By Tim du Toit, investing since 1987 and author of Quantitative Value Investing in Europe: What Works for Achieving Alpha. Last reviewed October 2026.
What is 12-1 month momentum? It is a stock's price return over the past 12 months, leaving out the most recent month. It still works over the long term. In our own 10-year test it was the best of four momentum measures, but it lost badly in two sharp market rebounds. Use it in calm, rising markets, combine it with value, and use less of it just after a crash.
Key findings
- 12-1 momentum shows up in up to 159 years of data and in more than 4,000 versions of the strategy (SSRN 5561720).
- In our test from January 2016 to January 2026, Magic Formula stocks ranked by 12-1 momentum returned 7.6% a year. The same stocks ranked by 12-month momentum returned 4.6% a year, and the Magic Formula alone returned -0.5% a year.
- All four momentum versions trailed the equal-weighted average stock in the same universe (9.9% a year) and had deeper losses (-41.9% against -24.7%).
- 12-1 momentum beat that average in 7 of 10 years. It lost most in 2021 and 2023, the years after sharp market rebounds.
- It works best in calm, rising markets: 2.32% a month against 0.94% a month in stress periods, in a 1975 to 2024 study (Dikhit, 2026).
This article is general research, not personalised investment advice. Past performance is no guarantee of future results, and you can lose money. Read our full disclaimer.
How the screener calculates 12-1 month momentum
In the screener this ratio is called Price Index 12m Minus 1m. It divides the share price one month ago by the share price twelve months ago.
So on 1 October 2026 it uses the price on 1 September 2026 divided by the price on 1 October 2025. A value above 1 means the price went up over those 11 months. A value below 1 means it went down.
You can use it as a slider (for example 0% to 10% gives you the top 10% of companies) or as an output column to sort, filter and export.
Why momentum works, and why you skip the last month
Investors react too slowly to news. When a company's earnings change, people stay anchored to their old view, and the price catches up over several months (Jegadeesh and Titman, 1993).
Investors also sell winners too early and hold losers too long. That slows price adjustment too, so trends continue.
The skip month exists because the average stock that jumps in one month tends to fall back the next. Investors overreact to short-term news, then correct.
But that reversal is not true for every stock. The most heavily traded stocks keep moving in the same direction the next month, in 22 developed countries (Medhat and Schmeling). So the skip month matters most for small, thinly traded companies.
Our test: 12-1 momentum against the alternatives, 2016 to 2026
I wanted to see which momentum measure does best on top of a value strategy, using our own historical screener data.
The rules: take the top 20% of companies worldwide by Magic Formula rank, with a market value above €100 million and excluding financial companies, real estate and utilities. Buy the top 25 by each momentum measure, in equal amounts. Hold 12 months, then rebalance. Returns include dividends.
Ten-year results, January 2016 to January 2026
| Magic Formula top 25, ranked by | Return a year | Total return | Largest fall | Worst year |
| 12-1 month momentum | 7.6% | 108.6% | -41.9% | -30.9% |
| Adjusted Slope 125/250 days | 5.2% | 66.2% | -41.1% | -19.6% |
| 12-month momentum (includes last month) | 4.6% | 57.4% | -44.6% | -27.1% |
| 6-month momentum | 3.1% | 35.4% | -41.5% | -19.2% |
| No momentum (Magic Formula rank only) | -0.5% | -4.4% | -55.2% | -24.3% |
| Average stock in the same universe | 9.9% | 156.3% | -24.7% | -13.9% |
Source: Quant Investing historical screener database, as at January 2026. Equal-weighted, returns include dividends, single-stock returns capped at +200%, delisted stocks valued at their last price. "Average stock" is the equal-weighted return of the 2,500 largest companies in the same countries and sectors, not a stock market index. Before dealing costs and tax.
Three things stand out.
First, momentum helped a lot. Without it, the cheapest Magic Formula stocks lost money over ten years. Adding 12-1 momentum turned -4.4% into +108.6%.
Second, skipping the last month helped in this test. 12-1 momentum returned 3% a year more than 12-month momentum. That is the opposite of the industry-level result in Dikhit's 1975 to 2024 study, where including the last month did slightly better. So the evidence on the skip month is mixed.
