Most angel investments lose money. A few return ten times or more, and those few decide whether your portfolio works. This free angel investment simulator runs 10,000 possible portfolios using your check size and number of deals, so you can see how often a portfolio like yours loses money and how much rides on a single winner. No signup, and we don’t capture any data from your entries.
Set your portfolio to simulate 10,000 possible outcomes.
How this is simulated
Each simulated portfolio makes equal-sized investments. Every deal's result is drawn at random from an outcome table: most deals return nothing or less than the money in, a few return 1 to 10 times, and a rare few return 10 to 300 times. The tool runs 10,000 portfolios and shows the spread. The angel-group table is calibrated to Wiltbank and Boeker, "Returns to Angel Investors in Groups" (Kauffman Foundation, 2007; 1,137 exits): 52% returned less than the money invested, and the 7% that returned more than 10× produced about 75% of all dollars returned. A UK study by the same lead author (NESTA, 2009; 406 exits) found the same pattern: 56% failed to return capital, while 9% returned more than 10× and produced nearly 80% of the positive cash flows. Both are voluntary surveys of angel-group members, so they may over-represent winners. The conservative table is an SVJ stress test with more losses than those studies report (about 65% of deals below 1×, under 4% at 10× or more); it is not drawn from a published study. Angel returns data is thin: investments are private and results are self-reported.
Simplifications: results are gross multiples, before fees, taxes, and dilution from later rounds. Follow-on investments are not modeled. Deals are independent (in reality, outcomes rise and fall together with the market). Every deal exits in the same year, so the annual return is an approximation. Outcome tables are SVJ editorial calibrations, reviewed October 2026. Educational estimates, not investment advice.
How to use the simulator
- Set how many startups you plan to back over the next few years, and your typical check size.
- Pick the outcome data. Start with conservative, a stress test with more losses than the published studies report. The angel-group setting is calibrated to surveys of angel-group members, which may over-count winners.
- Read the headline first: the share of simulated portfolios that lost money. Then change the number of investments and watch it move.
- Press Run again. The sample portfolio changes, but the percentages barely move. No single result is predictable. The odds are.
What the results mean
The power law
Startup returns aren’t spread evenly. In the most-cited angel study (Wiltbank and Boeker, 2007), 52% of exits returned less than the money invested, while the 7% that returned more than 10x produced 75% of all the dollars returned. The sample portfolio in the simulator shows the same pattern: a row of zeros and one deal doing most of the work.
Typical versus average
The Typical tile is the median portfolio. It sits below the average because rare, huge wins pull the average up. Most simulated portfolios finish below the average, so plan around the median.
Why the number of deals matters
On the conservative data, a 5-deal portfolio loses money about 46% of the time. At 20 deals that drops to about 22%, and at 50 deals to about 7%. More deals won’t make any single deal better. They raise the odds that one of your checks lands in the rare company that returns the whole portfolio.
Gross, not net
Results are before fees, taxes, and dilution from later funding rounds. If you invest through a syndicate or fund, its fees come out of these numbers.
Where the numbers come from
The simulator draws each deal’s result at random from an outcome table, adds up the deals in a portfolio, and repeats that 10,000 times. The two tables below are the whole model.
A caution first: angel returns data is thin. Startup investments are private, and the best studies rely on investors reporting their own results. No large, recent, public dataset of angel deal outcomes exists. What the studies below have going for them is that they agree on the shape: more than half of deals lose money, and a small share of big winners produces most of the returns.
Sources
- Robert Wiltbank and Warren Boeker, Returns to Angel Investors in Groups (2007), funded by the Ewing Marion Kauffman Foundation and the Angel Capital Education Foundation. A survey of 539 US angel-group investors covering 1,137 exits. The angel-group table is calibrated to it.
- Ramon DeGennaro and Gerald Dwyer, Expected Returns to Stock Investments by Angel Investors in Groups (Federal Reserve Bank of Atlanta working paper, 2010; later published in European Financial Management). Reanalyzes the same data and finds the evidence that it overstates returns “tenuous at best.”
- Robert Wiltbank, Siding with the Angels (NESTA, 2009). 158 UK angel investors, 406 exits: 56% failed to return capital, while the 9% that returned more than 10x produced nearly 80% of the positive cash flows. Overall 2.2x over just under four years.
- Chris Dixon, Performance Data and the “Babe Ruth Effect” in Venture Capital (Andreessen Horowitz, 2015). Horsley Bridge data on venture funds since 1985: about 6% of investments generated about 60% of total returns. Venture funds, not angels, but the same pattern.
Conservative table (stress test)
This table is an SVJ stress test, not a published study. It assumes more losses than the angel studies report, for anyone who suspects self-reported surveys are too rosy.
| One deal returns | Chance |
|---|---|
| Nothing (0x) | 50% |
| 0.1x to 1x | 15% |
| 1x to 3x | 21% |
| 3x to 10x | 10.4% |
| 10x to 30x | 2.5% |
| 30x to 100x | 0.8% |
| 100x to 300x | 0.3% |
Angel-group table
| One deal returns | Chance |
|---|---|
| Nothing (0x) | 35% |
| 0.1x to 1x | 17% |
| 1x to 2x | 36% |
| 2x to 5x | 4% |
| 5x to 10x | 1% |
| 10x to 30x | 5.8% |
| 30x to 100x | 1.0% |
| 100x to 300x | 0.2% |
Within each range, results are spread evenly on a log scale, which tilts them toward the low end: a deal in the 10x to 30x range is as likely to land between 10x and 17x as between 17x and 30x. The annual return shown is the portfolio multiple compounded over your years to exit: multiple1/years minus 1.
Frequently asked questions
Is angel investing profitable?
It can be, but the average hides a lot. The US angel-group study found an average return of 2.6x over about three and a half years, and a UK study found 2.2x over just under four years. In both, more than half of exits returned less than the money invested. Both surveyed angel-group members, so they may lean toward winners. A small portfolio can easily miss the winners entirely.
How many startups should an angel invest in?
Many experienced angels aim for 15 to 30 positions over time. In this simulator, going from 10 to 20 deals cuts the chance of losing money from about a third to about 22% on conservative data. If you can’t afford that many checks, write smaller ones rather than fewer.
Why do my results change when I press Run again?
Each run is a new random draw of 10,000 portfolios. The sample portfolio changes a lot; the percentages move by a point or two. Use Copy shareable link to save an exact run.
How accurate is this simulator?
It’s a teaching model. Deals are independent, every deal exits in the same year, and follow-on investments aren’t modeled. Real outcomes rise and fall together with the market, which makes real portfolios riskier than this. Use it to set expectations. It isn’t investment advice.
Want to model one deal in detail? Use the angel investment return calculator. And see all our free startup tools.