Most Angel Investors Lose Money. Here's the Data — and the Portfolio Size That Actually Works.
Most Angel Investors Lose Money. Here's the Data — and the Portfolio Size That Actually Works. Most Angel Investors Lose Money. Here's the Data — and the Portfolio Size That Actually Works. By Jeff Ba
Most Angel Investors Lose Money. Here's the Data — and the Portfolio Size That Actually Works.
By Jeff Barnes, MBA | July 1, 2026
- 52% of individual angel deals return less than the capital invested, according to the landmark Wiltbank & Boeker study of 1,137 exits.
- 7% of deals generate 75% of all dollar returns. Without 20-30+ investments, your odds of touching one of those winners are statistically poor.
- The median angel has only 10 investments and 2 exits, far below the minimum portfolio size the math actually demands.
- Hold periods run 4.8 to 8 years. You are locking up capital, probably longer than you think, for a return profile that underperforms an index fund for most participants.
According to the landmark Wiltbank & Boeker study published by the Kauffman Foundation and Angel Capital Education Foundation in 2007, 52% of all angel investment exits return less than the capital put in. Not underperform. Not disappoint. Return less than invested. More than half of individual deals are losses. That is not a fringe result from a bad sample. It comes from 1,137 exits across 539 angels who tracked and reported their results. If anything, that sample skews toward more disciplined investors than the average solo angel writing checks at pitch competitions.
The narrative you hear at startup events is different. Angels who hit big talk loudly. The ones who quietly wrote off $50,000 into a startup that dissolved do not give keynotes. That survivorship bias shapes the entire public perception of angel investing. Here is what the actual data says.
The Uncomfortable Truth: Most Angel Deals Lose Money
The 52% loss rate stops people cold. But it is not even the most clarifying statistic. The median deal is a loss. The average looks fine on paper: 2.6x over 3.5 years, roughly 27% IRR. That sounds good until you understand what drives it.
A small number of extraordinary outcomes pulls the average far above where most investors land. The UK data from Wiltbank's 2009 NESTA and British Business Angels Association study shows the same pattern: a 2.2x overall return in under four years, but 9% of exits generating 80% of positive cash flows.
The NBER Working Paper 33231 (December 2024, revised January 2026) by Karlsen, Kisseleva, Mjos, and Robinson at the Norwegian School of Economics and Duke University's Fuqua School is the most rigorous population-level test to date. Using Norwegian administrative data from 2003 to 2018 (actual tax and registry records, not self-reported surveys), they found that more than one-third of all realized angel investments are a total loss. Only about 25% return more than the capital invested on an undiscounted basis.
Three out of four individual deals either lose money or merely return what you put in, with no compensation for the years you waited, the risk you took, or the opportunity cost of capital that could have been in an index fund.
The Power Law: Why 7% of Deals Do All the Work
Angel returns follow a power law. This is not a metaphor. It is the actual mathematical structure of how returns distribute across deals. In the Wiltbank & Boeker dataset, 7% of exits returned more than 10x invested capital. Those 7% of deals account for 75% of all dollar returns in the entire study. The top 10% of angel performers earned 50% of total capital returns.
The NBER Norwegian study confirms the extreme right-skew. The top 1% of angel investments return more than 50x invested capital. The mean payback across the full dataset is roughly 2x, but that mean is driven entirely by that tiny top slice. For everyone else, the distribution looks much worse.
This structure has a practical consequence: angel investing is a volume game whether you want it to be or not. You cannot pick your way to a good outcome through superior judgment alone. The variance is too high at the deal level. The only reliable path to the power-law upside is being in enough deals to have a statistical chance of holding one of those 7% winners.
The portfolio math is not subtle. With only 10 investments, your probability of holding even one 50x-plus outlier sits at roughly 28%. Get to 25 investments and that probability rises to 57%. At 50 investments it reaches 82%. Below 20 deals, you are gambling that luck drops a winner into your small sample.
