A monthly crisis window can hide the crash
Crisis performance measured on monthly returns can post a gain while the asset fell 38% peak-to-trough inside the very same window. Report intra-window daily drawdown to measure a crash faithfully.
A +11.6% “crash”
Define the COVID crash as the months 2020-02 through 2020-03. Ranked by monthly total return over that window, Nvidia is the best crash performer at +11.6%. The number is arithmetically correct and completely misleading: the monthly window opens on February 1, so it captures the late-February melt-up that preceded the crash. The two monthly bars net to a gain.
What actually happened
On daily NAV inside the same window, Nvidia fell roughly −38% peak-to-trough. That is the crash. The monthly aggregate smeared the pre-crash rally, the collapse, and the start of the recovery into one positive number, and the drawdown disappeared.
The fix
Measure crises two ways and report both: monthly total return over the window (for cross-asset comparability) and intra-window daily peak-to-trough drawdown (for the experience that actually mattered). When the two disagree, the daily drawdown is the faithful measure of the crash; the monthly figure is a window artifact.
The general point
Aggregation frequency is a modeling choice, not a neutral detail. A window boundary one week early or late, or a switch from daily to monthly, can flip a crash into a gain. So the frequency and the exact window boundaries are part of the claim — state them, and cross-check a monthly headline against the daily path before trusting it.