The economics in this subject describe a portfolio of many small bets, not a single concentrated one - and somebody who can only make two or three checks is not really practising this asset class in any statistical sense, whatever it feels like while they are doing it.
9 min read · Level 3
I think the least honest thing said about this asset class is that anyone with a bit of spare money and a good eye for a founder can meaningfully participate in it, and the least honest thing believed about it, by investors who should know better, is that a handful of careful decisions amounts to the same activity as the twenty-position programs the published statistics describe. Both understate how much of this asset class's outcome is decided by sample size before a single dollar is deployed. A reader with genuine conviction and real surplus capital who chooses to make one or two concentrated bets, with eyes open about the odds, has made a defensible decision - speculation is a legitimate activity when it is named honestly and sized deliberately. What I would object to is the same decision dressed in this subject's portfolio language, because the two are not the same claim. One says "I am making a bet I understand to be high-variance and could lose entirely"; the other says "I am practising an asset class whose published economics I expect to benefit from", and only a program resembling the full sample size in this subject's own arithmetic actually supports that second sentence. The honest version of lesson six, for a great many readers with real interest and modest capital, is that they have been served correctly by discovering this is not, at their current scale, the right vehicle for what they are looking for - and that the right next step is either building toward a real program over time, or reading the venture funds subject that follows this one.
Lesson six in every asset-class subject in this curriculum asks what job a class does that the reader's other holdings do not, and this is the subject where the honest answer is inseparable from a question about sample size rather than about the asset itself.
The portfolio arithmetic in lesson two was built on twenty positions, and 70.42% of its total return came from a single one of them. That is not incidental colour - it is the load-bearing fact of this entire asset class. The economics this subject describes are the economics of making many small, independent bets and letting the distribution do its work across them. They are not the economics of any single bet, and a reader evaluating one company in isolation, however carefully, is evaluating a coin that on its own tells them almost nothing about how the whole class actually performs.
Work out what "many" requires in practice. If a representative cheque size is 25,000 and a genuine attempt at this asset class means something close to the twenty-position portfolio in lesson two, the initial capital commitment alone is 500,000, spread across years rather than deployed at once. Layer on a reserve for the pro-rata follow-ons that lesson two showed can be worth more than the initial decision - a representative reserve equal to half the initial commitment, or 250,000 - and the realistic program size is closer to 750,000, committed over a period likely to span most of a decade before the bulk of it is either lost, returned, or realised.
Set that beside what happens statistically at a much smaller scale. An investor who can genuinely afford three cheques of 25,000 each - 75,000 total - is not sampling the distribution in lesson two; they are drawing three cards from it. Applying that portfolio's own illustrative odds, roughly 55.0% of positions return nothing, and a specific investor's actual result depends enormously on which three of the twenty slots they happened to draw. Somebody with three positions who avoids the one standout entirely - a likely outcome, since only one slot in twenty carried it - has made three individually reasonable decisions and captured a portfolio-level return nowhere close to the 3.55x the full twenty-position set produced, through no error in judgement at any stage.
What that means honestly is this: a reader who cannot fund something resembling the full program - the checks and the reserves - is not really practising the startups-angel asset class in the sense this subject has described it. They are making a small number of individual venture bets, which is a real and legitimate thing to do, but it is a materially higher-variance and differently-shaped activity than the portfolio-level economics in the rest of this subject, and describing it with this subject's language overstates what a handful of positions can be expected to deliver.
Because the diversification argument that works for public equities does not transfer here in the way it is usually assumed to, and the reason is sample size rather than correlation.
In public markets, a reasonably diversified position of thirty or more holdings genuinely reduces company-specific risk while preserving most of the asset class's expected return, because individual public-company outcomes are not dominated by a single extreme tail the way this class's outcomes are. Here, the distribution itself is the point: the return is not smoothly distributed across positions waiting to be diversified away from an unlucky one, it is concentrated in a small number of extreme outcomes that most portfolios, by construction, will simply miss if they are too small to plausibly include one.
That has a direct and uncomfortable consequence for eligibility and access, which belongs in this lesson rather than being glossed over. Most jurisdictions restrict this kind of investing to accredited or otherwise qualified investors specifically because the combination of total-loss risk, illiquidity, and the sample-size problem described above makes it a poor fit for money that has not been assessed as genuinely surplus to the investor's other financial needs. Stating the eligibility gate honestly - rather than treating accreditation as a hurdle to clear on the way to a good idea - is the more useful framing: the gate exists because the asset class specifically punishes small, undiversified exposure, which is exactly the exposure an investor without substantial surplus capital is the most likely to end up with.
There is also a genuine access asymmetry worth naming plainly. The strongest deal flow in this asset class runs through networks - other founders, other investors, and institutional venture funds who see a very large number of opportunities and select from among them before most individual angels ever hear about a company. An individual investor without that network access is, on average, seeing a set of opportunities that has already been filtered by others' decisions not to invest, which is a real and structural disadvantage that no amount of diligence on any single deal corrects.
None of this is an argument that nobody should hold any of this asset class. It is an argument that the honest unit of analysis is the program, not the position, and that a reader whose realistic capital and time do not support something resembling a real program has been served well by concluding that this asset class, at the scale available to them, is not the right vehicle for the exposure they are looking for.
