12-minus-1 momentum: finance's most-documented factor, and how we test it on NSE
The deepest-supported factor in the literature, a peculiar name, and one construction detail that trips everyone up.
If you had to pick one factor with the deepest academic support behind it, it wouldn't be RSI or MACD — it would be cross-sectional momentum: the tendency for stocks that have outperformed over the past several months to keep outperforming over the next few. It has been documented across decades, markets and asset classes. The standard way to measure it has a peculiar name — 12-minus-1 momentum — and the "minus 1" is the interesting part.
We've added 12-1 momentum to StockLearn as a measured candidate, computed on our own NSE data. This piece explains what it is, why the last month gets thrown away, and the one construction detail that trips people up.
Why skip the most recent month?
12-1 momentum takes a stock's return over the past twelve months but excludes the most recent one — it measures month 12 back to month 1, skipping the final weeks. That sounds odd until you connect it to something we found in our own data: over very short horizons, recent strength tends to reverse. The last month is contaminated by short-term mean reversion that runs opposite to the longer momentum effect. Skipping it separates the durable twelve-month trend from the noisy one-month wobble — two effects that point in different directions and would otherwise cancel.
The construction detail that matters
There's a subtle right and wrong way to combine the numbers. The tempting shortcut is to take the one-year return and subtract the one-month return. That's an approximation, and for the exact stocks momentum cares about — the big movers — it's meaningfully off. The correct construction is geometric: compound the twelve months and divide out the last month, rather than subtracting percentages.
A stock is up 100% over the year and 20% in the last month. The lazy subtraction says its 12-1 momentum is +80%. The correct geometric calculation says +66.7%. For quiet stocks the two nearly agree; for the strong movers momentum is built to find, they diverge — which is why we use the geometric version from day one. (Illustrative.)
How we handle it
We compute geometric 12-1 momentum from our own price history and register it as a measured-only candidate — tracked and measured, not yet part of any verdict. Fixing the formula before we start accumulating measurement data matters: changing a candidate's definition midway would fragment the record and reset the clock. It needs roughly sixty trading days of clean data before it can prove itself, around September 2026.
Momentum and relative strength are cousins
12-1 momentum and relative strength versus the Nifty are closely related ways of asking "has this stock been strong, in a way that tends to persist?" We're measuring both, the same disciplined way, and if either earns its place it becomes a candidate to eventually replace weaker ingredients in the scoring — on evidence, not on reputation.
Key takeaways
- Cross-sectional momentum is the most-documented factor in finance.
- 12-1 momentum uses the past year's return but skips the most recent month.
- The last month is skipped because short-term reversal runs opposite to longer momentum.
- Use the geometric construction, not "one-year minus one-month" — they diverge for big movers.
- We track it as a measured-only candidate until the data earns it a place (~Sept 2026).
See these ideas on real stocks
StockLearn runs this read on ~2,000 NSE stocks every evening. Nifty 50 is free, no login.
Browse today's scan →This guide is educational and explains how StockLearn interprets common technical indicators, using illustrative examples. It is not investment advice or a recommendation to buy or sell any security.