Momentum investing turns a simple observation into a repeatable portfolio rule: stocks that have outperformed their peers over recent months have sometimes continued to lead for a while. Implementing it well, however, is not the same as buying whatever is suddenly popular. A usable strategy needs a defined universe, a ranking formula, position limits, a rebalancing calendar and a plan for costs and sharp reversals.

The short answer: A straightforward long-only version can rank liquid U.S. stocks by their total return from 12 months ago through one month ago, buy the highest-ranked 20 to 50 names, weight them evenly or by risk, and review the portfolio monthly while limiting unnecessary trades. This is an educational framework, not individualized financial advice or a guarantee of profit.

The signal has deep research roots. The Kenneth French Data Library forms momentum portfolios using prior months 2 through 12, while the current S&P Momentum Indices methodology measures the 12-month price change while excluding the most recent month. MSCI combines six- and 12-month risk-adjusted momentum scores in its index process. Those are research and benchmark conventions, not proof that the next winner will keep winning.

Start by writing the rules

Before looking at any stock, put the strategy on one page. State which stocks are eligible, the exact return window, how many names you will own, how they will be weighted, when you will rebalance and what forces a sale. If a decision cannot be reproduced from those rules, discretion has crept back into the process.

A seven-step implementation

  1. Choose a liquid universe. Start with the constituents of a broad, established large- and mid-cap index, or create minimum market-capitalization and trading-volume requirements. A stable universe reduces exposure to thinly traded shares, large bid-ask spreads and stocks that are difficult to exit.
  2. Collect adjusted prices. Use prices adjusted for stock splits and cash distributions so that the return calculation is comparable across companies. Decide in advance how you will handle delistings, mergers, missing prices and companies with less than a full year of trading history.
  3. Calculate the signal. At each review date, divide the adjusted closing price from one month earlier by the adjusted price from 13 months earlier, then subtract one. That approximates a 12-month return that skips the latest month. Excluding that month helps keep a very short-term reversal from dominating the ranking.
  4. Rank every eligible stock. Sort the universe from highest to lowest signal. A simple strategy might select the top 10% or the top 30 names. Set the breadth before seeing the list; otherwise it is easy to keep expanding the portfolio to include a favorite company.
  5. Apply risk controls. Cap each position and each sector. One workable starting policy is no more than 5% in one stock and 25% in one sector, but the appropriate limits depend on the investor. Remove any company that fails the predefined liquidity test.
  6. Set target weights. Equal weighting is transparent: 25 stocks would begin near 4% each. Inverse-volatility weighting gives smaller allocations to stocks with larger recent price swings, but it adds data and calculation risk. Do not borrow money or short stocks merely because academic momentum portfolios often include a loser leg.
  7. Rebalance by rule. Review monthly, but trade only when a holding falls below a sell threshold, such as the top 20% after entering in the top 10%. This buffer can reduce turnover. Use limit orders when appropriate and record the signal, target and actual execution for every trade.
Diagram of stock tiles passing through a narrow entry gate, a wider holding buffer and a scheduled replacement path
A wider sell threshold can keep borderline holdings in place and reduce avoidable replacement trades.

A small hypothetical example

Suppose 300 stocks pass the size and liquidity screens on June 30. You calculate each stock’s adjusted return from May 31 of the previous year through May 31 of the current year, rank the results and select the top 30. With equal weights, each target begins at roughly 3.3%. If a stock ranks 45th the next month, a buffer rule might retain it; if it falls to 90th, the strategy sells it and promotes the highest-ranked eligible replacement.

This example shows why a momentum system is a process, not a prediction. It never says a stock is cheap, that earnings are strong or that the business is safe. It says only that the stock has recently performed better than most peers under the chosen definition.

Control the risks that matter most

  • Reversal risk: Leaders can fall quickly when a market panic ends or a crowded trade unwinds. Research by Kent Daniel and Tobias Moskowitz found that momentum strategies can suffer persistent strings of losses, particularly after market declines when volatility is high and the market rebounds sharply.
  • Concentration risk: Momentum rankings can cluster in one sector. Stock and sector caps prevent a rules-based strategy from quietly becoming a single-industry bet.
  • Turnover risk: Frequent replacement trades create spreads, slippage and possible taxes. The SEC notes that transaction and ongoing fees reduce the amount left to compound.
  • Data risk: A backtest that uses today’s index members, ignores delisted companies or assumes trades at unavailable prices can look much better than a strategy a person could actually have followed.
  • Behavior risk: The hardest moment is often a drawdown. Changing the formula after losses means the investor is no longer running the tested strategy.

Account for costs and taxes

Calculate results after commissions, bid-ask spreads, market impact and any subscription or data costs. In a U.S. taxable account, the IRS generally classifies gains on assets held for one year or less as short-term, with net short-term capital gains taxed as ordinary income. Because momentum can sell positions within a year, taxes may materially change the outcome. Tax rules depend on the investor and can change, so a qualified tax professional can help evaluate an actual account.

Backtest before committing money

A credible test uses point-in-time universe membership, adjusted prices, delisted securities, a delay between signal calculation and trade execution, and realistic costs. Compare the result with a broad-market index over the same dates. Look beyond total return to maximum drawdown, turnover, sector exposure, the number of holdings and the longest period of underperformance.

Then run a paper portfolio for several scheduled rebalances. Paper trading cannot reproduce every fill, tax consequence or emotional response, but it can reveal missing data, ambiguous rules and a workload that is harder than expected.

When a fund may be simpler

Investors who want momentum exposure without maintaining data and trading individual stocks can compare rules-based momentum exchange-traded funds. Read each fund’s methodology and prospectus: lookback periods, weighting, rebalance frequency, expense ratio and sector concentration can differ substantially. A fund reduces operational work but does not remove momentum risk, market risk or the possibility of loss.

The bottom line

A sound stock momentum strategy is deliberately boring. It defines the opportunity set, measures the same signal on every review date, diversifies among leaders, limits concentration, trades on a schedule and evaluates results after costs and taxes. The discipline is the strategy. The recent winners are only its current inputs.

Sources