Switch to the ETF That's Rising? How Momentum Rotation Works and What to Check in a Backtest

Review your investments at the start of each month. The candidates are three ETFs—a ChiNext ETF, a Nasdaq ETF, and a gold ETF. Compare their returns over the past 21 trading days and hold the strongest one. Repeat the ranking process the following month, switching if the leader has changed.

This is the basic idea behind momentum rotation. Rather than predicting which assets will rise, you select an asset that is already strong and move to another candidate when it weakens. The process is simple, but results can vary significantly depending on the candidates, rebalancing frequency, and rules for handling declines.

Momentum Rotation Selects Assets by Relative Strength#

Momentum measures how much a price has moved over a given period. For a simple price return, it can be calculated as follows:

動量 = 現在価格 ÷ N営業日前の価格 - 1

For example, if ETF A rose 4%, ETF B rose 1%, and ETF C fell 3% over the past 21 trading days, the ranking would be A, B, then C. If the rule is to hold the top-ranked ETF, you would select A. You then recalculate the rankings on the next rebalancing date.

This approach uses the tendency of strong assets to continue rising for a while. That tendency does not last forever, however. During a sudden market reversal, the ETF that was leading until recently can fall sharply. Momentum is a selection rule based on past price movements, not a forecast that guarantees future profits.

First, Define the Candidates, Lookback Period, and Rebalancing Schedule#

When putting a rule into practice, decide on the following first:

• Which ETFs to include
• How many trading days of price movements to use for ranking
• How often to recalculate rankings and review holdings
• What to do when all candidates are falling

Choose ETFs with sufficient liquidity and reliable historical price data. Combining ETFs from different equity markets, regions, or sectors can reduce dependence on a single market. Adding more ETFs does not automatically make the strategy safer, however. If their underlying holdings or price movements are similar, the diversification benefits will be limited.

Shorter lookback periods respond faster and tend to produce more frequent ranking changes. Longer periods take more time to catch changes in price trends but are less sensitive to short-term fluctuations. As a starting point, you could test 20–60 trading days for a short-term window, 60–120 for a medium-term window, and 120–252 for a long-term window. The best choice depends on the candidates and the test period.

Rebalancing frequency involves the same tradeoff. Daily reviews can detect changes quickly, but they also increase trading and costs. Weekly and monthly reviews are alternatives; monthly rebalancing is one way to track rankings while limiting trading. Define the date precisely—for example, “the first trading day of each new month”—rather than using an ambiguous phrase such as “the last day of each month.” This makes backtest conditions easier to keep consistent.

Holding One ETF and Holding Several ETFs Are Different Rules#

The term momentum rotation may suggest a strategy that buys several top-ranked ETFs. In practice, some versions hold only the single strongest ETF, while others hold the top two or three in equal proportions. Their concentration and trading behavior differ, so they should be treated as separate strategies.

InvestPilot’s “Momentum Rotation Strategy” selects the single candidate with the strongest momentum. It differs from strategies that hold several top-ranked ETFs together. For example, even with three candidates, the default approach is to hold the leader rather than spread the investment across all three. Keep this distinction in mind when reviewing the strategy description and parameters.

Look Beyond Returns in a Backtest#

Backtesting means applying rules to historical price data to see what results they would have produced. When comparing results, keep at least the following consistent:

• Candidate ETFs and test period
• Momentum lookback period
• Rebalancing frequency and how rebalancing dates are determined
• Estimated trading costs
• Benchmark index or ETF

Check not only profits and losses but also maximum drawdown, volatility, number of trades, and trading costs. Even if annualized returns look high, whether you can stick with the strategy is a separate question if it suffers steep declines along the way or trades frequently enough to incur substantial costs.

InvestPilot’s documentation gives annualized returns of 8–15% and maximum drawdowns of 15–25% as examples of typical momentum rotation performance. These figures are not promises of future results. Outcomes vary with market conditions, candidates, parameters, and cost assumptions. You need to examine the backtest data to see whether those results hold under your own conditions.

Repeatedly changing backtest settings and selecting only the most attractive result is also risky. Settings tailored to a particular historical period may fail in another market environment. Test separate periods and check how results change with different candidates and parameters to avoid relying too heavily on a single figure.

In a Falling Market, Even the “Strongest” Can Be Negative#

A rule that simply selects the top-ranked candidate will continue holding something even when the entire market is falling. An ETF can rank first despite a negative return simply because it has fallen less than the other candidates.

InvestPilot provides a way to add a defensive ETF to the candidate list and switch to it when specified conditions are met. These include a “relative strength gate,” which triggers a switch when the momentum of risk assets falls below that of the defensive ETF, and an “absolute momentum gate,” which triggers a switch when the leader’s momentum is at or below a specified threshold. The former compares strength against a defensive asset; the latter checks whether the risk asset itself is rising.

When specifying a defensive ETF, you must also add its ticker to the candidate list. Entering the ticker in the settings field does not automatically add it to the tradable universe. A ticker missing from the candidate list may prevent the intended defensive behavior, so check both settings together.

Understanding InvestPilot’s Parameters#

InvestPilot’s momentum rotation strategy mainly uses the following parameters:

• momentum_period: The number of past trading days used to calculate momentum
• rebalance_frequency: Rebalancing frequency; d means daily, w weekly, and m monthly
• defensive_symbol: The ticker of the defensive ETF
• use_defensive_relative_gate: Whether to use a relative comparison against the defensive ETF
• absolute_momentum_threshold: The threshold at or below which the leader’s momentum triggers a switch to the defensive ETF

The example configuration uses 21-trading-day momentum to select the leader each month from a ChiNext ETF, a Nasdaq ETF, and a gold ETF. Rebalancing takes place on the first trading day of each new month. To use a defensive ETF as well, add it to the candidates and configure the switching conditions. Assess whether the example suits your own approach by reviewing its candidates and conditions.

What to Check Before Live Trading#

To trade in line with a backtest, you also need to consider execution price differences and how orders are placed. Bid–ask spreads, fees, and market impact can reduce the profits shown in a backtest. Even with monthly rankings, large price moves on the rebalancing date can prevent execution at the assumed price.

Planned withdrawals, taxes, and account-specific trading conditions also affect live implementation. Decide in advance whether you will keep following the rules or pause after a series of losses; waiting until performance deteriorates is too late. Document these decisions as part of your trading rules.

Momentum rotation is a method of periodically switching to relatively strong ETFs. In backtests, examine drawdowns, costs, and trading frequency as well as returns. If you are configuring your own candidates in InvestPilot, first choose the lookback period and rebalancing frequency, test the rules—including defensive conditions—against historical data, and then review the official configuration documentation and related examples.

Source: https://www.myinvestpilot.com/docs/strategies/momentum-rotation