The Residual

What this pack runs.

Every declining business has an agreed story, and it is nearly always real: a rate rise, a policy change, a pandemic, a competitor, a regulator. It is also nearly always a trigger rather than a mechanism — the decline started earlier, ran longer, and was visible in the company's own reporting before the named cause existed. This pack measures the difference. It takes what fell before the cause began, fixed by the dates; bounds the cause at its arithmetic ceiling rather than at an estimate of its effect; and returns the residual — the share of the fall the explanation cannot account for even at its most generous — computed a second time from comparable businesses, on evidence entirely independent of your own accounting. It stops on two conditions: the decline has never been dated against the explanation, and nobody has said how large the named cause could possibly be.

The casebooks these cases come from.

The pack reads nine sourced decisions through the framework. They are drawn from these casebooks, each one a set of verified records on a single decision type.

A verified record has passed a mechanical claim-versus-source check, a second model's read, and human approval. The count is what exists today, not a target.