The team is the most important factor across all sub-samples, particularly for early-stage and IT venture capital.

—— How Venture Capitalists Make Decisions, by Gompers, Gornall, Kaplan, and Strebulaev

Looking at other responses in the survey, 9% of investors admitted to using no financial metrics at all, a figure that rises to 17% among early-stage investors. An industry so reliant on qualitative judgment should have carefully considered its evaluation criteria and how to track outcomes.

Unfortunately, the answer remains an ambiguous promise—to invest in "the best founders"—without clarifying what that means or why.

The findings show that venture capitalists are not adept at reflecting on their decision-making processes. Even in controlled experiments where the amount of information considered was significantly reduced, venture capitalists lacked deep insight into how they made their decisions.

——Lack of Insight: Do Venture Capitalists Really Understand Their Decision-Making Process? by Andrew Zacharakis and G. Dale Meyer

Thus, the founder-first venture capital approach has fostered a pandemic of lazy thinking, permeated by bias and credentialism. This, in turn, is reflected in declining performance and frequent scandals involving fraud and negligence.

In 2022, University of Chicago Booth School of Business economist Diag Davenport put a price tag on the losses the industry suffered due to this oversimplified attitude.

Davenport built a machine learning model on a dataset of over 16,000 startups, representing more than $9 billion in committed capital. He trained the model using only the information available to investors at the time of their decisions and asked: Of the investments that venture capitalists actually made, how many could have been identified in advance as worse than investing the same amount in standard public market alternatives? The answer is about half.

By eliminating the worst half of investments and reallocating capital to public market alternatives, Davenport found that venture capital returns in the sample could have been 7 to 41 percentage points higher. In his analyzed data, this equated to over $900 million in avoidable losses. The cost of poor investments, measured as the spread relative to external alternatives, was approximately 1000 basis points.

Davenport trained two parallel algorithms: one to predict which startups would become the best investments, and another to predict which would become the worst. When he compared the signals each model relied on, a strange pattern emerged. The algorithm built on good outcomes depended on product features, while the one built on bad outcomes heavily relied on founder backgrounds. When investors made good decisions, they paid closer attention to the idea. When they made bad decisions, they seemed to focus more closely on the team.

To test for overweighting, Davenport built a separate model using only founder education data and asked: Would two companies that appear equally promising under the full model receive different investment outcomes due to differing performance under the education-only model? The model revealed that investors systematically overweight education, and they do so most severely for startups that later performed the worst.

Investors appear to be convinced that the founder-priority world model is correct. This may lead investors to overlook predictive features, and feedback loops that never notice or learn continue to persist, consistent with the model and evidence presented by Hanna et al. (2014).

——The Predictable Bad Investment: Evidence from Venture Capitalists, by Diag Davenport

Davenport’s paper is part of a growing body of research reaching similar conclusions, showing that investors overweight superficial founder attributes, leading to predictable poor investments (false negatives) and predictable missed opportunities (false negatives).

There is a structural explanation: in venture capital, "success" is more easily measured through incremental funding rather than distant exits, and if investment decisions become mere checkbox exercises, funding friction decreases.

At some point, the industry convinced itself that the ability to raise capital was itself an ideal founder trait, and this logic became recursive. Investors began pattern-matching against the archetype of founders most likely to raise their next round, making that archetype easier to fund and reinforcing the pattern. As a result, the quality of returns generally declined, while the speed of capital (and fee income) accelerated.

This cycle is explained by economist Daniel Kahneman, who described how even sophisticated professionals can be tempted by simple, coherent ideas when they align with the right incentives—even when those ideas lead to clearly poor outcomes.

The statistical evidence of our failures should have undermined our confidence in judgments about specific candidates, but it did not. It should also have led us to moderate our predictions, but it did not. We know as a general fact that our predictions are hardly better than random guesses, yet we continue to feel and act as if each specific prediction is valid.

——Don’t Blink! The Danger of Confidence, Daniel Kahneman

This creates an interesting puzzle. Data shows that over-weighting founder attributes leads to poorer investment decisions, particularly in the worst-performing deals. Yet some of the most successful companies in the industry are also the most founder-centric.

Founders Fund has spent two decades supporting unusual people before others were willing to. Peter Thiel also created the Thiel Fellowship for young entrepreneurs without college degrees, resulting in incredible success stories.

Y Combinator has operated for two decades on the premise of identifying exceptional founders. In fact, the program has proven to reduce credentialism in venture capital by providing investors with an alternative signal.

If founder-first thinking were merely a systemic pathology, the companies most committed to it should perform the worst. Instead, they are the best.

The answer is actually quite straightforward. When top investors say "founders first," they mean something far more nuanced than the superficial interpretations common across the industry.

The desire to reduce founder success to a checklist of predictable traits is a modern manifestation of the great man theory—the belief that history is shaped by exceptional individuals born with greatness, ignoring how success itself forges these qualities.

A successful company with a strong track record? The leader appears visionary, charismatic, and highly communicative. A company experiencing downturns? The same leader seems hesitant, misleading, or even arrogant.

—— The Halo Effect, Phil Rosenzweig

For example, entrepreneurs like Elon Musk have shaped investors’ expectations of hard tech founders through numerous stories highlighting his cross-domain fluency, discipline, and determination. As a result, investors seek these traits in first-time founders, unaware that Musk developed these qualities over time—thereby denying others the opportunity to do the same.

