Do full and new moons move Bitcoin? I tested 226 of themBitcoin / TetherUSBINANCE:BTCUSDTWenMoonLambo"Buy the new moon, sell the full moon" is one of crypto's most repeated pieces of sky lore. It's also easy to test, because a full or new moon comes around about every 14.8 days. On the chart, white circles mark full moons and blue circles mark new moons. The test: • Bitcoin daily closes, Aug 2017 to Oct 2026. • 113 full moons and 113 new moons, timed to the minute. • For each one, buy at that day's close and hold for a week, then compare with buying on any random day. Results (average 7-day return): • After a full moon: +0.21% • After a new moon: +1.35% • After any day: +1.08% That looks like a story: new moons beat full moons by more than a percentage point. It isn't one. Random sets of days differ by that much all the time. Across every holding period tested, the lowest p-value was 0.22, nowhere near significant. The small edge in the half-cycle version disappeared when the dates were shifted by a single day, which is what noise does. Why this matters even if you don't follow astrology: the same trap catches plenty of indicator back-tests. A pattern that's easy to see in a chart, isn't significant, and vanishes when you nudge the inputs isn't an edge. If you like tracking moon phases, use them as a journaling lens and note how you felt and traded around them. Don't use them as a signal. Educational content, not financial advice.