Daily-bar momentum on index futures
Return persistence at the daily horizon reads zero, and most of the index drift accrues overnight.
- Verdict
- Settled by the data
- Decided by
- Drift null
- The number
- Lag-1..10 autocorrelation ≈ 0; 73% of drift is overnight
The hypothesis, as pre-registered
Futures trend at the daily horizon, so a multi-day momentum hold should capture continuation.
What the test found
Daily return autocorrelation at lags one through ten is indistinguishable from zero on every major contract tested. There is no persistence for a momentum rule to harvest.
Worse, the profit a naive long-biased daily system does show is beta, not edge. Roughly three quarters of the index drift accrues overnight, between the close and the next open — so a ten-day long hold banks a substantial amount of pure exposure that a backtest with no benchmark happily reports as strategy performance.
Every long-skewed strategy tested on an up-drifting sample has to clear this bar. Most cannot. A drift null — the same trade count, the same holding period, entered at random — is the minimum honest comparison, and it is positive.
What would make this work
The strategy would have to beat a random-entry control of matched trade count and holding period, and keep its edge with the overnight segment removed.
Every insight ends with this line: the exact condition that would turn the finding into a tradeable edge. That is what makes each entry a next experiment rather than a full stop.
Screened April 18, 2026
Testing something similar?
The screener puts your version of this idea through the same gate stack, on your market, and answers in seconds. Your first verdict is free.