The Portfolio Math: Why 10 Deals Is Not Enough
Here is the problem. The typical angel investor is nowhere near 20 to 30 deals.
The Angel Capital Association's Angel Funders Report and the original Wiltbank & Boeker dataset both point to the same median angel profile: roughly 10 total investments, 2 exits or closures. Most angels are operating with portfolios that are less than half the minimum size required for the math to work in their favor.
At average angel check sizes of $25,000 to $50,000, building a 25-deal portfolio means $625,000 to $1.25 million deployed before the portfolio size gives you reasonable odds at the return profile that justifies the illiquidity. That is not a figure most retail-level accredited investors can absorb, especially with those dollars locked up for years.
| Number of Investments | Probability of 1+ Outlier (50x+) | Estimated Capital Required* |
|---|---|---|
| 10 deals | ~28% | $250K – $500K |
| 25 deals | ~57% | $625K – $1.25M |
| 50 deals | ~82% | $1.25M – $2.5M |
| *Assumes $25K–$50K average check size. Source: Portfolio probability analysis citing Kauffman Foundation and Cambridge Associates datasets. | ||
Most angels never get there. They make a handful of bets, experience the base-rate reality of losses on most, and either quit or attribute the result to bad luck. In most cases, luck had little to do with it. The portfolio was too small for the asset class's return distribution to resolve in their favor.
What the Norwegian Study Confirmed
The Karlsen, Kisseleva, Mjos, and Robinson paper addresses a persistent criticism of American angel return studies: self-selection and self-reporting bias. The Wiltbank data relies on angels who chose to participate in formal groups and report their results. Not a random sample.
Norwegian administrative records capture every angel investment at the population level, regardless of whether the outcome was good or bad. The results are not more optimistic than the self-reported US data. More than one-third of realized investments: total loss. Only about 25% return more than invested capital.
The paper also surfaced a finding that cuts against intuition: wealthier angels sort into better-performing startups but earn relatively lower returns than other investors in those same companies. The authors suggest non-financial motives (access, status, network effects) drive high-net-worth angel activity more than return expectations. At the top end of the angel market, the "smart money" is partially optimizing for something other than financial return.
The Time Problem
The 2.6x average return from Wiltbank & Boeker sounds acceptable. Until you factor in time. That average came with a 3.5-year average hold. But the 3.5-year figure applies to angel-stage deals that found an exit relatively quickly. More realistic scenarios look different.
CB Insights data, analyzed by David Teten and cited in Stephen Morrissette's "A Profile of Angel Investors", shows VC-backed M&A exits averaging five years from first investment to close, with IPO exits averaging seven years. Angels investing at the pre-VC seed stage add time before a VC even enters the picture. The realistic hold period for an angel investment is 4.8 to 8 years. Not 3.5.
At a 7-year hold, a 2.6x gross return is roughly a 14.7% IRR. That is before taxes on gains. Before the opportunity cost of capital that was not in the S&P 500 during one of the stronger equity decades on record. After accounting for losses on the 52% of deals that returned less than capital, the median angel's actual experience looks considerably worse than any headline IRR number suggests.
The Cambridge Associates US PE/VC Benchmark Commentary for first half 2025 is a useful reference: professional VC funds have struggled to consistently outperform public small-cap equities in recent periods. Individual angels, without dedicated sourcing teams or institutional due diligence, start from a harder position than those VC benchmarks reflect.
Who Should Actually Angel Invest
None of this means angel investing is a bad activity. It means it is a different activity than what most participants think they are doing.
If you are angel investing for financial return, be honest about the minimum conditions:
- You can deploy $25,000 to $50,000 per deal across at least 25 companies, keeping each position to no more than 5-10% of your liquid portfolio.
- You can write off 60-70% of those positions mentally before outcomes resolve, without that affecting your financial security.
- You have the time and expertise for real due diligence. Angels in the top quartile of diligence hours achieved a 7.1x multiple versus the 2.6x overall average. Diligence is one of the few controllable inputs.
