TechnicalResearch

Why Trend Following Is Harder Than It Looks | Systematic Investor | Ep.405

Top Traders Unplugged

Rob Carver and Niels Castellen discuss trend following challenges, examining why chasing recent performance is problematic, the risks of AI-driven backtesting and overfitting, and analyzing research showing how different asset classes contribute to trend-following returns across market regimes.

Summary

In this episode of the Systematic Investor series, Rob Carver and Niels Castellen explore several critical topics affecting trend-following strategies. They begin with market updates and observations, noting that recent optimistic pricing of geopolitical events has already been factored into markets, and discuss concerns about data quality degradation due to government budget cuts. They address a listener question about portfolio optimization, with Rob arguing against using drawdown metrics and favoring simpler approaches like Sharpe ratio and skew for strategy allocation decisions. The hosts examine two research papers: the Alpha Architect piece on factor concentration, showing that despite hundreds of documented factors, only a handful of 'species' truly explain returns, and a Substack post on trend-following's equity risk premium exposure. They debate whether trend-following portfolios should be adjusted based on their net-long equity bias depending on overall portfolio context. A significant discussion covers the CFTC's approval of perpetual futures contracts in crypto markets—effectively daily-settling futures without expiration dates—and the CME's legal response, raising concerns about market structure, hedging functionality, and industry implications. The centerpiece is a detailed analysis of Quantica's research paper comparing trend-following performance across three decades, revealing that the 2010-2019 period dominated by fixed income was anomalous, while the current commodity-driven regime (2020-2026) represents a return to historical patterns of diversified sources of returns. Rob emphasizes that continuous trend-strength systems naturally allocate toward stronger trends without requiring dynamic regime shifts. Finally, they address the growing problem of AI-generated backtest results and overfitting, with Rob arguing that the research process and robust testing methodology matter more than whether AI generates initial ideas, and that due diligence should focus on process rather than tool-specific concerns.

Key Insights

  • Carver argues against using drawdown depth and duration as meaningful statistics for evaluating individual strategies because drawdowns scale predictably with strategy performance and backtest duration, making them redundant when Sharpe ratio is already evaluated.
  • The Quantica research reveals that trend-following returns concentrated in fixed income during 2010-2019 was a regime anomaly driven by predictable central bank actions, not diversification, suggesting this period does not represent the historical norm for factor contributions.
  • Perpetual futures fundamentally change market structure by eliminating the need to roll contracts and reducing arbitrage between futures and spot prices, but potentially remove the critical price discovery mechanism that hedgers (farmers, producers) depend on for risk management.
  • Carver contends that using AI to generate trading strategy ideas is acceptable, but the critical factor is the robust testing process that follows, which should remain identical whether the idea originated from AI, data mining, or chart observation—the source matters less than methodological rigor.
  • Continuous trend-strength systems already naturally position portfolios to benefit from stronger trends by asset class, making explicit dynamic regime shifts unnecessary and potentially harmful through overcomplication and overfitting risks.

Topics

Trend Following Performance DriversPortfolio Optimization and Strategy SelectionPerpetual Futures Contracts and Market StructureAI-Driven Backtesting and Overfitting RiskCross-Asset Diversification in CTA StrategiesResearch Methodology and Due DiligenceFactor Analysis and Market RegimesDrawdown Analysis and Risk Measurement

Transcript

[0:01] Welcome to Top Traders Unplugged. In markets, success doesn't come from predicting what happens next. It comes from being prepared for what you can't predict. In each episode, we go deep with some of the world's most thoughtful minds in investing, economics, and beyond to understand how they think, how they prepare, and how they decide, and the experiences that shaped how they see the world. No noise, no shortcuts, just real conversations to help you think better and invest with confidence. [0:35] Welcome or welcome back to this week's edition of the systematic investor series with Rob Carver and I, Neilles Castellen, where each week we take the polls of the global market through the lens of a…

Full transcript available for MurmurCast members

Sign Up to Access

More from Top Traders Unplugged

Get AI summaries like this delivered to your inbox daily

Get AI summaries delivered to your inbox

MurmurCast summarizes your YouTube channels, podcasts, and newsletters into one daily email digest.