The Magic Number: When Being First Doesn't Mean Forever First ✛ ✎ ✦

Original: "Universal fluctuations of first discoveries in competitive exploration"
arXiv:2607.04252 · 2026-07-05 · arXiv non-exclusive · independent analysis · 1 min
The spectral dimension of space determines whether the advantage of the first to grab a resource will stick.

Two tourists on a beach. If the beach is a narrow strip, the one who steps ahead grabs almost everything. If it's a wide field, both find a spot. Physicists discovered that the outcome of any race is predetermined not by speed, but by the geometry of the arena. This invisible property — the spectral dimension: it shows how entangled the possible paths are. In a corridor world (dimension less than 2), the initial advantage never fades: uncertainty is eternal. In a three-dimensional environment (dimension greater than 3), fairness arrives swiftly, like a coin flip, and the expansion of possibilities levels the playing field. In between, the wait is infinite. Modeling showed: in one-dimensional space, in about every seventh case, one gets less than 40% — such is the certainty. Thus increasing disorder doesn't always smooth out inequality; in cramped worlds, it cements it. This is a general law for bacteria, patents, algorithms. The discoveries of Einstein and Markov confirm: even random movement obeys geometry.

In a one-dimensional race between two equal 'invaders', in nearly every seventh run, one ends up with less than 40% of the resources — the inescapable injustice of geometry.

\text{Var}(X_n) \sim \begin{cases} \text{Var}(X_\infty) > 0, & d_s < 2 \\ (\log n)^{-2}, & d_s = 2 \\ n^{-(d_s-2)}, & 2 < d_s < 3 \\ (\log n)/n, & d_s = 3 \\ 1/n, & d_s > 3 \end{cases}
Shows how quickly fluctuations die out as the number of discoveries n grows. For d_s < 2, the variance does not decrease—inequality persists forever. For d_s ≥ 3, the decay follows a 1/n law, as in ordinary Gaussian statistics.
Scientists
Claude-Louis NavierJean LerayEberhard HopfOlga LadyzhenskayaLuis CaffarelliRobert V. Kohn
Concepts
Gause's competitive exclusion principle trials factor Fisher's fundamental theorem of natural selection density fluctuation look elsewhere effect drug discovery initial conditions random walk non gaussianity quantum fluctuation inflation prior distribution variance disjunction effect standard deviation probability distribution measurement uncertainty sample size equipartition theorem random variable founder effect