Realised and unrealised
A portfolio can be up a great deal and have made no profit at all. It can be down and have realised substantial gains along the way.
Realised profit exists only where a sale happened. Everything else is unrealised: the gap between a current price and the cost of something still held. Realised profit is settled. Unrealised profit depends entirely on a price you cannot control.
A portfolio view that shows one combined figure has taken a position on whether unrealised gains count as performance. Both positions are defensible. Mixing them without saying so is not.
Which cost is matched to a sale
When you sell part of a position, something has to be decided: which part of your original cost is being released. There are two common approaches.
Average cost
One average price per asset, recomputed on every purchase. Simple, and it is what most portfolio displays use because a single figure is easy to show.
Its weakness is precision. Average cost spreads one acquisition across many, so a sale on the wrong side of a price move produces a different number from the lot-based treatment used for tax reporting. The two are not reconcilable without going back to individual lots.
Lot based
Each acquisition stays separate, and a sale matches specific lots — usually oldest first. More accurate, and considerably more machinery.
Why percentage returns mislead
A percentage return is a ratio, and ratios hide their inputs. Three portfolios can report the same percentage over the same period and be entirely different situations.
- Narrow range. A large gain on a small balance is a huge percentage and a small absolute number.
- Wide range. The reverse case. The percentage looks modest; the money is not.
- Deposits and withdrawals. Money moving in inflates gains, money moving out deflates losses. Neither is performance.
That third point is the one that matters most in practice, and it is the subject of the next section.
Time-weighted and money-weighted
| Measure | What it asks | What it answers |
|---|---|---|
| Time-weighted | How did the portfolio grow per unit of time, ignoring cash flows? | How well the holdings performed |
| Money-weighted | How did my own money grow, given when it arrived and left? | How well I did personally |
They answer different questions and will disagree whenever money moved during the period. A large deposit made just before a rise looks spectacular on a money-weighted basis and unremarkable on a time-weighted one, because only one of them credits the timing.
Neither is more correct. A performance figure without naming which one it uses cannot be compared with anything, including another figure from the same screen.
Currency matters more than people expect
Every crypto portfolio number is two numbers multiplied together: a token quantity and a price. The price is quoted in some currency, and if that currency is not yours, a third rate is involved.
Over a short period this barely matters. Over a period where your local currency moved significantly against the quote currency, a large part of your reported gain may be currency movement rather than anything to do with the asset.
What P&L does not tell you
- How much risk you took to get there
- How concentrated you are in what is driving the number
- How much of the return came from timing versus from holding
- Whether the position could be realised at the displayed price
- How the result compares with simply having held a broad index
These are not small omissions. A portfolio that has doubled while becoming far more concentrated has not unambiguously improved, and P&L on its own will say that it has.
Next: consolidating across exchanges and wallets