Does "sell in May and go away" work for sneakers? We tested the Halloween indicator

"Sell in May and go away" is one of the oldest lines on Wall Street. In 2002, two economists, Sven Bouman and Ben Jacobsen, put it to a proper test in the American Economic Review. They found that stock markets in most of the countries they studied returned more from November through April than from May through October, and named it the Halloween indicator: buy around October 31, sell in May.

Sneakers now trade like a market, with a bid, an ask and a price history for every pair. So we asked whether resale prices have their own Halloween indicator.

They do not. Over 22 months of StockX sales, the same shoes moved about the same in either half of the year. Winter months averaged a 0.17% fall a month; summer months a 0.19% rise. That gap is well inside what chance produces.

Each half of the year won once

If the Halloween indicator held, every winter half would beat the summer half next to it. Instead, the halves split:

The first summer beat the winter before it by six points. The second winter beat the summer after it by about two. Put all 22 months together and the difference between a typical winter month and a typical summer month is 0.36 percentage points, in the wrong direction for the indicator. A test that shuffles the season labels finds a gap that size or bigger 62% of the time by chance alone.

Over 22 months, a winter month and a summer month moved the same shoes by about the same amount.

December up, January down: twice, but only twice

One thing does repeat. In both years, the same shoes cost more in December than in November (+2.3% in 2024, +1.9% in 2025) and less in January than in December (-3.8% in 2025, -1.3% in 2026). That would fit holiday gifting: our holiday resale calendar found sales of the same shoes roughly doubled from Black Friday to mid-December 2025 and fell below normal in January. That post saw weekly prices hardly move; the monthly figures here suggest a small price lift on top of the volume surge.

Two Decembers are not a pattern you can test, though. Bouman and Jacobsen had decades of stock returns; we have less than two years of sneaker prices. Read the December bump as something to watch, not a rule.

What it means

For buyers and sellers, the half of the year says little about how a shoe's resale price moves. What moves a pair is the pair: how long since it released, whether it restocks, and what happens to its line. Our same shoes a year later piece shows that kind of shoe-specific drift is far larger than any seasonal swing here.

It is also a reminder of how easy it is to see a pattern in a market's history. The Halloween indicator held up in stocks across many countries and many decades. In sneakers, with 22 months of data, the best we can say is that there is no sign of it yet. We will rerun this test each spring as more months come in, alongside our monthly SneakerPulse Resale Index. This is past sales, not a forecast.