A return looks like a fact. It's a choice.
Hand the same book, over the same months, to a manager who wants to impress you, and they can produce a dozen different "returns" - every one of them technically true. Gross or net. Annualised or actual. From inception, or from the convenient low. Your money-weighted experience, or the book's time-weighted performance. Pick the right combination and a mediocre year becomes a great headline without a single false number in it.
The belief this piece is here to break: that a reported return is an objective measurement. It isn't. It's the output of half a dozen decisions, each of which can quietly flatter. The only honest thing a manager can do is tell you exactly which decisions they make - and which flattering ones they refuse. So here are ours.
We report one number: NAV per unit, net
The headline is not a composite, not a model, not a "representative account." It's the change in NAV per unit over the period, after all fees. One number, drawn straight from the same reconciled price we strike everything else against. If the unit went from 1.00 to 1.10, the period return was 10%. There is no second, friendlier number kept for the cover.
Time-weighted, because the cash flows aren't the skill
There are two honest ways to measure a return, and they answer different questions.
A money-weighted return blends in the size and timing of cash flows - it's your personal outcome, flattered or punished by when you happened to deposit. A time-weighted return strips that out: it measures what the book actually did, per dollar, independent of when money arrived. They are not interchangeable, and quoting one while implying the other is a classic sleight of hand.
We report the book's performance time-weighted, because that's the only version that reflects the manager's work rather than the luck of deposit timing. And it falls out for free: because deposits and redemptions move cash and units together, the change in NAV per unit already excludes cash-flow timing by construction. Your own statement also shows your money-weighted return - what you earned given your timing - clearly labelled as yours, not the book's. We never blur the two into one flattering average.
Every window, not the flattering one
The most common trick needs no false numbers at all: just start the clock on a good day.
So we don't get to choose the start date. Returns are shown over standard, fixed windows - month, quarter, year, since inception - and the worst stretch is left in, not smoothed away. "Since the low" is not a window we will ever quote, because measuring from the bottom of a drawdown is how you turn a recovery into a fake triumph. If a period was bad, it stays on the page at full size.
We don't annualise a short record
Take three strong months, multiply up, and you can advertise a huge annual figure for a book that has barely existed. It is the most seductive number in early-stage investing and the emptiest.
We won't print it. A short record is reported as exactly what it is - a short record, over the actual elapsed time, with no extrapolation. "Would have" is the most expensive phrase in asset management, and "annualised from a quarter" is its close cousin. When the record is long enough to annualise honestly, we will. Not before.
Net of everything, gross of nothing
The number we publish is after the performance fee - the figure you actually keep, not the one before the haircut. A gross return is a menu price; you live on the net one, the same gap that separates quoted yield from kept alpha. We don't lead with gross and bury net in a footnote. The headline is the kept number.
Here is what all of this looks like in one picture - the same book, the same year, reported the legitimate-looking ways:
What you can check
In the spirit of how you can verify us: the return isn't a figure you have to take on faith, because it's derived from a number you can already see. Take the NAV per unit at the start of any period and at the end; the percentage change between them is the time-weighted return, net of fees, full stop. No adjustment we could make is hidden inside it. If our published return and that arithmetic disagree, the gap is a question you're entitled to ask.
A return is only a fact once you know which choices were made to produce it. Tell me the choices, and the number becomes evidence. Hide them, and it's just a headline.
The one line to keep
There is no flattering version of our return kept somewhere for the cover - the number we show you is the number you can recompute.
Anyone can report a good number. The honest skill is reporting the same number in a bad year that you'd report in a good one.