- You can engage with companies regularly. Angels interacting with founders a couple of times per month achieved a 3.7x multiple in four years. Those checking in a couple of times per year averaged 1.3x. Passive angel investing is passive loss absorption.
- Your capital can be illiquid for 5 to 8 years per deal. Not on paper. In practice.
If you are angel investing for non-financial reasons (to support founders, stay close to innovation, build networks), those are legitimate motivations. The NBER study suggests many high-net-worth angels are doing exactly this. Just be clear about what you are buying. It is not reliably a high-return financial instrument for most participants.
Jeff's POV: What I Tell Founders Who Want Angel Investors
When founders ask me how to think about angel investors, I tell them two things.
The best angels are not the ones writing the biggest checks. They are the ones with domain expertise, meaningful post-investment engagement, and portfolios large enough that they are not desperate for your specific deal to work out. An angel with 8 investments who needs one home run to justify the activity is a different partner than one with 40 investments who can support you without the pressure of over-concentration.
Most angels pitching their "value-add" are also underinformed about their own return profile. They do not track IRR across a properly marked portfolio. They remember the wins and categorize the losses as learning. Evaluate angels on their actual networks and documented track record of helping companies. Not their confidence about the asset class.
The Wiltbank NESTA study and the Hustle Fund / Angel Squad analysis point to the same conclusion: the angels who do well are running a process. Due diligence. Portfolio size. Post-investment engagement. These are controllable inputs. Most angels skip the process and chase the story. That is why most angels lose money.
Frequently Asked Questions
What is the average angel investor return, and why is it misleading?
The Wiltbank & Boeker study found an average of 2.6x over 3.5 years: roughly 27% IRR. That number is pulled up by a small number of extreme outliers. The median deal is a loss. 52% of exits return less than invested capital. The average tells you what top performers did. The median tells you what most investors experienced. For most angels, the median is the relevant number.
How many angel investments do I need before the math works in my favor?
Research consistently points to 20 to 30 investments as the minimum for reasonable statistical diversification. With 10 deals (the typical angel's median), you have only a 28% chance of holding even one 50x-plus outlier. At 25 deals, that rises to 57%. At $25,000 to $50,000 per check, you need $625,000 to $1.25 million deployed before the portfolio size gives you a reasonable shot at the return profile this asset class can produce.
Do angel investors really hold for 7-8 years?
Yes. The Wiltbank 2007 study reported a 3.5-year average hold, but that reflects deals that exited, not the full distribution including companies still alive and illiquid. CB Insights data shows VC-backed M&A exits average five years from first investment; IPO exits average seven. Angels investing pre-VC add time before institutional capital even arrives. A 4.8 to 8 year realistic hold period is what the exit data shows, not a pessimistic assumption.
Is angel investing worth it if most angels lose money?
It depends on why you are doing it. For financial return, it can work. But only with 25-plus deals, rigorous due diligence, active post-investment engagement, and capital you can lock up for nearly a decade. For most accredited investors who cannot meet those conditions, a diversified public equity portfolio delivers better risk-adjusted returns with full liquidity. Angel investing makes financial sense as a deliberate, high-volume strategy. It rarely makes sense as a casual activity driven by enthusiasm for individual deals.
Related Reading on AIN
- How to Evaluate Angel Investment Deals: A Due Diligence Framework
- Accredited Investor Requirements 2026: What Has Changed and What It Means for You
- Power Law Returns in Venture Capital: Why the Math Forces Portfolio Thinking
- Angel Investing vs. Venture Capital Funds: Which Is Right for Your Situation
Author Disclosure: Jeff Barnes, MBA has no personal position in any company, fund, or platform named in this article. Angel Investors Network has no current commercial relationship with any party mentioned. AIN provides marketing and education services, not investment advice. Past performance does not guarantee future results. All investments involve risk, including loss of principal.
Looking for investors?
Browse our directory of 750+ angel investor groups, VCs, and accelerators across the United States.
About the Author
Jeff Barnes, MBA