Methodology

Glossary

The same definitions the (i) dots carry, in one place.

Invested return

How the money you actually invested performed, ignoring uninvested cash. A different metric from Total return, not the same one measured differently — deposit 10,000 and buy 21 of an ETF, and a 10% gain on that ETF is a 10% invested return but only a 0.02% total return. Same profit, different denominator.

Total return

How every euro you committed performed, uninvested cash included. This is the canonical figure for the portfolio and the one that answers “what happened to my money”. Cash earns nothing on its own, so holding a lot of it usually pulls this below your Invested return; the gap is your cash effect.

Cash effect

Total return minus Invested return, in percentage points — everything holding cash did to your return, not only the dilution the name once implied. Breaks down into three parts: dilution (uninvested money earning nothing), interest (what that cash actually paid), and currency (revaluation of any foreign balance). The three sum to this figure exactly, and it is not always negative — interest or a favourable currency move can outweigh the dilution.

Total value

Everything this scope is worth today at market prices, uninvested cash included. A position no price could be found for is carried at what you last paid for it and counted here, flagged rather than dropped.

Market value

What your open positions are worth today at market prices — uninvested cash is NOT included here. It shows up in Overview's Total value instead, which sums positions and cash together.

Uninvested cash

Cash sitting in this scope, not yet put into a position. Already counted in Overview's Total value, alongside your positions — this card states the same money as its own fact, not a second holding.

Net invested

Every deposit and transfer in, less every withdrawal, running from the first one you ever made. It moves only on the days money crossed the boundary, never with prices — the gap between it and Total value is your profit.

Fees paid

Commissions and charges actually booked in this scope's ledger, not an estimate. An absent fee is an absent fee, not a zero one: if your broker export carried no fee column, nothing appears here.

Tax paid

Tax actually withheld in this scope's ledger. Kept separate from fees because you can act on one and not the other, and never summed with an estimate of tax you might still owe.

Weight

A position's share of this scope's gross value today, cash included. It moves with prices as well as with trades, so a position you have not touched still drifts.

Average invested weight

The share of your capital that was actually exposed to the market over the window, averaged across it. 100% means fully invested; 20% means four fifths of the money sat in cash.

Net return

Profit and loss over the selected window, in currency — market movement plus income, after fees actually charged. Not a ratio: the percentage beside it divides by invested capital.

Contribution to return

This holding's share of the return above, in percentage points. Its weight at the previous close times that day's return, summed across the window and compounded, so every row plus Trading and income adds up to the total exactly. Weight times return will not reproduce it — those two are window averages and this is a daily sum. Splits into Local and Currency alongside it.

Local

The share of this holding's contribution earned in its OWN trading currency — what it would have contributed had its currency never moved against yours. Zero currency effect for a holding already in your own currency.

Currency

The share of this holding's contribution that came from its own currency moving against yours, not from the holding itself. Local plus Currency equals the Total exactly — it is what Local leaves over, not a separately measured figure.

Average weight, held

The mean size of the position across the days you held it, not across the whole window. A holding opened last week is a 40% position that existed for five days, and averaging its absence in would report 2%. Anything opened or closed inside the window therefore leaves this column summing to more than 100%.

Own return

What the holding itself did over the days you were exposed to it, compounded and independent of how much of it you held. The day you bought it is excluded: the move that happened before you owned it was not yours.

Trading and income

The part of the return no holding's weight times return explains: trades placed during a day, income paid out to cash, and any position that could not be priced. It gets its own row so every holding above stays exactly right.

Realised on sales

The part of this window's market movement that selling actually banked — what you sold for, against what those shares cost. Gross of fees, which the statement subtracts once on their own line below.

Still unrealised

The part of this window's market movement still riding on positions you hold. It moves with prices until you sell, and selling converts it into Realised without changing the total above.

Transfers and corporate actions

Shares that moved at a price no market set: a custody transfer, a split, a redemption. Real changes in what you hold, with no trade and no market move behind them, so they sit apart from the two lines above. Usually zero, and hidden when it is.

Local price movement

The part of Market movement your holdings earned in their OWN trading currencies — what this window's price change would have been had none of your foreign holdings' currencies moved against yours.

Currency effect

The part of Market movement that came from a foreign holding's currency moving against yours, not from the holding itself. Local price movement plus Currency effect equals Market movement exactly — it is what Local leaves over, not a separately measured figure. Zero on an account with nothing held outside your own currency.

Drift

How far a position's current weight sits from its target, in percentage points. Only defined once a strategy has been set — with no target there is nothing to drift from.

Risk contribution

Each position's share of total portfolio volatility, accounting for how it actually moves with everything else you hold — not just its own size. Uses the same covariance matrix as the efficient frontier and correlation grid, so a highly-correlated pair contributes more together than their weights alone would suggest.

Fee ratio

Fees paid per unit transacted — a cost-per-trade measure. Undefined when nothing has been traded, which is different from zero.

Cost ratio p.a.

Total fees and tax as an annualised share of average portfolio value — an expense-ratio-style figure for what ownership drags per year. Averaged only over the span the portfolio held material capital, so a tiny seed balance early on doesn't inflate it.

Turnover p.a.

Buys plus sells as an annualised share of average portfolio value — how much trading activity happened relative to what was actually held. Averaged only over the span the portfolio held material capital, the same rule Cost ratio p.a. uses, so a tiny seed balance early on doesn't inflate it.

Time-Weighted Return

Return with the effect of cash flows removed — what the money would have earned held constant, regardless of when you added or withdrew.

