Momentum Decay and Rotation Timing

A stock or sector that outperformed by 30% over the past six months has momentum. But momentum does not last forever—it decays. The rate and shape of that decay matter enormously for rotation strategies, because they determine the optimal window for identifying which sectors are hot (and buying them) versus which are cooling (and selling them). Momentum decay and rotation timing explains how to calibrate rebalancing frequency to capitalize on momentum without chasing yesterday’s winners.

What Is Momentum and Why Does It Decay?

In investing, momentum is the tendency for assets that have risen recently to continue rising in the near term, and for assets that have fallen to continue falling. A sector up 20% in the past six months often outperforms one that fell 20% in the same period—at least for another few weeks or months. This is momentum.

Momentum can arise from several mechanisms. Information diffuses slowly: not all investors may have recognized a positive earnings surprise immediately, so buying pressure continues for weeks. Behavioral biases like herding mean investors follow the crowd into successful sectors. Technical factors like forced buying (option expiration, rebalancing by index funds) can extend momentum. Tax-loss harvesting at year-end can temporarily reverse trends.

But momentum does not persist indefinitely. Eventually, the new information is fully priced in, the herding reverses (as contrarian investors bet against overextended moves), technical tailwinds fade, and the momentum regime flips. A sector that rose hard for six months begins to consolidate, then to fall. This fade—the decay of momentum from peak strength to diminished or negative returns—is the central dynamic that determines how often a rotator should rebalance.

The Shape of Momentum Decay

Empirical research on momentum decay reveals a general pattern, though specifics vary by asset class and time period.

Months 1–3 post-peak: Momentum is strongest immediately after a large move. A sector up 25% in months 1–6 often continues to outperform in months 7–9, though not by as large a margin. The decay is noticeable but not steep. For a momentum-based rotation strategy, a 6-month lookback captures the best part of the trend without having to time the exact peak.

Months 4–6 post-peak: The fade accelerates. A sector that was on fire months ago is now consolidating or rolling over. Its relative performance from the 6-month peak has rotated into the rearview mirror. Strategies using a 6-month lookback begin to capture declining returns; a 12-month lookback dilutes the signal.

Beyond month 6: Momentum often reverses. Assets that were on the largest uptrends begin to underperform. This reversal is sometimes called “mean reversion”—prices eventually return toward fair value—and it is often reinforced by profit-taking and contrarian positioning.

The decay curve is not linear. It has a sharp decline in the first 2–3 months, then flattens. This non-linearity is crucial: it means there is a “sweet spot” lookback period. Too short (1–2 months), and noise dominates; too long (18 months), and the signal includes trailing losers. For most equity sectors, 6–12 months is the empirically optimal window.

Lookback Period and Rotation Frequency

The relationship between lookback period and rebalancing frequency is direct. If you identify the strongest-momentum sectors using a 6-month lookback and rebalance into them monthly, you are likely rebalancing before momentum has materially decayed; you are capturing fresh momentum and avoiding stale winners. If you rebalance quarterly, you allow one round of natural decay but still capture multiple months of continued performance. If you rebalance annually, you are allowing significant decay and risk buying sectors whose momentum has already faded.

Monthly rebalancing with a 6-month lookback: This is an aggressive, high-turnover strategy. It requires strong trading discipline and low transaction costs. It works well in volatile, trend-driven environments but generates tax drag in taxable accounts. A rotation strategy rebalancing monthly should expect 10–15% annual turnover.

Quarterly rebalancing with a 6-month lookback: This is a common middle ground. Every three months, you rank sectors by their 6-month returns and shift weights toward the strongest. Three months is long enough to let some decay happen and short enough to capture multiple cycles of momentum. Annual turnover is 4–8%, lower than monthly but still meaningful.

Semi-annual rebalancing with a 12-month lookback: This is conservative and suited to longer-term rotators or those averse to trading costs. A 12-month lookback is wide enough to filter out noise and include multiple seasons of data. Rebalancing every six months means you are harvesting momentum that is already partly decayed but still present. Turnover is 2–4% annually.

Annual rebalancing: This is so low-turnover that it is hardly a “rotation” strategy—it is closer to value investing with a slight momentum tilt. Use annual rebalancing only if you are optimizing for tax efficiency (in taxable accounts) or if you believe momentum is so noisy that the expected value is near zero.

