What Is Momentum Investing, and Does It Still Work?

The 10% of US stocks with the highest returns over the prior year, leaving out the latest month, made 16.8% a year from January 1927 to August 2026, before costs. The market made 10.3%. The 10% with the lowest returns lost 0.8% a year.

That is momentum investing: holding what has recently done best. This guide defines it, shows the result for every calendar year since 1927 and checks whether the edge survived after 2000.

Momentum investing ranks assets by recent return and holds the winners. It is a trading signal: a fixed rule that turns price data into a hold or avoid instruction.

Trend following is a close relative. It compares one asset with its own past, where momentum compares assets with each other.

Narasimhan Jegadeesh and Sheridan Titman wrote on the idea in 1993, in Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency (Journal of Finance, volume 48, pages 65 to 91).

Kenneth R. French's Data Library publishes 10 portfolios of US stocks formed each month on past return.

The measure is the return over the 11 months that end two months before the holding month. For the portfolio held in August 2026 that is August 2025 to June 2026, with July 2026 left out.

Group 10, the winners, holds the highest tenth. Group 1, the losers, holds the lowest tenth.

The breakpoints come from stocks listed on the New York Stock Exchange (NYSE). Each group is weighted by company size and held for one month.

The data cover 1,196 months, January 1927 to August 2026. The portfolios are re-formed every month, so real trading would carry costs. These results are before costs.

Each step up in past return came with a higher yearly return. The yearly returns are compound rates over the whole period.

They run from −0.8% for the losers to 16.8% for the winners, with the market at 10.3%. The winners beat the market by 6.5 points a year.

For each of the 99 calendar years from 1927 to 2025 the chart shows the winners' return minus the losers' return, in percentage points. The winners were ahead in 80 years and the losers in 19. The winners also beat the market in 70 of the 99 years.

The worst year was 2009. The winners made 9.7%, the losers 130.1% and the market 28.6%, a gap of −120.4 points. The next worst were 2003 at −38.2, 2021 at −37.7 and 1932 at −29.4.

The worst single month for the gap was August 1932 at −77.7 points. Four of the six worst months fell in 1932 or 2009: July and August 1932 and March and April 2009. Each pair came one and two months after the market's low, in June 1932 and February 2009.

The latest data show a sharp reversal too. In July 2026 the winners lost 15.4% and the losers made 7.3%, the 14th-worst month of 1,196 for the gap. Even so, from January to August 2026 the winners made 38.8%, the losers 12.1% and the market 12.7%.

The dot is the average yearly gap between winners and losers in percentage points: the average monthly gap times 12.

The bar is the 95% range, a rough guide to how far the average could sit from the true figure. It treats each month as an independent draw, which real returns are not.

From 1927 to 1999 the average gap was 16.6 points, with a range of 10.7 to 22.5. From 2000 to August 2026 it was 5.4 points, with a range of −6.8 to 17.6. A range that includes zero means the data cannot tell this gap from none.

By period, the average yearly gap was 22.8 points in the 1990s, 3.0 in the 2000s and 8.2 in the 2010s. From 2020 to August 2026 it was 4.8.

The average gap differs from the gap between the compound rates (19.2% and −1.5% in 1927 to 1999) because averaging and compounding treat big swings differently.

The chart follows $1 from the end of 1999. By August 2026 the winners had grown to $14.72, the market to $8.90 and the losers to $1.31. The winners' worst fall in these years was 51.3%, against 50.3% for the market.

The winners made 10.6% a year from 2000 against 8.5% for the market. Since 2010 they made 17.4% against 14.3%.

Before 2000 the winners had beaten the losers in 63 of 73 calendar years. From 2000 to 2025 they did in 17 of 26.

From 1994 to August 2026, the years after the 1993 paper, the winners made 14.3% a year, the losers 2.2% and the market 11.0%.

The gap is a difference between two long portfolios. Shorting the losers would add borrowing costs and risks that this data does not contain.

The winners' worst fall in the full record was 75.8%, against 83.7% for the market. The winners fell 15.5% from their June 2026 peak to August 2026.

Twenty-six calendar years since 2000 is a small sample. The test also covers one country's stocks.

Quant investing for beginners covers how to read a result like this one.

Over 99 years the winners beat the losers in 80 calendar years and lost to them by 120.4 points in 2009. Since 2000 the average gap has been 5.4 points, in a range that includes zero.

This explainer is part of the YX Insights Academy. Each one takes a single idea and checks it against real data.

The same approach runs through everything else we publish:

Good places to start on the website:

Momentum investing holds the assets that have risen most over the past year and avoids those that have risen least or fallen. The test here ranks US stocks each month on their return over the prior 11 months, ending a month ago, then compares the highest tenth with the lowest tenth.

From 2000 to August 2026 the top group of US stocks made 10.6% a year against 8.5% for the market. The average gap to the bottom group was 5.4 points a year, with a 95% range of −6.8 to 17.6 points. The data cannot separate that gap from zero.

Momentum investing compares assets with each other and holds the highest past returns. Trend following compares one asset with its own past, for example holding it while its price is above a moving average. Both use past prices, but they ask different questions.

In the 1927 to 2025 record the worst year was 2009. The top group of US stocks made 9.7% and the bottom group 130.1%, a gap of −120.4 points, while the market made 28.6%. The next worst years for the gap were 2003, 2021 and 1932.

DISCLAIMER: This article is strictly educational. Any information or analysis in this note is not an offer to sell or the solicitation of an offer to buy any securities. Nothing in this note is intended to be investment advice and nor should it be relied upon to make investment decisions. Any opinions, analyses, or probabilities expressed in this note are those of the author as of the note's date of publication and are subject to change without notice.