I Retested the Full Moon Two Years Later. The Edge Faded.
Back in 2024, I did a study I still think about a lot. I asked whether the full moon actually makes the Nasdaq crazy, pulled five years of data, and found the honest answer: it doesn't. If anything, full-moon days were quieter than normal.
The Strawberry Moon just passed on June 29, so it felt like the right time to do something most trading writers never do: go back and check my own homework. Same study, same method — but now with seven years of data instead of five. About 500 more trading days.
Here's what two more years did to the numbers. This is the real lesson.
The full moon: −10% became −5%
In 2024, full-moon days ran about 10% below the typical day. That looked like a real, if modest, effect.
Two years later, with more data, that gap shrank to about −5%. Still quieter than normal — the direction held — but the size of the effect halved.
| Phase | 2024 (5 yrs) | Now (7 yrs) |
|---|---|---|
| Baseline range | 231.5 pts / $463 | 264.0 pts / $528 |
| Full moon | −10% ($416) | −5% ($502) |
| New moon | −2% | −1% |
| First quarter | +3% | +4% |
| Last quarter | +16% ($538) | +10% ($582) |
Look at that full row, and then look at the last-quarter row.
In 2024, the last-quarter moon looked like the widest phase by a mile — +16% over baseline. I flagged it back then as something to watch, not trade, because one bump across five years can be noise. Good instinct: two years later it cooled to +10%, drifting back toward the pack.
This is the whole point
Here's what I want you to take from this, and it's bigger than the moon.
Early edges fade. When you find a pattern in a few years of data, some of what you're seeing is real and some of it is just the randomness of a smaller sample. As more data comes in, the noise averages out — and the "edge" you were excited about often shrinks toward ordinary. That's not the market playing tricks on you. That's statistics doing exactly what it does.
The full moon effect halving and the last-quarter bump cooling are textbook examples. If I'd taken the 2024 numbers, declared "trade small on full moons, size up on the last quarter," and never looked again, I'd be running a strategy built partly on noise that has since evaporated.
Instead I did the boring, unglamorous thing: I checked again. And I'll check again in two more years.
What this means for you
Two practical takeaways:
On the moon specifically: it still doesn't move the Nasdaq in any way you can trade. The full moon is quiet, not wild — that part's held up across seven years now. The wilder-moon folklore is just folklore.
On edges in general — this is the real gift: be suspicious of any edge built on a small sample, including your own. The fix isn't to stop looking in interesting places. The fix is to keep re-checking as your data grows, and to hold your findings loosely until they've survived more of it. An edge that's still there after seven years is worth more than one that dazzled you for one.
I test the mystic stuff because I'm genuinely curious about it. But curiosity without re-checking is just belief with extra steps. The re-checking is the job.
Sharpen your edge by telling yourself the truth — and then telling yourself again later, when you have more of it.
— 23HT
How this was done: MNQ daily bars from the 2019 contract launch, tagged by moon phase from astronomical data, compared by median daily range against the all-days baseline. 2024 study: ~1,300 days through mid-2024. This study: ~1,800 days through July 2026. Same method, more data.