Calibrating for Decay: Practical Examples

Suppose you manage a sector-rotation strategy. You rank all 11 sectors by their 6-month returns at the beginning of each quarter and allocate 20% of your portfolio to the top two sectors, 15% to the middle three, and 10% to the bottom three.

At Q1 start, energy (up 30%) and financials (up 18%) are your top two. You buy them.

By Q2 start, energy is up another 5% and financials are up 3%. Momentum is positive but fading. You rebalance and find that materials (which lagged in Q1) have now risen sharply, and energy momentum is flatlining. Your new allocation shifts weight toward materials. You sell some energy and buy materials.

By Q3 start, energy has actually fallen 8% from its Q2 high. Momentum has reversed. Your 6-month lookback now includes the Q1 surge but also the Q2–Q3 decline. Energy’s 6-month return is now +15% (not the +30% from Q1). It drops from “top sector” to “middle-tier.” You rotate out.

In this example, quarterly rebalancing with a 6-month lookback allowed you to capture energy’s momentum in Q1, begin harvesting it in Q2, and exit before the reversal fully took hold. If you had rebalanced monthly, you would have harvested even more of the rise but also incurred higher costs. If you had rebalanced annually, you would have held energy longer into its decline.

Decay and Transaction Costs

Momentum decay is not the only factor determining optimal rebalance frequency—costs matter equally. Each trade carries a commission, bid-ask spread, and (in taxable accounts) potential tax consequences. A rotation strategy that rebalances excessively will lose returns to costs that exceed the value of the momentum captured.

Suppose a quarterly momentum strategy expects to outperform a buy-and-hold index by 1.5% per year from momentum capture. But rebalancing costs (spreads, commissions, taxes) are 0.3% per quarter, or 1.2% annually. Your net advantage shrinks to 0.3% per year—marginal and exposed to noise.

In contrast, a semi-annual strategy might capture only 1% of momentum (because some has decayed), but costs only 0.4% annually (two rebalances instead of four), netting a more attractive 0.6%.

Costs determine the practical lower bound on rebalancing frequency. If you cannot rebalance for less than 0.15% per trade (in a large, liquid portfolio), you cannot profitably use a monthly rebalancing frequency unless momentum is exceptionally strong and stable.

Momentum Decay Across Asset Classes

Momentum decay patterns differ across assets.

Equities. Individual stock momentum is quite strong over 3–12 months. Sector momentum (as discussed here) persists over 6–18 months. A 6-month lookback is standard for sector rotation.

Bonds. Duration trends (long bonds outperforming or underperforming relative to short bonds) can persist for 1–2 years, so a 12-month lookback is more appropriate. Rebalancing quarterly or semi-annually is common.

Commodities. Commodity trends are driven by supply-demand cycles and tend to have longer duration (12–24 months). A 12-month lookback with semi-annual rebalancing is typical.

Currencies. Forex momentum is much shorter-lived than equities, with meaningful decay in 1–3 months. Short-term rotators use 1–3 month lookbacks and rebalance monthly.

Seasonality and Decay Inflection Points

Momentum decay is not always smooth. Certain calendar windows see sharp reversals or accelerations.

Year-end and tax-loss harvesting. Late December often sees temporary momentum reversals as investors harvest tax losses. A sector that was outperforming may drop sharply in December only to recover in January. A rotation strategy rebalancing in December risks buying into a temporary trough; rebalancing in January may mean buying after the bounce has already occurred.

Earnings season. During quarterly earnings periods, momentum can be significantly disrupted. Sectors with strong momentum into earnings often face profit-taking if guidance is cautious. A rebalancing window that straddles earnings season may capture elevated volatility and decay.

Central bank events. Federal Reserve meetings, employment reports, and inflation data can suddenly shift momentum regimes. A strategy rebalancing just before or after these events may encounter stale momentum.

Experienced rotators are aware of these inflection points and adjust rebalancing timing accordingly—for example, rebalancing in early January rather than late December to avoid year-end tax-driven noise.

The Role of Trend-Following Filters

Rather than using a fixed rebalancing schedule, some strategies add a trend-following filter: only buy sectors with positive momentum (positive 6-month returns) and only when that momentum is accelerating (recent returns stronger than lagged returns). This approach reduces whipsaw by filtering out decaying momentum and sideways-grinding sectors.

For example, a strategy might say: “Rank sectors by their 6-month returns, but only hold positions in the top five sectors if they also show positive 3-month returns and are trading above their 200-day moving average.” These filters prevent buying sectors whose momentum is about to reverse sharply.

See also