Endowment, Trends, and Moves

Definition

Analyzing 40 candidate variables across 2,393 large companies over 15 years, the authors found 10 that dominate: three describe a company's starting endowment (size, debt level, past R&D investment), two describe the industry and geographic trends it rides, and five describe the moves it makes (see Five Big Moves). Each variable only helps once a company clears a specific relative threshold against peers — "getting one point better doesn't seem to do much; you need to put yourself in another league" — and the resulting model predicted which third of the Power Curve a company would occupy a decade later with 86% accuracy.

In the Book

Chapter 5 reframes the "average" 8% odds of jumping from the middle to the top of the Power Curve into company-specific odds, using a Bayesian analogy: knowing only that someone is "a person" gives you the global average income, but adding nationality, age, gender, and occupation narrows the estimate sharply — the same logic applied to the 10 variables turned an 8% base rate into individually modeled odds of 69%, 87%, or 76% for three real companies (Dai Nippon Printing, UNFI, and Precision Castparts), each of which then moved as predicted. The resulting scorecard is called a "mobility dashboard" and is illustrated later with Harris Corp, which ticked seven of ten boxes (favorable trend, low debt, high R&D, and four of the five big moves) and carried an 80% modeled probability of a middle-to-top jump, which it then achieved.

Why It Matters

Turning "improve your strategy" into ten scored, threshold-based variables lets a team see its actual, calibrated odds of success before committing resources, rather than relying on confidence built from a well-rehearsed internal narrative. The general move — decompose an outcome into a small number of empirically validated predictors, each with a real threshold rather than a vague direction — is exportable to any domain where people mistake "more effort in the right direction" for "enough effort to change the outcome."