Sequential Achievement and the Phase Transition in Outlier Productivity

Bessis observes that mathematical talent, as perceived by practicing mathematicians, follows a Pareto-like distribution with orders-of-magnitude gaps between individuals, and argues that this rules out genetic determinism as a primary explanation. However, the question of how a Gaussian input becomes a Pareto-like output is left as an informal conjecture. Taking Ramanujan as a motivating example, this paper proposes a quantitative mechanism. In Part I, we develop a method for inferring the parameters of a population’s ability distribution from the observable characteristics of an elite subpopulation and a single extreme observation, and show that the scaling character of the rank-#1 residual serves as a diagnostic that deduces the lognormal distributional family rather than assuming it by convention. In Part II, we model complex achievement as a sequential stochastic process requiring consecutive successes, where any single failure resets progress. We derive a sharp phase transition in expected completion time: below a critical score, completion time grows exponentially in the task depth; above it, completion time is approximately linear. This consecutive-success mechanism is the missing bridge: it converts modest, continuous differences in latent ability into differences of many orders of magnitude in realized output, producing the Pareto-like distribution Bessis identifies. We propose a realist interpretation of the underlying score as divergence resistance—the absence of a common tendency to extrapolate inferential structures incorrectly—inverting the standard framing by treating the outlier not as possessing a rare gift but as lacking a common impediment. We calibrate the model empirically and find that the framework simultaneously demystifies the outlier, who is a draw from the same distribution as everyone else, and validates their irreplaceability, since the exponential barrier facing lower-quantile individuals is a structural impossibility, not a matter of degree.