Every mental model you trust was built from evidence that no longer exists. The American pragmatist philosopher William James made a disturbing observation about this in his 1907 lectures on pragmatism: beliefs, once formed, become self-confirming — not because the world cooperates, but because we stop looking for disconfirming data once a belief feels settled. The model becomes the lens, and the lens filters out whatever would disturb it. What James couldn't fully anticipate was how this compounds in finance, where confidence in a model often grows precisely as the conditions that validated it quietly erode. The useful habit isn't skepticism about your models in general — it's identifying, for each model you currently rely on, what the original confirming evidence actually was, and whether that specific evidence-type is still present in the world today. Not the conclusion the model reached. The evidence that made the conclusion feel earned.

For your most-used financial model right now: what specific conditions made you originally trust it, and are those conditions still observable — or are you running on the memory of trust?

Drawing from American Pragmatism — William James (Pragmatism: A New Name for Some Old Ways of Thinking, 1907)