Quants that choose empirical risk models typically seek the benefits of adaptiveness. Theoretical risk models are relatively rigid, meaning that the risk factors are not altered often (otherwise the theory would not have been very strong in the first place). Yet the factors that drive markets do change over time. For a while in early 2003, daily reports about the prospect, and later the progress, of the U.S. invasion of Iraq drove stock, bond, currency, and commodity markets almost singlehandedly. More recently, in early 2008, commodity prices were a significant factor. At other times, expectations of how much the Federal Reserve might cut or raise rates are the key drivers of market behavior. As markets evolve, the data that the markets produce reflect this evolution, and these data drive empirical risk models. For these reasons, an empirical model may be more adaptive to ever‐changing market conditions by detecting through new data whatever factors are implicitly driving markets. There are two stages to this adaptation. During the early phases of a market regime change (for example, when equity investors rapidly change their behavior from risk seeking to risk aversion), the quant is using now irrelevant historical data to determine relationships and measure risk factors. Thus, during this phase, the empirical risk model will be modeling market risks incorrectly. Later, if the new behavior persists, the empirical risk model eventually will catch up to the newly prevailing theme driving markets, and all will be well again.