Momentum

What it measures¶

Momentum is the tendency of prices that have risen over the past year to keep doing better, on average, for some months — one of the most studied patterns in finance. The tile asks two questions, its votes:

• M1 — was the share's return over the twelve months to one month ago positive? The latest month is skipped because very recent moves tend to partly reverse.
• M2 — is the latest close within 25% of the 52-week high — the highest price of the past year (at least 75% of it)?

Beside the votes, the tile shows the RSI(14) — a gauge of how stretched recent moves are, from 0 to 100 — as a reading only. It never votes: a share near its high often reads above 70, and at this horizon that is not a weakness.

Why it is on the page¶

Two separate lines of research support the two votes: the sign of the past year's return predicted the next month's direction across many markets (time-series momentum), and nearness to the 52-week high predicted the following six to twelve months better than past returns did. They are counted as two votes because they are two different findings.

Thresholds, their source and the population they were fitted on¶

• Sources: Moskowitz, Ooi and Pedersen (2012), "Time series momentum", for M1; George and Hwang (2004), "The 52-week high and momentum investing", for the idea of M2; Mark Minervini's trend template for its 25% line; J. Welles Wilder (1978) for RSI's 30 and 70.
• Fitted on: M1 — 58 futures contracts, 1985–2009; M2 — US shares, with the effect found in ten of twenty other markets. The 25% line is a practitioner's, not fitted.
• Evidence grade: peer-reviewed and replicated (the effects); practitioner convention (the 25% line, RSI's zones).

Published patterns weaken once known: a study of many of them found returns about a third lower after publication, and momentum has suffered sharp reversals after market crashes.

A worked example¶

Honeywell International, on bars to 23 October 2026: down 2.9% over the twelve months to a month earlier (M1 failed); 15.0% below its 52-week high, inside the 25% line (M2 passed): 1 of 2 votes passed · 50% · Mixed –, with RSI 66.

Apple, the same day: up 29.4% (M1 passed) and 1.1% below its high (M2 passed): 2 of 2 · 100% · Strong ✓✓, RSI 64.

What “No verdict” means here¶

Momentum needs 252 daily bars — about a year of trading. With fewer, the tile reads No verdict — fewer than 252 daily bars and its votes leave the strip's count; the rest of the strip still shows. When the newest daily bar is older than the previous weekday, the tile keeps its verdict and is marked stale.

What the verdict is and is not¶

The verdict describes the past year's price path against two published lines. It is not a forecast — momentum fails, sometimes sharply, especially after market falls — and it is not an instruction to trade the shares.