How does SPY react to macro events?
Algorithmic detection of 8 high-leverage event types — first Fed cut, first Fed hike, yield-curve inversions/steepenings, VIX spikes/normalizations, credit blow-outs/recoveries — built directly from FRED time series with no manual calendar required. For every historical occurrence, the SPY forward-return distribution is computed at 1, 5, 21, 63, and 126 trading-day horizons.
How to read
Each card is one event class. currently_active = the event has triggered in the past few days; cards in green are live signals. Forward returns are SPY mean / median / hit-rate measured from the day of trigger going forward N trading days.
If multiple events are simultaneously active, the system computes a composite expected forward return across active themes.
Limitations: events compute on FRED's public daily series — trigger detection has up to ~1 trading day of lag. Sample sizes per event are small (8-25 occurrences across 35 years) so confidence intervals around forward-return summaries are wide. This is a directional tool, not a precision instrument.
Currently active events
Events that have triggered recently and are still considered "active" by the detector.
8 algorithmically-detected event classes
Sorted by recency of last trigger. Click a card for occurrence list (forthcoming).
Past forward-return distributions do not predict future returns. Sample sizes per event class are inherently small (often single-digit occurrences over decades). This page surfaces statistical patterns, not trading signals. Always pair with regime context (regime page), positioning data, and risk constraints before acting.