← All tested claims

Should You Avoid the Daily Rollover?

True — and it produces the biggest numbers you should least trust.

The claim as it is usually statedThe 21:00–22:00 UTC rollover is thin liquidity to be avoided.

The 21:00–22:00 UTC rollover shows the largest directional effects anywhere on our board — EUR/GBP closes higher 59.6% against 48.2% elsewhere. It is also the thinnest liquidity of the weekday: EUR/USD moves 59% of its normal size there. A strong-looking edge you cannot actually collect, which is why this site flags rollover buckets rather than promoting them.

What the data shows

Across 19 markets: direction: 12 of 19 markets differ beyond chance · movement: quieter in only 11 of 19. Median movement in the window, 83% of a normal hour.

MarketIn the windowRestzMovementSamples
EUR/GBP59.6%48.2%14.0119% of normal4,049
EUR/USD53.3%49.7%6.859% of normal9,861
GBP/JPY54.3%50.7%5.887% of normal6,949
USD/CAD52.2%49.4%5.482% of normal9,893
AUD/USD47.5%50.2%-5.0103% of normal9,551
EUR/JPY53.0%50.5%4.583% of normal8,741

Strongest six by statistical strength, whichever way they point. z measures how far from chance a result sits — we treat |z| ≥ 3 as significant, and print it even when it is nowhere near.

Who says it

Daily Price Action

How this was measured

Every claim here reduces to a window on the clock or the calendar, so it can simply be checked. We compare the claimed window against the rest of the week on 23 years of hourly data across 19 markets, measuring two separate things the folklore tends to conflate: direction (does the window close higher more often?) via a two-proportion z-test, and movement (is the average move bigger or smaller?). A window can be perfectly ordinary in direction and genuinely quieter in movement — saying which is the entire point.

The bar here is |z| ≥ 2, deliberately more lenient than the |z| ≥ 3 this site applies to its own numbers: when testing someone else's claim, failing to reject it even at a lenient bar is the stronger statement.

Check any hour yourself →

Free on the four major pairs, with the full 23 years and the sample size on every number.