Thesis Drift

Thesis drift is what happens when a trade's original reason for existing silently changes without the trader explicitly deciding to change it. The position stays open, but the justification has shifted — often from a specific, falsifiable thesis to a vague hope that it will eventually work out.

The term appears in art-of-trading-with-light-su-zhu-and-hasu as a specific failure mode: a trade entered with a clear thesis and defined invalidation point gets held past its invalidation, with the trader unconsciously substituting a new, weaker justification rather than admitting the original thesis failed.

Why It Happens

Thesis drift is a form of loss aversion operating through rationalization. Admitting a thesis is wrong means realizing a loss. Instead of doing that, the mind generates a new reason to hold — one that is conveniently less falsifiable. "I'm early, not wrong" and "the market hasn't understood the narrative yet" are common drift phrases.

The related failure mode is timeframe conflation: a trade entered as a short-term momentum play gets silently reclassified as a long-term investment after the short-term thesis fails. The position hasn't changed; only its justification has. Light's formulation in the source: a trader enters a 1-week trade, it doesn't work, and they say "actually, this is a 1-year trade." The decision to extend the timeframe was never consciously made.

The Fix

Explicit thesis documentation at entry forces the question at exit: is the original thesis still intact, or has it drifted? A trade journal with a written entry thesis, invalidation condition, and target creates a checkpoint that drift cannot quietly bypass.

decision-quality-vs-outcome frames this more broadly: the quality of a decision is judged by the process at the time it was made, not by the outcome. A position held past its invalidation on a drifted thesis is a bad decision regardless of whether price eventually recovers.