Momentum Factor Decay and Half-Life

The momentum factor—the tendency for assets that have outperformed to continue outperforming in the near term—is one of the most persistent patterns in financial markets. Yet its predictive power erodes over time. Understanding momentum factor decay means recognizing that the same price trend that drove outsized returns in weeks two through twelve of a holding period often reverses by month eighteen, as the signal loses force and mean reversion reasserts itself. The half-life of momentum—the time at which the original signal retains only 50% of its explanatory power—typically spans three to twelve months depending on asset class and market regime.

Why momentum signals decay

Momentum arises from several mechanisms, and each decays at a different rate. Underreaction (market participants slowly digesting information) typically fades within weeks to months as news fully propagates and is priced in. Herding and trend-chasing (behavioral investors piling into winners) can persist longer, but eventually new capital dries up and the crowd reverses. Risk repricing (winners being riskier or riskier assets offering higher expected returns) is more durable but still shifts as fundamentals evolve.

Once the initial catalyst loses novelty—new product launches are old news, earnings surprises are priced, or the macro regime pivots—demand from momentum-following investors tapers. Crucially, the same conditions that created the initial momentum often revert. A firm that beat earnings and saw its stock surge may face normalizing comparisons in the next quarter. A currency that weakened from carry unwinding may stabilize once flows reverse. An asset that rallied on fear-driven buying may face selling as confidence returns.

Measurement: lookback and holding windows

Empirical momentum studies typically define the signal using a 3-to-12-month lookback window (how far back to measure prior returns) and then measure subsequent performance over a 1-to-12-month holding period. A classic construction: the return over the prior 6–12 months predicts returns over the next 1–6 months.

The decay curve is usually nonlinear. In the first month after forming the signal, momentum predictability is strongest. By month three, it has typically eroded 30–50%. By month six, it is often halved again. By month twelve, many studies find predictability is statistically insignificant or has reversed.

This nonlinearity is crucial for active managers. A momentum strategy that rebalances monthly captures the high-decay window and avoids holding losers. A buy-and-hold momentum portfolio, by contrast, captures the decay and may turn negative.

The half-life concept explained

The half-life is borrowed from radioactive decay (in physics, how long until half the material remains). Applied to momentum, it is the holding period at which the signal’s information coefficient (predictive power) falls to half its original level.

If a momentum signal has an information coefficient of 0.10 (suggesting 100 basis points of annual excess return per unit of momentum exposure), and its half-life is four months, then by month four the signal’s information coefficient is roughly 0.05. By eight months, it approaches 0.025. This nonlinear decay means the majority of momentum’s edge is captured in the first half-life, and exiting near that point maximizes risk-adjusted returns.

The half-life varies:

• Stocks (developed markets): 3–6 months; fastest decay among major asset classes
• Cryptocurrencies: 2–4 weeks; extremely rapid decay, often driven by retail herding and quick reversal
• Commodities: 6–12 months; slower decay, likely because supply-demand shifts persist longer
• Foreign exchange: 1–3 months; faster decay due to quick mean reversion and central bank resistance
• Emerging market stocks: 4–8 months; decay slower than developed, partly due to lower analyst coverage and slower information flow

Why exit timing matters

Many momentum traders focus on entry—identifying which assets have the strongest recent outperformance. Fewer focus systematically on exit. Yet the decay profile shows that timing the exit is equally important.

A momentum portfolio held for three months captures the sweet spot: full exposure to underreaction and early-stage herding, before reversal signals emerge. Held for nine months, the same portfolio has eroded significantly and now carries concentration risk to overshooting and reversal.

Consider a simple example: a stock rallied 40% over six months (the momentum signal). A trader enters a six-month holding period expecting 8–10% of additional returns based on historical momentum. But by month three, as the signal decays, the stock may have already delivered half those gains; by month six, reversals and mean reversion may erase the remaining edge. The trader who exited at month three, satisfied with high risk-adjusted returns, captured more value than the one who held hoping for further gains.

Reversal: the decay endpoint

After momentum’s half-life, mean reversion begins to dominate. Assets that have surged tend to underperform; assets that have lagged tend to outperform. This reversal can be gradual (fading over months) or sharp (when a large crowded trade unwinds).

The reversal is strongest in the 6–12 month window after the initial momentum signal. A stock that had soared on a beat in month three may crash in month nine when growth slows or multiple compression hits. This is not coincidence—momentum traders often use stop losses that trigger at the inflection point, and crowded momentum trades are vulnerable to cascading forced selling.

Understanding this reversal is critical for factor investing practitioners. A momentum portfolio is not “buy the strongest and hold forever”; it is “capture the signal’s edge, then exit or reverse.”

Regime dependence and decay acceleration

The half-life is not fixed. During low-volatility, stable regimes, momentum decays slowly because the underlying risk repricing happens gradually. During stress, the decay accelerates sharply—momentum often crashes in the first week of a bear market as technical selling and forced liquidations hit winners first.

The 2020 COVID crash saw extraordinary momentum decay. Momentum winners from January vanished in March as liquidity seized and volatility spiked. The half-life collapsed to days. Similarly, in 2022, as the Federal Reserve tightened unexpectedly, growth momentum decayed in weeks instead of months.

This regime dependence means a trader cannot set a static exit schedule. A momentum manager must monitor both the signal itself and broader market conditions, adjusting exits if volatility or correlation regime changes sharply.

Momentum factor decay in different markets

Equities: Decay is fastest, with half-lives around 3–6 months. Stocks are heavily researched, information propagates quickly, and the crowd is large and reactive. Momentum edges erode quickly.

Commodities: Decay is slower, 6–12 months. Commodity supply-demand dynamics shift gradually, and momentum is less crowded than in equities. The underlying economic drivers persist.

Currencies: Decay is fast, 1–3 months. Central banks resist sustained currency trends, and carry trades unwind predictably, flattening the momentum signal.

Bonds: Decay is moderate to slow, 4–8 months. Bond momentum is driven partly by duration repricing, which can persist, and partly by supply flows, which evolve gradually.

Practical implications for momentum investing

1. Rebalance frequently: Monthly or quarterly rebalancing avoids holding winners too long past their peak predictability. Transaction costs must be weighed, but the edge is largest early.

2. Set explicit exit rules: Define a time-based exit (e.g., sell after six months) or a signal-based exit (e.g., reverse if momentum flips below zero). Avoid vague “hold if it’s still strong” rules.

3. Diversify momentum lookbacks: A portfolio using both 3-month and 12-month lookback periods captures different phases of decay and reduces idiosyncratic drawdowns.

4. Watch for reversals: Once the signal decays below half its original strength, reversal risk rises sharply. Position sizing should shrink.

5. Adjust for regime: During high-volatility or stress periods, shorten holding periods and expect decay to accelerate. Have a pre-set contingency for market dislocations.

See also