Money-Weighted Return

Return that accounts for the size and timing of your cash flows — what you actually earned. A balance correction moves this rate without changing the interest already booked, so the two figures don't have to tie.

XIRR

The internal rate of return on irregular cash flows — the money-weighted return, annualised.

CAGR

Compound Annual Growth Rate — the geometric (compounded) growth rate, what a single lump sum would have needed to grow at to reach the same total return. Usually a little below the arithmetic "Annualised return" when returns vary year to year.

Average return (annualised)

The mean daily return scaled to a year — the arithmetic counterpart to CAGR above. CAGR compounds; this averages, so it sits a little higher whenever returns vary from day to day, and the gap between the two is itself a rough measure of how volatile the ride was.

Best day

The single largest one-day gain over the period.

Worst day

The single largest one-day loss over the period.

Month to date

Compounded return since the start of the current calendar month, regardless of the lookback window selected above.

Quarter to date

Compounded return since the start of the current calendar quarter, regardless of the lookback window selected above.

Year to date

Compounded return since the start of the current calendar year, regardless of the lookback window selected above.

Sharpe ratio

Return earned per unit of volatility, net of the risk-free rate. Higher is better risk-adjusted performance.

Sortino ratio

Like Sharpe, but penalises only downside volatility — upside swings don't count against it.

Calmar ratio

Annualised return divided by maximum drawdown — return earned per unit of worst-case loss.

Volatility (annualised)

The standard deviation of daily returns, scaled to a yearly figure.

Downside deviation

Like volatility, but counts only returns below your risk-free target — upside swings don't count against it. The same figure the Sortino ratio divides by.

Value at Risk (95%)

The loss not expected to be exceeded on 95% of days, based on the recent return distribution.

Conditional VaR (95%)

The average loss on the worst 5% of days — what VaR doesn't tell you about the tail.

Skew

The asymmetry of the return distribution. Positive means a long tail of rare, large gains outweighing more frequent small losses; negative is the reverse — frequent small gains punctuated by rare, sharp drops. Zero is symmetric.

Kurtosis

How fat the tails of the return distribution are, relative to a normal one. Higher means extreme days — in either direction — happen more often than a bell curve would predict.

Beta

Sensitivity to the benchmark's moves. 1.0 tracks it exactly; below 1.0 moves less than the market.

Alpha

Return earned beyond what beta and the benchmark's own move would predict.

Correlation

How closely your daily moves track the benchmark's, from -1 (opposite) to +1 (identical). Distinct from beta: correlation is about how closely two series move together, beta is about how much.

Maximum drawdown

The largest peak-to-trough fall in your time-weighted return over the period. It measures performance, not balance: deposits and withdrawals are removed before it is computed, so paying money in cannot end a drawdown. Your portfolio can be worth more than ever while this one is still running.

Average drawdown

The mean depth ACROSS EVERY drawdown episode in the period, each one weighted equally regardless of how long it ran — a two-week dip and a two-year slump each count once. Different from "Max drawdown" above, which is the single deepest episode.

Average drawdown duration

The mean duration across every drawdown episode, the same equally-weighted average "Average drawdown" uses for depth. Different from "Max drawdown duration" and "Longest time underwater" below, which each describe one specific episode rather than an average of all of them.

Max drawdown duration

How long the DEEPEST drawdown lasted, in calendar days from the peak that preceded it to the day that peak was regained — or to today, if it has not been. Depth and duration are different kinds of pain: a 30% fall recovered in a month and one still running after three years are not the same experience. Coincides with "Longest time underwater" below when the deepest episode also happens to be the longest one (common at account scope, where one dominant drawdown tends to swamp the rest) — but not in general, so both stay on the table.

Max drawdown period

The peak and the date the deepest drawdown recovered to it — or "ongoing" if it hasn't. The days count above is the same episode's length; this is when it happened.

Longest time underwater

The most days EVER spent below a previous high, whichever episode that was — often a different, shallower one from the deepest drawdown, not the same one measured twice. Shown alongside "Max drawdown duration" rather than instead of it, because a portfolio can have a sharp fall it recovered from quickly and a separate, mild slump it sat in for years.

Real return

Return adjusted for inflation (CPI), in your account's region — what your money actually gained in purchasing power.

Net of fees

Return after fees and trading costs actually paid, not an estimate.

Cost basis

What you paid for a position, tracked by average cost across every buy and sell.

Win rate

The share of days with a positive return — a day-by-day measure, not a per-trade one.

Average up-day

The mean return on days that were positive, excluding every flat or negative day.

Average down-day

The mean return on days that were negative, excluding every flat or positive day.

Payoff ratio

Average up-day divided by the size of the average down-day — how much a typical good day gains relative to what a typical bad day costs. Above 1.0 means winners are bigger than losers on average, independent of how often each occurs (that's win rate's job).

Profit factor

Total gains from every positive day divided by total losses from every negative day. Above 1.0 means gains outweighed losses over the period; unlike the payoff ratio, this also accounts for HOW OFTEN each occurred, not just their average size.

Risk-free rate

The annual rate used as the "do nothing" baseline throughout this page — Sharpe and Sortino measure return in excess of it, and it sets downside deviation's hurdle. A rate you set in Settings, not a live market rate.

Observations

The number of return observations — trading days for a business-calendar portfolio, every calendar day for one holding crypto — behind every statistic on this page. A small count is worth noticing: a Sharpe ratio from three weeks of data carries far less information than one from three years.