Angel Investing Portfolio Strategy: The Power Law, the 30-Check Threshold, and How to Size Your Checks

Angel investing returns follow a power law: a small number of outliers drive nearly all the gains, while most individual checks return less than the money put in. The Kauffman Foundation study of 3,097 investments found that 52% of exits returned less than 1x, while 7% of exits returned more than 10x and produced 75% of all dollars returned (Wiltbank and Boeker, ACA/Kauffman). To capture that distribution, you need roughly 20 to 30 investments, with 30 or more approaching the point where aggregate returns turn reliably positive. The Angel Capital Association caps angel allocation at 5 to 10% of net worth. The math is simple to state: total angel allocation divided by about 30 checks gives your per-check size, deployed across 3 to 5 vintage years to spread market-timing risk. Cheryl Kellond, CEO of Play Money, frames the whole problem through three variables you control: check count, position size, and vintage spread.

This is educational content, not investment advice. Angel investing carries the risk of total loss of capital. Past return data does not guarantee future results.

The power law: why angel returns are not normally distributed

Angel and early-stage venture returns do not cluster around an average. They follow a power law. A June 2020 AngelList analysis by Abe Othman of the platform's own fund data put the shape parameter at roughly 2.3. A power law is a distribution where a few extreme outcomes account for most of the total, and a lower shape parameter means a heavier right tail: more probability sitting out at the extreme high end. Public equity returns sit closer to 3.0. The lower the number, the more your result depends on catching one of the rare giants.

Here is what that tail looked like in the AngelList data:

• 1% of positive AngelList investments returned 22x or more.
• The single best-performing investment returned over 100x.
• The top 10% of investments generated 85 to 90% of all cash proceeds (Angel Capital Association).
• Tech Coast Angels: 8 companies, just 3% of 247 tracked outcomes, produced 77% of all dollars returned (ACA, same source).
• 500 Startups' 10-year, 2,500-investment dataset: about 50% of startups returned 0x, 40% returned 1 to 3x, 8% returned around 20x, and 2% returned 50 to 100x (Play Money: Diversification, Returns, and How to Pick).

The same AngelList data dismantles the high-conviction, low-volume strategy for most angels. A manager making just 10 investments was outperformed by the broader AngelList portfolio 74% of the time, and the most common outcome from a 10-deal portfolio was a result slightly above zero, well below market returns. Othman's conclusion for pre-seed and seed investors: an indexing approach, spreading capital across many credible deals, beat roughly 75% of early-stage venture funds. For an individual angel, that reframes the central question. The skill is not picking the one winner. The skill is buying enough credible shots that the math has a chance to work for you.

"Back in 2020 AngelList published a blockbuster insight on early-stage investing. The conventional wisdom was (and still is): Make a small number of highly concentrated bets and double down on the ones that get hot. 'Spray and Pray' is often used with derision to describe investors who take the opposite approach. But when AngelList analyzed decades of data, they determined something surprising. The best strategy for pre-seed and seed investors is often the opposite: Broadly index the market by investing in many credible deals."

From Play Money's weekly letter: Diversification, Returns, and How to Pick.

The 30-check threshold: where diversification starts to pay

The AngelList analysis points to about 30 investments as the point where diversification benefits become statistically meaningful for an early-stage angel portfolio. Below that, you are too exposed to the chance that your handful of bets simply misses the tail. The probability work from Hustle Fund's Angel Squad makes the stakes concrete. If the outlier rate, the chance any single check returns 5x or more, is about 5%, then your odds of landing at least one outlier climb steeply with portfolio size:

• 10 deals: 40% chance of catching at least one 5x+ outlier.
• 20 deals: 64%.
• 30 deals: 79%.
• 50 deals: 92%.

The ACA 2017 American Angel Report found a related threshold from the other direction: angels who built 12 or more investments over 5 or more years had a 75% probability of a 2.6x return. The same study found the median active US angel held just 7 investments, with a mean of 11.4. Most angels are sitting well below any data-supported minimum.

The return gap between an underdiversified portfolio and a diversified one is large, and it is almost entirely a story about outlier capture. Portfolios of fewer than 10 deals returned roughly 0.8x in ACA-cited historical research, a net loss. Portfolios of 15 or more deals returned roughly 2.6x. The difference is not that the larger portfolios picked better companies. It is that they bought enough tickets to be holding one of the rare winners when the power law paid out. At 10 deals, there is a 60% chance the outlier never shows up in your portfolio at all. At 30 deals, that miss probability falls to 21%.

