A stock, measured against its own past
In 1967, Robert Levy divides a stock's price by its own average over the past months. This single division is relative strength — then as now.
No balance sheet, no earnings estimate, no knowledge of the industry. The result is a number around 1.0 — above 1.0 the price is above its own average, below 1.0 it is below. From this number Levy builds a ranking and updates it every week.
This page explains where the measure comes from, how it is calculated and what it is not. That it was refuted three years later is covered here as well — it is the more important part of the story. What research made of it over the following fifty years is the subject of the second part: Evolving the RSL strategy.
The idea in one sentence
A price that is above its own six-month average is more likely to stay up than to go down.
That is how Levy himself defines it, as C/A26: the price of the current week, divided by the average of a series that ends with that same price and includes the 26 weeks immediately before it — 27 prices in total. In the original:
The price for the current week divided by the average of the series of prices
ending with the price for the current week and including the prices for the 26 weeks
immediately preceding.
Robert A. Levy 1967, Section II “Price Ratios”
Why half a year, and why an average? Levy explains both in the same section. He chose the periods because they are familiar — half a year — and because of the six-month holding period in US tax law. He uses the moving average because it smooths out exaggerated swings and is widely used in practice. Robert A. Levy 1967, Section II “Price Ratios”
That is all there is to it. The measure relates a price to its own past and needs no balance sheet, no earnings estimate and no knowledge of the industry. This very simplicity is why it can be recalculated — and why it has been testable for more than fifty years.
This list calculates it the same way, day after day, for every stock in the indices it covers. Which other columns stand next to it, and why, is explained in the second part.
The origin
Robert A. Levy publishes a measure in the Journal of Finance that needs no balance sheet, no earnings estimate and no knowledge of the industry: the closing price of the current week, divided by the average of that same price and the 26 weekly closing prices before it. He calls it C/A26. From this ratio he builds a ranking of all stocks and updates it every week.
He tested it on a manageable sample: the weekly closing prices of 200 stocks listed on the New York Stock Exchange over 260 weeks, from October 24, 1960 to October 15, 1965. Throughout, he works with the strongest 10 percent of the ranking. Over the following 26 weeks they rose considerably more on average than the weakest 10 percent — computed with dividend-adjusted prices and before costs. The two figures are shown in the tiles below.
One finding from his Table 2 is rarely cited today: the best results came from stocks that were relatively strong and relatively volatile. That is why this list does not rate the Vola6M column in the second part: it shows how much a stock fluctuates without coloring it green or red. Barroso and Santa-Clara answer a different question — they adjust over time how much a strategy puts to work overall, and for that, calm is the better sign. For choosing which stocks, it is not.
The paper goes back to his doctoral dissertation; at the time, Levy was president of Computer Directions Advisors in Silver Spring, Maryland. This is where this list starts. What came afterwards is the reason it does not stop with him.
Source: Robert A. Levy 1967, Section II (“Price Ratios”, “Construction of the Data File”), empirical part, Table 2, Pedro Barroso 2015
The setback
Three years later, Michael C. Jensen and George A. Benington re-examine two of Levy's rules — using the monthly data of the Center for Research in Security Prices for the New York Stock Exchange, 1,952 stocks. From these they draw 29 independent samples of 200 stocks each over consecutive five-year periods from 1931 to 1965. Levy's own five years are among them; but his rule is now tested on 35 years instead of five.
The result: after costs, the rules on average earn no significantly more than buy-and-hold. And because their portfolios were riskier, even this comparison flatters them — against buy-and-hold with the same risk, the two rules fall short by 0.31 and 2.36 percent after costs. Their verdict:
… the behavior of security prices on the N.Y.S.E. is remarkably close to that predicted by the efficient market theories of security price behavior, and Levy’s (1967a) conclusion that “… the theory of random walks has been refuted,” is not substantiated.
Michael C. Jensen 1970, Summary and Conclusions
The second objection concerns selection. In his dissertation, Levy had worked through “some 68 variations” of trading rules, of which only very few beat buy-and-hold — on the same data on which he then demonstrated the published rules. Anyone who picks the best of 68 rules is very likely to find one that looked good, even if not a single one of them is worth anything. Given enough computer time, Jensen and Benington write, one could find a rule that “works” even in a table of random numbers — as long as it is tested on the same table.
Levy names the third objection himself. His sample only included stocks that were listed on the NYSE for the whole period and appeared in Moody’s handbook in May 1965 — selected, in other words, with the benefit of hindsight. He considered this harmless:
Although the sampling procedure was ex post, the author considers it unlikely that the test results have been materially biased.
