Does Valentine's Day Move the Market? (A Small, Fun Look)
Not every study is a serious edge. Some are just fun questions worth a look — and part of telling yourself the truth is being honest about which is which. This is a fun one. Small sample, loose conclusions, worn lightly.
The question: on Valentine's Day, do traders show up in love with the market — buying — or do they sell it and go home early to their dinner reservations?
I pulled every February 14th the Micro Nasdaq has traded (skipping the years it fell on a weekend). Here's the honest catch up front: that's five days. Five. You cannot build a strategy on five days and I'm not going to pretend otherwise. But the direction is cute enough to share.
The numbers (all five of them)
Against a baseline day of 240.6 points ($481):
- Valentine's Day range: 217.8 pts ($436) — about 10% below normal. A little quieter than average. Maybe everyone's distracted.
- Direction: 40% trend-up, 0% trend-down. Not one of the five Valentine's Days closed as a down-trend day.
So the tiny, tongue-in-cheek read: the market seems to feel the love more than the heartbreak. No down-trend days, a couple of clean up days, and a slightly quieter session overall — like the tape took a softer, friendlier tone.
The honest part
Five days is nothing. Two of those days were wildly different from each other — one ran just 71 points (2020, dead quiet), another ran 363 (2022, a big mover). That spread across five days tells you how little we can conclude. The "0% down-trend" stat sounds strong but with n=5 it could flip with a single bad year.
So I file this under charming, not tradeable. It's the kind of thing I'll re-check every February and watch build. If ten years from now Valentine's Day is still never a down-trend day, that starts to be interesting. For now, it's a Valentine to the data, nothing more.
Buy the love if you want. Just don't size it like an edge.
— 23HT
How this was done: every February 14th traded by MNQ from 2019 through Feb 2025 (weekends excluded), compared by median daily range and day-type against the all-days baseline. n=5 — a fun look, not a finding.