A note on what those historical multiples mean today. The 2.6x figure comes from a hotter exit environment than the one angels face now. The ACA 2025 Angel Funders Report found that across 2024 non-shutdown exits, the median multiple on invested capital was 1.3x, and 25% of exits returned less than the original capital (Figure 71). The direction of the diversification effect is unchanged: more credible deals, better expected outcomes. But treat 2.6x as a historical ceiling for an experienced, well-diversified angel, and 1.3x as the more sober current median to calibrate your own expectations against. One more data point worth holding: ACA groups averaged 16 deals per year in 2024, which makes the 30-deal threshold about two years of steady activity, not a lifetime project.

Position sizing: from net worth to check size

Once you accept that you need 20 to 30 checks, the next question is how big each one should be. Start at the top, with how much of your total net worth belongs in the angel asset class at all. The practitioner sources cluster in a tight band. The Angel Capital Association puts it at 5 to 8% of net worth. Seraf Investor, from Christopher Mirabile, says 5 to 10%. Rockies Venture Club, from Peter Adams, says under 10%. The ACA's best-practices work notes that half of angels commit no more than 15%.

Where you land in that band depends on how active you are. Passive angels who rely on SPVs and syndicates can tolerate the higher end, because portfolio construction is handled systematically. Active angels running full due diligence on every deal should stay at the lower end, because time is the binding constraint and spreading yourself thin across 30 hand-checked deals is hard.

The follow-on reserve almost nobody mentions

Here is the step generic "how many deals" content skips. Seraf Investor recommends reserving $1 for every $1 you commit to initial checks, a 50% follow-on reserve. That single rule doubles your real capital requirement. A 20-deal portfolio at $10K per initial check is not a $200K commitment. It is $400K once you reserve for follow-ons. Leave that out and your position sizing looks twice as affordable as it actually is, which is exactly the mistake that leaves angels unable to support their winners later.

Right-sizing your angel portfolio by net worth, including the follow-on reserve:

• $1M net worth: about $100K to angel investing at a 10% allocation. 10 initial checks (the minimum), $5K to $10K each, $50K held in reserve, $100K total.
• $2M: about $200K allocation. 15 to 20 checks, $5K to $10K each, $100K reserve, $200K total.
• $5M: about $500K allocation. 20 to 25 checks, $10K to $25K each, $250K reserve, $500K total.
• $10M+: $1M or more allocation. 30 to 40+ checks, $25K+ each, $500K+ reserve, $1M+ total.

These figures use a 10% allocation as a working benchmark; the ACA range is 5 to 8% and Seraf recommends 5 to 10%. Use the lower end if you do full due diligence on every deal. Follow-on reserve follows Seraf's $1:$1 rule. Neither this table nor this pillar is investment advice.

Cheryl Kellond, CEO of Play Money, has written roughly $1,500 checks across more than 50 investments, a deliberately small-check approach that buys portfolio breadth over concentration in any one deal. It is a credibility marker, not a template: the right check size is the one that lets you reach 20 to 30 deals at your own allocation. For context, the ACA 2025 Angel Funders Report put the average ACA-group deal at $239K in 2024 (median $110K), but those are pooled group commitments, not the solo checks an individual angel writes.

Stage diversification and vintage-year risk

Spreading across the number of deals is one axis. Spreading across stage and across time are the other two.

Early versus later stage

Early-stage deals (pre-seed and seed) fail more often, with 60 to 70% of companies returning 0x per Hustle Fund's Angel Squad, but the power-law tail is richest there, and your capital and connections matter most when a company has the least of both. Later-stage deals (Series A and B) fail at roughly half the seed rate and sit closer to liquidity, but higher entry valuations compress your multiples, and a company broadly soliciting angels late can raise the question of why institutional money is not leading.

"Healthy diversification across stages, and across different return profiles, strengthens a portfolio. It expands your learning. It increases exposure to different exit timelines. And occasionally, it brings upside sooner."

From Play Money's letter: Early vs. Later-Stage Angel Investing.