Third, and this is the uncomfortable part: none of the momentum versions beat the average stock in the universe, and all of them fell much further in bad years. A 25-stock value portfolio is a bumpy ride.
12-1 momentum year by year against the average stock
| Year held | 12-1 momentum | Average stock | Difference |
| 2016 | 16.0% | 11.2% | +4.8% |
| 2017 | 29.0% | 21.9% | +7.2% |
| 2018 | -30.9% | -13.4% | -17.5% |
| 2019 | 30.4% | 26.3% | +4.1% |
| 2020 | 34.1% | 17.6% | +16.6% |
| 2021 | -14.8% | 13.9% | -28.6% |
| 2022 | -1.9% | -13.9% | +12.0% |
| 2023 | -5.2% | 15.7% | -20.9% |
| 2024 | 15.2% | 10.5% | +4.8% |
| 2025 | 26.4% | 17.5% | +8.9% |
Each year runs from the first trading day of January to the first trading day of the next January. Same source and rules as the table above.
Look at the three losing years. 2018 was the year volatility came back and markets sold off hard in the last quarter. 2021 and 2023 both followed sharp rebounds, after the COVID crash and after the 2022 bear market.
See the 12-1 momentum winners from a year ago
Your free demo has Price Index 12m Minus 1m as a ready-made ranking factor across 22,000+ stocks. Add the Magic Formula filter to copy the test above, then sort 12-1 momentum from high to low. The demo runs on data from a year ago, so the names at the top are the stocks this rule would have picked twelve months ago.
No credit card needed. Cancels automatically after 30 days.
When 12-1 momentum works best, and when to avoid it
Dikhit split every month from 1975 to 2024 into four market types, using last month's return and volatility (SSRN 6009657, summarised by Larry Swedroe).
Average monthly return of 12-1 momentum by market type
| Market type | What it looks like | 12-1 result | What to do |
| Stable trend | Above-average returns, low volatility | Best: 2.32% a month | Use it fully |
| Overheated | Strong returns, high volatility | Medium | Use it, cut the most extreme winners |
| Slow breakdown | Weak returns, low volatility | Medium | 12-month momentum is more stable here |
| Stress or crash | Weak returns, high volatility | Worst: 0.94% a month (12-month: 1.35%) | Use less of it |
Source: Dikhit (2026), long-only portfolios of the 5 best of 48 industries, 1975 to 2024, no trading costs.
Volatility mattered more than direction. Calm months with weak returns still beat volatile months with strong returns.
The real danger is the first rebound after a fall. In 2009 a portfolio that bought past winners and sold short past losers lost 55% while the market rose 36% (Verdad). The losers jumped the most. Momentum crashes come after market declines, when volatility is high, as the market rebounds (Daniel and Moskowitz).
Verdad found one more warning light: when the extra interest that risky companies pay over government bonds rises above 7%, momentum has done poorly.
If you only buy stocks, and never sell short, your risk in a rebound is smaller. You mostly lag the market for a while. Our 2021 and 2023 results show what that lag looks like.
The alternatives compared
12-1 vs 12-month momentum. 12-month momentum includes the last month. In our test it returned 4.6% a year against 7.6% for 12-1. In Dikhit's industry study it did slightly better and gave steadier returns. My take: keep the skip month for small companies, where the one-month reversal is strongest.
12-1 vs 6-month momentum. 6-month momentum reacts faster but returned only 3.1% a year in our test. Its worst year (-19.2%) was much milder than 12-1 (-30.9%).
12-1 vs Adjusted Slope. Adjusted Slope rewards smooth trends and punishes jumpy ones. It returned 5.2% a year, with a worst year of -19.6%. If big losing years bother you more than return, it is the calmer choice.
Removing the most extreme winners. Between 2000 and 2015 a standard 12-month momentum strategy made only 0.72% a year. After removing the top and bottom 3% of stocks by strength, it made 8.76% a year (Yang and Zhang). Most of the gain came from avoiding crashes.
