Market Analysis

A signal that looks clean in backtest is a different object once it touches a real order book. This pillar covers what happens at that boundary: how trading on a signal moves the price against you, how an edge erodes as it gets crowded or arbitraged away, and how the liquidity available at the moment you trade determines whether a paper alpha survives execution.

It also covers the structural context that decides whether a signal is even measuring what you think. Performance metrics tell you how to evaluate a signal honestly; the high- vs low-frequency comparison frames the capacity-and-cost tradeoff that sets your operating regime; cross-asset correlation work shows where signals share a common driver versus carry independent information. Sitting over all of it is regime: the recognition that the relationship between a signal and forward returns is conditional, and that a metric estimated in one regime can invert in the next.

Read together, these pieces move from the mechanics of a single signal hitting the market, through how you measure and compare signals fairly, out to the regulatory and regime conditions that govern when any of it holds. The throughline is that live-market behavior, not in-sample fit, is what separates a tradable signal from an artifact.