Vintage-year risk

Deploy all your capital in a single calendar year and you concentrate vintage-year risk. A portfolio built entirely in a peak-valuation year like 2021 has a very different expected outcome than one built across 2019 to 2023. Seraf Investor recommends pacing 4 to 6 new deals a year over a 3 to 5 year horizon to average across cycles. Hustle Fund frames it as dollar-cost averaging into the private market: same logic as staggering your entry into an index fund, applied to startup entry points.

Sector as a third axis

Sector concentration is a quieter risk. AI seeds are currently priced at $50M+ post-money on revenue structures that take a forensic accountant to decode. Adjacent sectors run cheaper with established exit markets: consumer brands reaching $100M in revenue on less capital than SaaS, women's health with the fastest median seed-to-exit of any sector, and climate hardware where seed deals got cheaper in 2024 even as acquirers grew more active.

"AI seeds are priced at $50M+ post-money based on revenue gymnastics you'd need a forensic accountant to untangle. On the flip side, these 'out of vogue sectors' have low entry points, highly liquid M&A markets, and they solve very real problems. 'This makes life better' is as load-bearing an investment thesis as 'this makes money'. That's an edge institutional capital can never have."

From Play Money's letter: Smart Angels Invest Where VCs Won't.

Follow-on strategy: two valid stances

Follow-on investing, putting more money into a company's next round, is the most underdiscussed variable in portfolio construction. There is no single right answer, but the two dominant approaches are each internally consistent. Pick one on purpose.

Stance A: reserve for the winners

Reserve $1 for every $1 of initial checks, then deploy that reserve only into your top performers, the companies where the power-law tail is starting to show. This mirrors how institutional venture firms behave: concentrate follow-on dollars on the deals flashing early product-market-fit signals. Seraf Investor calls it the $1:$1 rule.

Stance B: breadth first, no follow-on

If any next deal could be the outlier, then every follow-on dollar into an existing winner competes with discovering a new one. Play Money's letter calls this the "Infinite Regret" school: if every investment could be the next Uber, your best move may be to keep funding new companies rather than doubling down, maximizing shots on goal. (Play Money #15)

Which fits you depends on signal and capital. The reserve model suits angels with 2+ years of investments and visible winners, plus enough capital to do both. The breadth-first model suits early angels building their first 20 to 30 deals, where portfolio signal is thin and access to new deals is the real constraint.

"Without a diverse pool of investors, founders miss out on capital. And angels miss out on potential returns."

From Play Money's letter: Portfolio Diversity and Returns.

The long-run baseline: what decades of data show

The most-cited longitudinal study on US angel returns is Rob Wiltbank's 2007 Angel Investor Performance Project, which tracked 3,097 investments from 538 organized angel-group members, with 1,137 realized exits. The result: a 2.6x gross return and a 27% gross IRR over an average 3.5-year hold. A 2016 Angel Resource Institute follow-up landed in the same neighborhood: 2.5x and 22% gross IRR over 4.5 years.

The current picture is cooler. The ACA 2025 Angel Funders Report covers 2024 exits from 69 reporting ACA groups: median MOIC of 1.3x, average MOIC of 3.5x (pulled up by a single 22x outlier), and 25% of exits returning less than 1x. M&A is still the dominant exit at 85% of transactions, but the 30x+ outcomes that flattered 2022 averages are absent.

A few honest caveats on the Wiltbank baseline, because the headline number gets quoted out of context. These were organized angel-group members, a sample skewed toward experienced, high-volume investors. The figures are gross, before fees, taxes, and opportunity cost. And the data predates the AngelList, Republic, and Wefunder era, so deal access has changed materially. Read 2.6x as a ceiling for a well-diversified, experienced angel, not as the expected value for a newcomer with 7 deals.

The power-law analysis and the long-run return data arrive at the same place from opposite directions. The angels hitting 2.6x and above are getting there through diversification and deal volume, not through superior stock-picking at low volume. The strategy is the diversification.

If you are still building toward your first handful of checks, the mechanics of getting started come before the portfolio math: our 5-day beginner guide to angel investing walks through accreditation, deal access, and writing a first check, and Play Money's letter on diversification and how angels pick winners covers the selection side of the problem this pillar only touches.

Written by Cheryl Kellond, CEO of Play Money. Serial founder, MIT Sloan MBA, active angel investor. This is educational content, not investment advice. Angel investing carries the risk of total loss of capital, and past return data does not guarantee future results. Consult a qualified financial advisor before making investment decisions. Last updated: June 2026.