Percentile rank
Education only · our voice · free public data
Definition
The share of historical observations (e.g., over a trailing 252 trading days) that fall below the current reading, expressed 0-100. It places today's value within its own historical distribution.
How to read it
A percentile rank answers 'how does today compare to its own history?' A 95th-percentile reading means the current value is higher than 95% of observations in the lookback window; a 5th-percentile reading means it is near the bottom of its range. The '252d' tag uses roughly one trading year of history. Unlike a z-score, percentile rank is distribution-free — it does not assume normality — so it handles skewed, fat-tailed sentiment and volatility data gracefully. It is the preferred way to define 'extreme' for indicators whose raw scale drifts over time.
How practitioners use it
Used as context among multiple indicators — never as a standalone signal to act.
Less common professional uses
Percentile is robust to outliers and skew (a single crash print doesn't blow up the scale the way it inflates a z-score) — a key reason desks favor it for tail-y series. Window sensitivity: a value at the 99th percentile of 252 days may be mid-pack over 5 years; always state the lookback, and cross-check a short window against a long one. It compresses distance information — 90th and 99th percentile can be far apart in raw value; pair with a z-score when magnitude matters, not just rank. In a strong trend, an indicator can print successive new highs so that it sits at the 100th percentile for weeks — 'extreme' by rank without any reversal.
Sources & provenance
Standard percentile-rank statistics (educational overview); Portal desk education notes
This page is educational content published by Pachira Aquatica Global LLC. It is not investment advice and not a recommendation.