DNS Research

Trend Following Performance: Historical Returns and Drawdown Analysis

advertisement

Historical Return Profiles of Trend-Following Strategies

Trend following is among the most extensively documented systematic strategies in financial literature, with return data stretching back over a century through managed futures programs and, more recently, transparent index proxies. The historical return profile of trend following is characterized not by equity-like compounding at a steady rate, but by episodic, fat-tailed return streams punctuated by prolonged flat or negative periods. Academic studies—including seminal work by Hurst, Ooi, and Pedersen on time-series momentum, and Moskowitz, Ooi, and Pedersen’s 2012 paper “Time Series Momentum”—document that diversified trend-following programs have historically delivered long-run returns comparable to equities with substantially lower correlation to traditional assets. The SG Trend Index, the most widely cited benchmark for the strategy, illustrates the pattern: strong performance in sustained macro trends (the 1980s disinflation, the 2008 financial crisis, and the 2022 inflation shock) interspersed with multi-year stagnation (2011–2014 and 2016–2018). BarclayHedge’s managed futures database shows similar cyclicality across four decades of live and simulated track records. Critically, returns are not normally distributed: trend followers exhibit positive skewness and convexity, meaning they tend to perform best during sharp market dislocations—precisely when equity portfolios suffer most.

The Crisis Alpha Phenomenon and Correlation Dynamics

The concept of “crisis alpha”—coined by Kathryn Kaminski—describes trend following’s tendency to generate positive returns during extended equity market drawdowns. Empirical analysis of the SG Trend Index versus the S&P 500 Total Return Index demonstrates that trend-following correlations to equities are state-dependent: near zero or slightly positive in calm markets, and strongly negative during sustained equity bear markets. During the 2008 Global Financial Crisis, the SG Trend Index returned approximately +20% while global equities fell roughly 40%. In 2022, as the Federal Reserve aggressively raised rates, trend followers captured powerful short bond and long commodity trends, with the SG Trend Index gaining roughly 27% against a -18% S&P 500 return. This convexity is structural, not accidental: trend systems reduce exposures as prices fall and flip short, mechanically converting prolonged declines into profitable positions. Regression analysis across multiple studies shows trend-following returns load positively on market volatility and negatively on flight-to-quality episodes, giving the strategy genuine diversification value within multi-asset portfolios. However, this same convexity means correlations can spike positively during sharp V-shaped reversals, when whipsaws hurt performance precisely as equities rebound—a dynamic observed in late 2020 and early 2023.

Drawdown Characteristics: Depth, Duration, and Recovery

Drawdown analysis is where trend following’s risk profile diverges most sharply from conventional assets. The strategy historically exhibits frequent shallow drawdowns but occasional deep, prolonged ones. SG Trend Index data since 2000 reveals maximum drawdowns in the 15–25% range during the worst episodes—notably 2011–2014, when central bank intervention suppressed volatility and trend persistence, and the index suffered a drawdown exceeding 20% lasting over three years. The 2016–2018 period produced another extended flat-to-negative stretch as markets whipsawed on trade-war headlines and volatility spikes without sustained direction. These drawdowns are qualitatively different from equity bear markets: they stem not from crash risk but from “trendless” regimes—choppy, mean-reverting markets where signals repeatedly reverse. Recovery periods can be long; the 2011–2014 drawdown required roughly two years of subsequent strong performance to fully recover. Quantitative analysis using the Calmar and MAR ratios shows diversified trend programs typically achieve superior drawdown-adjusted returns relative to long-only equity over full cycles, though the path is uncomfortable. Crucially, drawdowns tend to cluster when cross-asset correlations rise and volatility regimes shift abruptly—conditions that also challenge risk-parity and other diversifiers simultaneously.

Volatility Scaling, Sharpe Ratios, and the Role of Position Sizing

Historical Sharpe ratios for institutional trend-following programs typically range from 0.4 to 0.8 gross of fees, with diversified multi-market programs at the higher end. Volatility-targeting overlays—scaling positions inversely to realized volatility—have become standard since the 1990s and materially improve risk-adjusted returns by stabilizing portfolio risk across regimes. Analysis of the SG Trend Index shows realized volatility clustering near a 10–12% target, with return-to-volatility ratios improving when volatility scaling is applied versus static sizing. However, volatility targeting introduces its own drawdown dynamics: in sudden volatility spikes, deleveraging can lock in losses and delay recovery. Research by AQR and Man Group demonstrates that trend-following Sharpe ratios are positively related to the breadth of markets traded—programs diversified across 50–100+ futures markets (equities, rates, FX, commodities) consistently outperform concentrated programs on a risk-adjusted basis. Fee structures matter enormously: gross index returns of 8–10% annualized historically translate to net returns of 5–7% after typical 2/20 fee arrangements, a drag that compounds significantly over decades and explains the persistent gap between academic backtests and investor realized returns.

Regime Dependence: When Trend Following Thrives and When It Struggles

Trend-following performance is fundamentally regime-dependent, and historical analysis identifies clear patterns. The strategy thrives in environments of sustained macroeconomic trends—the 1970s inflation (commodity trends), the 1980s disinflation (bond trends), the 1990s dollar and equity trends, 2008 (broad risk-off), and 2022 (inflation and rate normalization). It struggles in mean-reverting, low-volatility regimes with frequent policy intervention: 2011–2014 saw central banks suppress rate volatility and “buy the dip” dynamics neutralize equity trends; 2016–2018 featured whipsaw markets around political events; 2023–2024 saw choppy, event-driven reversals. Research from Westbeck and Newfound Analytics quantifies regime sensitivity: trend-following returns correlate positively with the absolute magnitude of macro surprises and with realized volatility persistence, and negatively with central bank balance-sheet expansion that compresses term premia. The strategy’s returns also exhibit seasonality and autocorrelation patterns—momentum crashes, analogous to equity momentum crashes, occur when crowded trend positions unwind rapidly, as in March 2020’s initial chaos before trends reasserted. Understanding these regime dependencies is essential for setting realistic return expectations and sizing allocations appropriately within diversified portfolios.

advertisement

latest posts

Something went wrong. Please refresh the page and/or try again.