The pre-mortem: how prospective hindsight improves investment decisions

In 1989, Deborah Mitchell, J. Edward Russo, and Nancy Pennington published "Back to the Future: Temporal Perspective in the Explanation of Events" in the Journal of Behavioral Decision Making. They asked subjects to generate reasons for a future event in one of two framings. The prospective group was told the event might happen and asked to list reasons. The prospective-hindsight group was told the event had happened and asked to explain why. The hindsight framing produced reasons that were measurably more specific and concrete than the prospective framing of the same event.

Gary Klein adapted the finding into a decision-making technique he named the "pre-mortem," published in Harvard Business Review in September 2007 under the title "Performing a Project Premortem." The procedure is mechanical: before committing to a plan, the team is told to assume the plan has already failed catastrophically, and each member writes down the reasons. Klein reports that pre-mortems consistently surface failure modes that the same teams missed when asked the standard question "what could go wrong?"

The reason the framing matters is that human reasoning about future events defaults to scenario-building from current evidence outward, which biases toward confirmation. Reasoning about a past event runs in the opposite direction: from a known outcome backward to its causes, which forces specific causal claims. The same brain produces different reasoning under the two framings.

Applied to an investment thesis, the pre-mortem replaces "what are the risks?" with "it is 18 months from now and this position is down 40%. Write the post-mortem." Investors who run this exercise produce risk lists that are sharper, more mechanism-specific, and harder to dismiss than the same investors produce under the conventional framing. The post-mortem narrative also tends to surface contradictions with other positions in the book, because the failure path of one thesis often passes through the success path of another.

The output of a pre-mortem is the raw material of an invalidation condition. The most plausible failure narrative names the specific observable event that would constitute the failure — and that event, written down before entry, becomes the trigger that ends the position before the loss compounds.

Tensile prompts a pre-mortem at thesis creation and persists the resulting failure narrative alongside the invalidation conditions it produces, so the reasoning that ended the position later can be audited against the reasoning that started it.