Recovery Factor — Return
Log-normalized recovery: ln(1 + total_return) / ln(1 / (1 - max_dd)). Scales linearly regardless of compounding.
- Computed from
- Equity curve
- Scope
- Single report
- Range
- Any real number
- Direction
- Higher is better
The Recovery Factor measures profit against pain: it divides the strategy's total profit over the whole record by its maximum drawdown — the biggest fall from a high point down to the lowest point after it. It answers a blunt question: across the account's whole life, how many times over did it earn back its deepest hole?
Despite the name, Recovery Factor does not measure how long an account takes to recover. It measures how much profit was made relative to the worst drop — not the speed of climbing back out. For recovery time, see time under water.
How it's calculated
The idea is simple: total profit divided by the worst drop.
Recovery Factor = total profit / worst drawdown
- total profit
- cumulative gain over the whole record (NOT annualized — the lifetime figure, the key difference from Calmar)
- worst drawdown
- the largest fall from a peak to the lowest point after it, as a positive magnitude
ln(1+return) / ln(1/(1−dd)) — which stops a very large compound return from dominating the ratio out of proportion to its risk, keeping strategies at different return levels on one footing. On the P&L curve it is the plain dollar ratio.On the Return curve this is your money-weighted lifetime gain measured against the drawdown you personally lived through — deposits and withdrawals affect both. (This basis uses the log-scaled form.)
What it tells you
The bands below are for the plain (P&L) ratio and assume a meaningful sample — at least a year, and ideally a record that has survived a real drawdown. On the log-scaled Return/TWR curves the same records read lower, and on a short record the band means little (see the pitfalls).
| Value | Reading | Notes |
|---|---|---|
| < 1 | Underwater | Total profit hasn't yet matched the worst loss. |
| 1 – 3 | Modest | Earned back the drawdown a few times over. |
| 3 – 5 | Solid | Comfortable profit relative to the worst fall. |
| > 5 | Strong | Large lifetime profit per unit of deepest pain — over a multi-year record. |
Worked example
An account finishes the record up +120% in total, and its deepest drawdown along the way was 20%.
On the P&L curve this is the plain ratio:
Recovery Factor = 120% / 20% = 6.0
— the profit is six times the deepest drawdown. On the Return and TWR curves, where EquityTruth log-scales both terms, the same record reads lower:
ln(1 + 1.20) / ln(1 / (1 − 0.20)) = ln(2.2) / ln(1.25) ≈ 3.5
Same account, two numbers — because the log form compresses large returns. That's why the "6× the drawdown" intuition is exact only on the P&L basis; on the others, treat the figure as a comparable score.
Pitfalls
- Not time-adjusted. A longer record accumulates more total profit, mechanically inflating the Recovery Factor — so it cannot fairly compare a 1-year account with a 5-year one. Use Calmar (annual return ÷ drawdown) when comparing across records.
- A tiny drawdown makes it meaningless, not just large. As the worst drawdown approaches zero the ratio tends to infinity — a record with no drawdown returns an infinite Recovery Factor. So a young or lucky account that has simply never been tested posts a spectacular figure that says nothing about resilience. The metric mechanically flatters the least-tested accounts; treat a very high Recovery Factor on a short record as "not enough drawdown history yet," not as strength.
- One event sets the denominator. A single worst drawdown defines the whole ratio. It says nothing about how often or how long the account suffered — pair it with the Ulcer Index and time under water.
- Silent on recovery time. As the red flag up top warns: it measures profit relative to the deepest fall, never how long the account took to climb back out. A fast and a slow recovery score identically.
Recovery Factor vs Calmar
Both divide return by the same maximum drawdown — the difference is the numerator. Calmar uses annualized return, so it is time-adjusted and comparable across records of any length. Recovery Factor uses total cumulative return, so it is the lifetime version — it tells you how many times over the strategy has recovered its worst loss to date, but inflates with age. In the language of managed futures, Recovery Factor is essentially an un-annualized MAR ratio (MAR = annual return ÷ max drawdown, which is what Calmar measures). Read Recovery Factor for a record's whole history; read Calmar to compare strategies head-to-head.
Related
The companions that fill its blind spots: the Calmar ratio is the per-year, comparable version; max drawdown is the denominator on its own; the Ulcer Index and time under water add the frequency and duration of pain that Recovery Factor ignores.