Portfolio construction
How many investments
Why the answer is larger than most new angels expect, and how to think about the number without pretending to a precision the maths does not support.
The question every new angel asks after their second investment is how many they should make. The honest answer has two parts: more than you think, and the exact number matters less than being on the right side of the threshold.
The reasoning comes straight from the power law. If the portfolio's return depends on holding at least one outlier, then the portfolio must contain enough positions that holding one is likely rather than lucky. A portfolio of five is a bet that one of five is exceptional. A portfolio of thirty is a much safer bet on the same underlying rate.
What follows is arithmetic about probability, not a prediction about any individual company. It cannot tell you that thirty investments will work. It can tell you that five is a materially different proposition from thirty, and why.
The arithmetic of holding an outlier
Suppose, purely for illustration, that a given investment has a one in twenty chance of returning more than twenty times your money. The probability that a portfolio of N investments contains no such outcome is 0.95 to the power of N.
At five investments that is about 77% — three times in four you hold nothing exceptional. At ten it is about 60%. At twenty, about 36%. At forty, about 13%. The curve is steep in exactly the range where angels actually operate, which is why the difference between ten and thirty investments is so much larger than it appears.
Change the assumed rate and the numbers move, but the shape does not. Whatever the true underlying probability, a small portfolio is dominated by the chance of holding nothing exceptional, and that risk falls quickly as the portfolio grows.
| Investments | Chance of no outlier | Chance of at least one |
|---|---|---|
| 5 | about 77% | about 23% |
| 10 | about 60% | about 40% |
| 20 | about 36% | about 64% |
| 30 | about 21% | about 79% |
| 40 | about 13% | about 87% |
Illustrative, assuming a one-in-twenty chance per investment of an outsized outcome. The assumed rate is a modelling choice, not an observed figure — change it and the levels move while the shape stays the same.
What constrains the number
Three things bound how many investments you can sensibly make: capital, time and access.
Capital is the obvious one, and it is the reason cheque size and portfolio size are the same decision. An angel with £200,000 to deploy over five years is choosing between twenty cheques of £10,000 and five of £40,000, and the power law says the first is the better structure.
Time is underrated. Each investment carries real ongoing work — updates to read, follow-on decisions, consents to sign, occasionally a founder who needs help. Thirty positions is a genuine commitment, and the answer for most people is to be less involved per company rather than to hold fewer.
Access is the binding constraint for many angels. Making thirty good investments requires seeing several hundred opportunities, and that flow does not appear on its own. Syndicates, angel groups and platforms exist largely to solve this, at the cost of fees and carry.
Building the portfolio over time
The number should be reached over years, not months. Deploying a whole allocation in one year concentrates you in a single vintage, and vintages differ enormously — the same company raises at very different prices depending on when it happens to need money.
A common structure is to plan a fixed number of new investments per year over four or five years, holding back capital for follow-ons throughout. That produces both portfolio size and vintage diversification without requiring any market timing.
It also has a practical benefit: your judgement improves. The investments made in year four are usually better than those made in year one, and pacing means more of the capital is deployed by the better version of you.
In short
What to take away
- Plan for twenty or more positions. Five is a bet on being lucky rather than on being right.
- Portfolio size and cheque size are one decision — divide the allocation, do not raise the cheque.
- Spread deployment over four or five years to diversify vintage as well as company.
- If access is your constraint, that is what syndicates are for — price the carry against not being in the deal at all.
From the decoder