Robert A. Levy 1967, Section II “Construction of the Data File”
The technical term for the second objection is common knowledge today; in 1970 it was a new argument. It is also the most honest point of this presentation: the history of relative strength begins with a refutation, not with a success. Why this objection also applies to our own choice of measures is explained in the second part under “What of this is in this list”.
Source: Michael C. Jensen 1970, data description, Summary and Conclusions, Robert A. Levy 1967
What this list is not
The effect is documented — that price strength persists for a while. What is not documented is that any particular number in this table predicts tomorrow's price.
Research measures broadly diversified portfolios over months, not individual stocks over days. Two quantities with similar names that are not the same thing:
| Levy's RSL — what we calculate | Momentum — what research measures | |
|---|---|---|
| Quantity | Price divided by its own moving average | Return over the past 3 to 12 months |
| Reference | The stock against its own past | The stock against all other stocks |
| Result | A ratio around 1.0 | A ranking by past return |
The figures on this page are Levy's own and those of the researchers who re-examined them. Levy only bought, without short selling — but he calculated before costs and taxes, for groups of stocks rather than individual ones, on 200 stocks from 1960 to 1965, and with a sample he put together after the fact. Jensen and Benington calculated after costs and were left with nothing. Both describe what was measured at the time. Neither says anything about what an individual stock in this table will do next.
There is no backtest of our own here either. We have not run one, and claiming one would be made up. This list is a calculation tool: it applies a disclosed formula to publicly available prices and shows the result.
Not investment advice and not a recommendation. Past performance is not a reliable indicator of future results, and investing in securities involves the risk of losing your entire investment.
And then?
Up to this point, relative strength is a measure from 1967 that was refuted in 1970. That would be a poor reason to calculate it today.
What has happened since — independent confirmation on other data, adoption into the industry's standard models, the crashes that come with it, and the three refinements that appear as separate columns in this list — fills the second part: Evolving the RSL strategy.
References
Every figure on this page is shown with its exact location in the source — section, table or page. The links lead to the paper via its DOI; the full texts are mostly behind publishers' paywalls, but the bibliographic details are enough for any library.
- Robert A. Levy (1967): Relative Strength as a Criterion for Investment Selection. The Journal of Finance 22(4), pp. 595–610. doi:10.1111/j.1540-6261.1967.tb00295.x
- Michael C. Jensen, George A. Benington (1970): Random Walks and Technical Theories: Some Additional Evidence. The Journal of Finance 25(2), pp. 469–482. doi:10.1111/j.1540-6261.1970.tb00671.x
- Narasimhan Jegadeesh (1990): Evidence of Predictable Behavior of Security Returns. The Journal of Finance 45(3), pp. 881–898. doi:10.1111/j.1540-6261.1990.tb05110.x
- Bruce N. Lehmann (1990): Fads, Martingales, and Market Efficiency. The Quarterly Journal of Economics 105(1), pp. 1–28. doi:10.2307/2937816
- Narasimhan Jegadeesh, Sheridan Titman (1993): Returns to Buying Winners and Selling Losers: Implications for Stock Market Efficiency. The Journal of Finance 48(1), pp. 65–91. doi:10.1111/j.1540-6261.1993.tb04702.x
- Mark M. Carhart (1997): On Persistence in Mutual Fund Performance. The Journal of Finance 52(1), pp. 57–82. doi:10.1111/j.1540-6261.1997.tb03808.x
- Tobias J. Moskowitz, Mark Grinblatt (1999): Do Industries Explain Momentum?. The Journal of Finance 54(4), pp. 1249–1290. doi:10.1111/0022-1082.00146
- Thomas J. George, Chuan-Yang Hwang (2004): The 52-Week High and Momentum Investing. The Journal of Finance 59(5), pp. 2145–2176. doi:10.1111/j.1540-6261.2004.00695.x
- Clifford S. Asness, Tobias J. Moskowitz, Lasse Heje Pedersen (2013): Value and Momentum Everywhere. The Journal of Finance 68(3), pp. 929–985. doi:10.1111/jofi.12021
- Pedro Barroso, Pedro Santa-Clara (2015): Momentum has its moments. Journal of Financial Economics 116(1), pp. 111–120. doi:10.1016/j.jfineco.2014.11.010
- Kent Daniel, Tobias J. Moskowitz (2016): Momentum Crashes. Journal of Financial Economics 122(2), pp. 221–247. doi:10.1016/j.jfineco.2015.12.002
- Kenneth R. French (2026): Data Library: Momentum Factor (Mom), monthly and annual. Tuck School of Business, Dartmouth College, dataset, CRSP data as of 202607. Data library
The values of the momentum factor are the only figures on this page that we calculate ourselves: from the raw data of the data library, CRSP data as of 202607, monthly values from 1927 to 2026. They belong to a long-short portfolio with short selling and are not a return of this list.