Smooth trends over spikes. The crash risk sits in stocks trading close to their 12-month high. The part of momentum measured from that high back to today's price gave positive returns without crashes (Büsing, Mohrschladt and Siedhoff).
How I would use 12-1 momentum
These are general rules, not advice for your portfolio. You are of course free to decide for yourself.
- Combine it with value. Never use momentum alone. In our test it turned a losing value screen into a winning one.
- Check the market type once a month. Is the market above its 200-day average? Was last month calm or volatile? After a deep fall with high volatility, use less momentum for the next few months.
- Cut the most extreme winners. The top 3% by price strength reverse most often.
- Keep the skip month for small companies. For large, heavily traded companies, 12-month momentum is good enough.
- Use a stop loss. We use a 20% trailing stop loss in our own strategies. It limits the damage when a trend reverses.
Ask an AI assistant to run it for you
If you have a Best Deal subscription, you can connect Claude or ChatGPT to the screener and ask in plain English. For example:
"Show me the top 20% of stocks worldwide by Magic Formula rank, market value above €100 million, sorted by Price Index 12m -1m, top 25."
"Back-test my saved screen from 1 January 2016, top 25 names, held 12 months, sorted by Price Index 12m -1m."
Setup takes about 15 minutes: how to connect Claude and ChatGPT to Quant Investing. More example questions are in the Quant Investing MCP command list (MCP stands for Model Context Protocol, the standard that lets AI assistants use outside tools).
Forward-test 12-1 momentum yourself in five minutes
Open the demo, sort by Price Index 12m Minus 1m, and write down the top ten names. Those are the stocks the rule would have picked twelve months ago. Look up today's prices on any five of them and you have run a real one-year forward test, with no back-test software needed.
Forward-test 12-1 momentum now
No credit card needed. Cancels automatically after 30 days.
Already convinced? Skip the demo and subscribe direct. 100% money-back guarantee for the first 30 days. See subscription options →
Frequently asked questions
Does 12-1 month momentum still work?
Yes, over the long term. It shows up in up to 159 years of data. In our test from 2016 to 2026 it lifted a Magic Formula portfolio from -0.5% to 7.6% a year, but it still trailed the average stock (9.9% a year).
Why skip the most recent month?
Because the average stock that jumps in one month tends to fall back the next. This reversal is strongest in thinly traded stocks. The most heavily traded stocks tend to keep moving in the same direction.
When does 12-1 momentum fail?
Mostly in sharp rebounds after a market fall. In our test it lost 14.8% in 2021 while the average stock gained 13.9%, and lost 5.2% in 2023 while the average stock gained 15.7%.
Is 12-1 momentum better than 6-month momentum or Adjusted Slope?
In our test it gave the highest return: 7.6% a year against 5.2% for Adjusted Slope and 3.1% for 6-month momentum. But its worst year was deeper: -30.9% against about -19% for the other two.
Where can I find 12-1 momentum in the screener?
It is called Price Index 12m Minus 1m. Use it as a slider or as an output column. Best Deal subscribers can also ask Claude for it directly.
Sources
- Dikhit, D. (2026). The Informational Role of the Most Recent Month in Industry-Level Momentum Strategies. SSRN 6009657.
- Swedroe, L. (2026). The Skip-Month Mystery. Alpha Architect.
- Jegadeesh, N. and Titman, S. (1993). Returns to Buying Winners and Selling Losers. Journal of Finance.
- Daniel, K. and Moskowitz, T. (2016). Momentum Crashes. Journal of Financial Economics.
- Medhat, M. and Schmeling, M. Short-term Momentum. SSRN 3150525.
- Büsing, P., Mohrschladt, H. and Siedhoff, S. (2021). Decomposing Momentum: Eliminating its Crash Component. University of Münster.
- Momentum Factor Investing: Evidence and Evolution (2025). SSRN 5561720.
- Verdad (2024). A Momentum Crash Course.
- Yang and Zhang, summarised in Enhancing the Performance of Momentum Strategies, Alpha Architect.
This article is general research, not personalised investment advice. Back-test results are hypothetical, before dealing costs and tax. Past performance is no guarantee of future results, and you can lose money. Read our full disclaimer.
