Value-Growth Rotation Tested: Backtested Rules for Regime Shifts

A systematic backtest of value and growth rotation strategies dismantles the binary choice that most traders carry. The output is not a declaration of which style is permanently superior; it is a decision rule governed by the factor spread and the interest-rate regime. A strategy built on static allocation to either bucket misses the mechanism that actually drives relative performance. The backtest replaces a philosophical debate with a tradeable signal.

The Duration Mechanism, Not Valuation

Value stocks outperform when the discount rate applied to future cash flows rises, because their cash flows are front-loaded. Their effective equity duration is shorter. Growth stocks compound earnings far in the future, which makes them longer-duration assets highly sensitive to changes in the discount rate. The simple read is that value is cheap and growth is expensive. The better read is that the spread between value and growth is a direct expression of duration risk being repriced every time the rate outlook shifts.

When Treasury yields climb and the equity risk premium widens, the present value of distant growth collapses faster than the present value of near-term cash flows. That is why value leadership usually coincides with rising real rates. The same logic explains why growth names such as Apple (AAPL) and NVIDIA (NVDA) rally when the market expects rates to fall or stay low – lower discount rates inflate the terminal value embedded in their multiples. The trade is not about buying cheap stocks; it is about measuring the factor spread and recognizing which side the rate environment is supporting.

Backtest Elements That Survive Real Trading

A backtest that ignores regime shifts produces results that look pristine on paper and break in execution. Three elements separate a useful backtest from a data-mining exercise.

• Factor spread as the entry signal: Measure the valuation gap between the cheapest and most expensive quintiles – a normalized price-to-book or price-to-earnings spread over a rolling 10-year window. An absolute cheapness screen loads up on cyclicals at the worst point in the cycle. The spread tells you how extreme the current dislocations are relative to history.
• Momentum filter: A value entry without a momentum overlay catches falling knives; a growth exit without a momentum deterioration signal risks riding a blow-off top into a drawdown. Require positive 6-month price momentum for value additions and negative 6-month momentum for reducing growth exposure. The filter prevents the backtest from celebrating paper gains that never materialize.
• Implementation costs: Subtracting 50 basis points per round-trip for large-cap names and 150 basis points for small-cap names – or using actual execution data – reveals how much theoretical alpha evaporates. A backtest that does not account for the bid-ask spread on illiquid value names is an academic exercise, not a trading plan.

Watchlist Rotation, Not Factor Timing

The practical output from a properly specified backtest is a rotation rule for a live watchlist. The signal does not call for a binary switch into a value or growth ETF. It dictates a tilt. When the factor spread widens beyond one standard deviation, the probabilities shift. A value tilt becomes actionable; the backtest identifies the conditions that have historically rewarded it. At that point, screening for value names with improving fundamentals and growth names with deteriorating momentum becomes the trade, not blind factor exposure.

The stock market analysis tools that matter here are factor-exposure reports and sector-level flow data, not the daily commentary that declares value dead or growth permanently ascendant. The Apple (AAPL) and NVIDIA (NVDA) positions that work in a falling-rate regime will hurt when the regime reverses, and the backtest quantifies that relationship.

The edge in value-growth rotation never comes from picking one side and sticking with it through a cycle; it comes from knowing when the market is pricing one side too cheaply relative to the other. The next step is to pull five years of factor-spread data, code the entry and exit rules, and run the test. If the signal does not survive a simple transaction-cost adjustment, the strategy stays on the research shelf. If it survives, the backtest becomes a regime-monitoring tool that tells you exactly when to tilt your watchlist.