McNamara Fallacy

principle

The McNamara Fallacy is mistaking what can be counted for everything that matters. Quantitative data improves decisions only when interpreted alongside context, judgment, and relevant lived expertise.

During the Vietnam War, consequential decisions were made purely from the data by people who lacked actual experience of war. The numbers looked like a firmer basis for judgment; the missing experience was precisely what made that confidence dangerous.

When the measurement becomes the battlefield

A measurement captures a selected feature of reality, not reality whole. Once decision-makers treat the resulting dataset as complete, anything unmeasured—local context, tacit knowledge, or an expert’s sense that the situation has changed—effectively disappears. The failure is therefore not using numbers; it is allowing the numbers to define what counts as relevant evidence. Apparent objectivity then conceals a subjective choice about what was measured and what was excluded.

Where it shows up

War managed from the numbers

The Vietnam example exposes the gap between legible indicators and conditions on the ground: people without war experience could calculate from the available data while lacking the practical knowledge needed to interpret what those calculations meant.

The answer is not fewer numbers

This principle does not show that data is inferior to intuition or that experience is automatically correct. Judgment can also be biased. The boundary is narrower: data becomes hazardous when its omissions are forgotten and interpretation is treated as unnecessary.

Add an omission review

Before acting on a dashboard or score, ask the relevant practitioner to name one consequential condition the measurement does not capture and explain how it could change the decision. Keep the number—but record the missing context beside it.

Episodes that teach this

Things Go South When You Trust DATA Too Much - Goodhart's Law - FutureIQ 4,228 views
The Vietnam War example is described as decisions made "purely based on the data" by people without actual war experience.