Analytics and metrics
The definitions behind every number Quanify shows you on a strategy page and on Home. The account-level Analytics tab is not open yet; it arrives with the Trade Journal.
Reading a metric honestly
Two rules apply everywhere Quanify shows a number.
- Thin samples are flagged, not hidden. A four-trade Sharpe is not a Sharpe. Where there is not enough history for a ratio to mean anything, Quanify withholds it and says so rather than printing a confident-looking figure.
- A backtest and a live record are different things. On a strategy page, forward-recorded trades are appended to the historical record with the boundary marked, so you can always see which part is which. Both are hypothetical: simulated fills, not brokerage trades.
The core numbers
| Metric | What it means |
|---|---|
| Net P&L | Total result over the window, after the costs included in the data. |
| Win rate | Winning trades as a share of all trades. On its own it says almost nothing. A 30% win rate can be excellent and an 80% one can be ruinous. |
| Profit factor | Gross profit divided by gross loss. Above 1 is profitable; the interesting question is how it is built. |
| Expectancy | What an average trade returns in currency. Win rate and average win and loss folded into one figure. |
| Average win / average loss | The two sides separately. These are what actually explain a profit factor. |
| Best / worst trade | The extremes. Worth checking whether one outlier is carrying the whole record. |
| Total trades | Sample size. Read every ratio below with this number in mind. |
Drawdown
Max drawdown is the largest peak-to-trough fall in the equity curve over the window. It is the single most useful number for deciding what you can actually run, because it is the loss you have to be able to sit through: not the loss you hope for.
The drawdown view on a strategy chart plots the same thing continuously: how far below the previous peak the curve was at every point, and therefore how long each recovery took. A shallow drawdown that lasted nine months is not a small drawdown.
The ratios
| Ratio | What it measures |
|---|---|
| Sharpe | Return per unit of total volatility. Higher is smoother. It penalises upside volatility just as much as downside, which is its main weakness. |
| Sortino | Return per unit of downside volatility only. Usually the fairer read for a strategy whose good months are large. |
| Calmar | Return against maximum drawdown. The ratio that speaks directly to whether you could have held the position through the worst of it. |
| Recovery factor | Net profit divided by maximum drawdown. How many times over the strategy earned back its worst stretch. |
Sharpe and Sortino here are annualised, so they are comparable across strategies with different trade frequencies. Both need a real span of history to be meaningful, which is why they are withheld on short records.
The numbers on Home
- Unrealized P&L is what your open positions are worth against the live quote right now. Realized P&L is what closed trades have banked in the session. The two are kept separate on purpose: a good realized day sitting on a bad open position is not a good day.
- Account Health and the Strategy Leaderboard share one window: day, week, month or all time.
Where each number comes from
- A strategy page computes from that strategy's own trade history: its backtest, plus its live-recorded signals where it has been running on Quanify. It is the strategy's record, independent of your routing.
- Home computes from your accounts as the broker and the quote feed report them. It describes you.
They will not agree with each other, and they are not supposed to. A strategy's record includes signals that never filled for you; your accounts include trades no strategy ever fired.