Unlike other sources of alpha, trend following is difficult to arbitrage away. The guest argues that as more people adopt the strategy, their collective actions tend to amplify and extend existing trends, creating a self-reinforcing dynamic rather than a diminishing one.
To combat unreliable backtests, CFM is building "meta-models" that quantitatively predict whether a new model's results are overfitted. This systematic approach aims to replace human judgment with a data-driven process for deciding if a trading model is robust enough for production.
Investors frequently give up on trend-following strategies after a few flat years, right before they rebound. This is attributed to a deeply ingrained behavioral bias to chase recent performance, which causes them to sell low and miss the subsequent recovery, ensuring they underperform the strategy.
CFM maintains a strong academic presence not just for research, but as a core talent acquisition strategy. By having its leaders publish papers and hold professorships, the firm attracts top-tier PhD talent who are already familiar with their work and view CFM as a destination for serious, cutting-edge research.
CFM operates on the belief that in the short-to-medium term (up to a year), market prices are driven primarily by investor flows, not fundamental value. This "inelastic market hypothesis" means their strategy focuses on predicting what people will buy and sell, rather than analyzing company balance sheets.
The 1987 market crash highlighted a critical flaw in the Black-Scholes options pricing model: it assumes a world without large, sudden crashes. This intellectual gap spurred Jean-Philippe Bouchaud to move into finance and develop new quantitative models that could account for these real-world "jumps."
To overcome the limitation of having only ~100 years of real financial data, CFM is exploring the use of Generative AI to create vast synthetic market histories. This would allow them to train and test their quantitative models on a scale of a "million years," making them more robust.
Even a highly systematic quant shop like CFM acknowledges the need for human intervention. For truly unprecedented events like the Brexit vote or sudden tariff announcements, the firm concluded its models were blind to the unique context, requiring a manual human judgment call to manage risk appropriately.
Physicist Jean-Philippe Bouchaud applies concepts from theoretical physics, like granular media, to finance. He views markets as complex systems where a small event, like a single grain of sand, can trigger a massive, unpredictable "avalanche" or market crash, a core idea behind CFM's quantitative models.