Also consider Thiel’s investment in Mark Zuckerberg, the Harvard dropout. Today, it is often cited as an example of Thiel’s early ability to identify exceptional founders. However, contemporary records show that Thiel was drawn to Facebook itself, its early traction, and Zuckerberg’s specific framing of online identity issues.

If Zuckerberg were running a flower delivery startup, would Thiel recognize anything in him? It’s hard to imagine. It was Thiel’s vision of how a college social network should function—and the specific form Zuckerberg had already given it—that held the magic Thiel was looking for.

Indeed, at Andrew Ross Sorkin’s DealBook conference, Peter Thiel was asked how he evaluates founders, and his response aligned with the Facebook example.

I don’t separate ideas, business strategies, and technology too much from people—they’re all part of some complex bundled deal.

—Peter Thiel, Co-founder of Founders Fund

He said he cannot evaluate the founder without assessing the quality of the ideas they are researching, and he cannot evaluate the ideas without understanding how the founder shapes them. The two are inseparable.

Academia has also developed a complementary argument. In a 2022 paper published in the Journal of Business Venturing Design, Mattia Bianchi and Roberto Verganti from the Stockholm School of Economics and Politecnico di Milano argue that entrepreneurship has been systematically misunderstood as an activity focused on solving problems, when in fact it is primarily an activity of identifying problems.

In their framework, the founder’s most important creative act is identifying and defining a problem worth solving. Everything else—whether the pitch deck, market entry plan, or product roadmap—derives from the quality of this initial definition.

Viewing the identification of problems as a design act rather than merely a discovery expands the potential impact of design practice—from creatively generating solutions to creatively generating the problems themselves. Re-defining problems speculatively is another lever for breakthrough innovation, as unconventional problem formulations can open up unexpected pathways to solutions. — Bianchi and Verganti, “The Entrepreneur as a Designer of Worthwhile Problems”

If this framework is correct, then the core dichotomy between jockey and horse is flawed. Founders should be evaluated based on the problems they choose to solve and the specific frameworks they use to understand those problems. Ideas cannot be assessed in isolation, as they are material expressions of the founders’ beliefs about what the world will look like a decade from now. Each illuminates the other, and any investor claiming to evaluate them separately is likely to fail at both.

Nabeel Hyatt of Spark Capital articulated this combined approach well. When asked how to distinguish genuine executors from founders who merely appear to meet many criteria, his answer was surprisingly direct.

We distinguish between slick con artists and real doers by looking at what they’ve actually built. I’ve never evaluated a company and said, “This person deserves a $15 million check” based solely on seeing a product or using a website. You look at the product, and then you learn about the person behind it by evaluating the product. —Nabeel Hyatt, General Partner at Spark Capital

The product is a reflection of the founder's ambition, deeply revealing their judgment, priorities, and the problems they choose to solve.

An investor who says “I’m an investor” but hasn’t thoroughly researched the product is either investing in superficial models or in charisma and personal appeal—both of which are precisely the habits that reliably lead to predictable poor investments.

When Sam Altman shared his application screening heuristics at the 2016 Khosla Ventures summit with Keith Rabois, he expressed the same idea in slightly different wording:

The hardest trait to identify is determination. There are several other themes we pay attention to along the way: clarity of vision, communication skills, and the non-obvious brilliance of an idea—we examine these very carefully. These are things you can’t always judge correctly, but you usually have a fair amount of data to go on, unlike determination, which is much harder to assess. — Sam Altman, former President of Y Combinator

He didn’t talk about the founder’s brilliance; he talked about the brilliance of the idea, specifically defined as “non-obvious,” indicating that the founder chose a novel problem. He also emphasized the clarity of the vision, which shows how well they perceive and articulate that problem. Of course, there’s also their determination to commit to the process.

In the language of Bianchi and Verganti, he is the founder as a designer of worthy problems to solve.

When investors say they are investors, they may mean two different things.

The first view holds that attributes such as background, resume, charisma, and past fundraising success signal more than what founders choose to spend their time on. Essentially, this perspective treats founders as interchangeable commodities that can be ranked. This is the version most directly refuted by Davenport’s data.

The second, rarer version holds that the subject being evaluated is a unique alchemy of people and ideas. The investor’s job is to assemble a complete picture: the selection of the problem, the form of the solution, and the character of the team. Only then can they fully perceive the opportunity before them.

They are easily confused because they use the same vocabulary. Both express themselves in language that supports people and celebrates human potential. The first is lazy and is fully rewarded by industry norms. The second is difficult, often misunderstood, but clearly the path to higher-quality investments.

The argument is not that investors should abandon qualitative team analysis and return to the approaches of Perkins and Valentine. The conclusion is simply that teams cannot be effectively evaluated outside the context of what they are doing—and attempting to do so is precisely where investors fall into problematic pattern-matching.

That’s why the atomic unit of entrepreneurship is neither the founder nor the idea, but the unity of both. Venture capitalists must step back far enough to see both elements simultaneously and evaluate them as a single entity.

Instead of dwelling on the old question of jockey versus horse, an investor’s job is to identify the centaur.

Note: A 2009 paper provided empirical support for evaluating companies based more on ideas by analyzing how many companies had already replaced their leadership teams or core products at the time of their IPO. However, this covered a period when VCs frequently brought in new executives before going public, and now appears less relevant.

Run your venture capital firm on your phone with Odin.