This is educational information, not investment, securities or financial advice. Rules differ by country and change — get advice from a qualified professional in your jurisdiction before acting.
Educational, not personalised financial advice. Nothing in the Investor Academy is a recommendation to buy, sell or hold anything — what is right for you depends on your own circumstances and the rules where you live.
Sources: U.S. SEC - Investor.gov; FINRA - Investor Insights
Three investors, each with 75,000 available for this asset class, make different decisions about how to use it.
Investor one commits the full 75,000 into a single company they have strong personal conviction about, having read all six checks in lesson four and genuinely liked what they found. If that company fails - the 55.0% base rate from lesson two's illustrative table - the entire 75,000 is a permanent loss, with no other position in the same asset class to offset it. If it becomes the standout 50x outcome, the position is worth 3,750,000: a spectacular result, but one that depended on picking correctly from a set where four in five attempts historically land at 1x or below.
Investor two spreads the 75,000 across three positions of 25,000 each, drawn from the same illustrative distribution as lesson two's twenty-position table. A plausible draw of three from twenty slots where eleven are zeros and only one is the 50x standout gives this investor roughly a 15% chance of including that specific slot by chance alone (1 favourable slot out of 20, times 3 draws, is a rough illustrative approximation rather than a precise combinatorial claim). In the much more likely case that the standout slot is not among the three drawn, this investor's outcome depends entirely on the other, far less extreme outcomes in the table - a real possibility of ending up near or below 1x on the whole 75,000, despite having diversified.
Investor three recognises that 75,000 does not support a real program, and instead allocates that capital toward the venture funds subject that follows this one - buying a professionally sourced, already-diversified slice of many more than twenty companies through a single commitment, at the cost of the fees that subject describes in full.
None of the three investors made an unreasonable decision. What differs is which of them is actually practising the portfolio-level economics this subject describes, and which of them, however careful their diligence, is really making a small number of individually high-variance bets that this subject's arithmetic does not straightforwardly apply to.
Evaluating a single position as though the 3.55x portfolio-level economics in lesson two applied to it individually. They do not - that figure describes what a full, diversified set of twenty positions produced together, dominated by outcomes most individual positions will not achieve.
Treating three or four positions as a diversified portfolio in this asset class the way thirty holdings would be in public equities. The concentration of return in a small number of extreme outcomes means a small sample can easily miss the outcomes that make the whole class work, through no error in selection.
Sizing the initial cheques without reserving for follow-ons, then discovering the reserve requirement only once a position starts working. Lesson two showed a follow-on decision that was worth more than the initial one on a position that had already proven itself - a reserve that does not exist when it is needed forfeits exactly the outcome the whole program was built to capture.
Treating the eligibility gate as an obstacle rather than as information. Most jurisdictions restrict this activity to accredited or similarly qualified investors precisely because the combination of total-loss risk, illiquidity and the sample-size problem is a poor fit for capital that has not been assessed as genuinely surplus.
Assuming access to deal flow is uniform. An individual investor without a strong network is, on average, seeing opportunities that institutional investors with far greater reach have already passed on, which is a real disadvantage that no amount of diligence on any single opportunity corrects.
Representative cheque size, times a number of positions large enough to plausibly include an extreme outcome, plus a reserve of roughly half that amount for follow-ons. On the numbers in this lesson that is close to 750,000 over several years, not the size of any single cheque.
If the honest answer is that you can fund two or three positions rather than something resembling twenty, say so plainly rather than describing the activity with this subject's portfolio-level language.
A professionally sourced, already-diversified fund - the subject of the next part of this curriculum - trades a visible fee for the sample size an individual program of two or three cheques cannot achieve on its own.
The restriction to accredited or similarly qualified investors in most jurisdictions is a direct response to the total-loss rate, the illiquidity, and the sample-size problem this lesson describes - not an arbitrary barrier standing between an investor and a good idea.
Before committing to this asset class at all, when deciding whether a small number of individual positions constitutes a genuine program, and whenever the portfolio-level statistics from earlier lessons are being applied, explicitly or implicitly, to justify a single concentrated position.
Do not use this lesson to conclude that a small number of positions can never be a reasonable decision - a reader with genuine conviction, surplus capital, and an honest acceptance of the odds may reasonably choose to make one or two individual bets. What this lesson refuses is describing that choice using the portfolio-level economics that only apply at a materially larger sample size.
Published figures about this asset class's aggregate economics - a multiple, a headline return, a share of outcomes described as strong - are widely available and widely quoted, and they are frequently invoked by an individual investor to justify a single concentrated position rather than a genuine program. The statistic is not wrong; it is simply describing a different unit of analysis than the one being defended by it. A portfolio-level number computed across twenty or more independent positions says very little about what any one of them, chosen in isolation, is likely to do - and the gap between the two is rarely acknowledged in the moment the number is being used to justify a decision.
Before citing a portfolio-level statistic to justify a decision, ask how many positions it was actually computed across, and whether your own decision resembles that sample size or a single draw from it.
Create a free account to read the rest and apply it to your own investing.