Analyst upgrades across 2,717 names
Analyst upgrades were measured across 2,717 names over the last three years; 50% of them have a positive median reaction at 30 minutes. An analyst raises their rating on a stock. The market's response is the label — not the upgrade itself.
Analyst upgrades do not move one way.
Direction is not automatic: over the last three years NVDA upgrades faded (median −0.67% at 30 minutes, 27% positive) while AAPL and AMD popped. The realized move is the signal, not the call.
Across all 2,717 measured names, 50% have a positive median 30-minute reaction and 116 have a median of exactly 0.00%. No single median is published for the class: with the names split that evenly, the middle of that distribution is zero, and it is zero on all six of these pages.
The strongest and weakest reactions to analyst upgrades, among the 111 best-measured names.
Strongest median 30-min reaction
Weakest median 30-min reaction
Last 3 years, pulled 2026-09-21. Reaction = percent move vs. the price at the news timestamp. Counts and medians only. Both tables rank only the 111 names with at least 50 measured analyst upgrades; the full population publishes every name with at least 10, where a handful of events can put a name at either extreme.
Email me when a name reacts to analyst upgrades.
One email per qualifying move — the headline, the measured reaction, and how it compares to that name's baseline. Nothing else.
All six event types, side by side →
Other event types: Analyst downgrades → · Coverage initiations → · Rating reiterations → · Price-target raises → · Price-target cuts → · Every measured ticker →