A defensive strategy can meet its downside objective and still spend years lagging equity. The analysis evaluates those two outcomes separately rather than treating benchmark outperformance as evidence that downside protection succeeded.
The long-history analysis compares equity buy-and-hold, a monthly rebalanced 50/50 equity-cash allocation, and a fixed 12-month defensive trend rule. Rolling 5-, 10-, 15-, and 20-year holding periods are used to study both drawdown protection in the strategy's own wealth path and the benchmark-relative path experienced while obtaining it.
Four statistics summarize the central 20-year results in the July 1927–June 2026 Kenneth R. French Data Library sample.
Four selected working-paper figures summarize downside protection, the benchmark-relative path, an illustrative 20-year episode, and the role of persistent equity exposure. Click any figure to inspect the full-resolution public image.
The four-state decomposition shows that realized downside protection does not imply benchmark outperformance. At long horizons, protected-but-underperforming periods remain common.
Open full-size figure ↗
Protected holding periods can remain below a prior benchmark-relative high-water mark for years. Depth and duration are separate dimensions of the relative performance path.
Open full-size figure ↗
The June 2005–May 2025 illustrative path combines materially lower maximum drawdown with a 195-month benchmark-relative underwater spell and lower terminal wealth than equity.
Open full-size figure ↗
Across fixed tactical shares, retaining more persistent equity exposure materially reduces benchmark-relative drawdown depth, while the median longest underwater duration changes much less.
Open full-size figure ↗Protection can arrive in concentrated stress episodes while benchmark-relative opportunity cost accumulates across much longer intervals.
June 2005–May 2025: the trend rule reduces maximum drawdown by 27.1 percentage points, yet ends at 70.4% of equity's terminal wealth and spends 195 months below a prior benchmark-relative high-water mark.
The 20 largest months of qualifying defensive outperformance account for 36.2% of total positive monthly active log returns after the signal becomes available; the fixed 50/50 allocation accounts for 19.6%.
The trend rule records 112 qualifying months under the paper's fixed criterion, with a median gap of two months. These counts are descriptive because adjacent crisis months can belong to the same market episode.
Measure whether the defensive allocation reduces maximum drawdown relative to equity within the same holding period.
Record whether terminal wealth finishes above or below equity buy-and-hold.
Measure benchmark-relative drawdown depth and the longest spell below a prior relative high-water mark separately.
Persistent equity exposure changes the severity of the benchmark-relative shortfall more than its duration.
Median benchmark-relative drawdown depth across all 20-year periods. The corresponding median longest underwater spell is 196 months.
Median relative drawdown depth across all 20-year periods, while the median longest underwater duration remains about 190 months.
The fixed blends are descriptive portfolio-design comparisons, not estimated optimal allocations. Their role is to show how benchmark-relative depth and duration move as persistent equity exposure increases.
The qualitative depth-versus-duration pattern remains visible in an independently reconstructed historical series over the matched February 1935–June 2023 sample, although the magnitudes differ from the Kenneth R. French Data Library results.
Fixed turnover costs worsen benchmark-relative performance but do not remove the coexistence of realized downside protection and prolonged relative shortfall.
The long-history results describe U.S. return paths and do not establish that the same magnitudes hold across other markets, asset classes, or future regimes.
Rolling periods overlap and are strongly dependent. Reported frequencies summarize the historical sample; they are not probabilities estimated from independent trials.
Return histories do not identify an investor's actual abandonment threshold. Benchmark-relative tolerance remains an implementation constraint rather than an estimated behavioral parameter.
The timing rule and blend weights are fixed design choices. The analysis characterizes historical trade-offs rather than estimating an optimal tactical allocation.
These companion articles explain methods used directly in the study or needed to interpret its design and evidence.
Explains the portfolio accounting, weight drift, monthly rebalancing, turnover conventions, and next-period timing needed to interpret the portfolio paths used in this research.
Explains drawdown, high-water-mark, underwater-duration, recovery, and censoring conventions used to interpret path-dependent downside and recovery in this research.
Explains drift-aware turnover, transaction-cost conventions, and the mapping from before-cost portfolio performance to performance net of modeled transaction costs.
Explains how momentum and trend signals are defined, timed, and stress-tested across lookback choices before they are translated into portfolio decisions.
Explains rolling long-horizon outcome analysis, overlapping-window dependence, and the distinction between endpoint outcomes and the path experienced while reaching them.
Explains active return, tracking error, information ratio, relative wealth, and benchmark-relative drawdown as distinct portfolio-evaluation quantities.
The public replication repository reproduces the long-history analysis, fixed blends, trading-cost sensitivity, independent reconstruction, and all five working-paper figures. Raw third-party source files are not redistributed.
Python pipeline, tests, reference hashes, generated results, and manuscript figures.
Open GitHub ↗Latest tagged public package: v1.1.1. The repository is synchronized to the final public SSRN paper.
Open release ↗