How often does theory match experiment?

TL;DR

This paper investigates how often and under what statistical conditions theoretical predictions match experimental results across various sciences, focusing on the distribution of these occurrences over time.

Contribution

It introduces a statistical framework to characterize the distribution of experimental observations matching theoretical predictions, highlighting the randomness and interval-based nature of these matches.

Findings

Theoretical predictions often occur within specific intervals of experimental results.

The distribution of matching events can be characterized statistically.

Matches between theory and experiment are generally random over time.

Abstract

In every sphere of science, theories make predictions and experiments validate them. However, common experience suggests that theoretically predicted exact magnitude for a parameter, constitute a small subset of all the experimentally obtained magnitudes for that particular parameter. Typically, irrespective of the branch of science and the particular problem under consideration, the set of obtained experimental results form an interval [x_min,x_max], within which the theoretically predicted magnitude, say 'x', occurs with time, apparently randomly. We attempt here to find the characteristics of the statistical distribution of events of experimental observation of the occurrence of theoretically predicted 'x'; in other words, characterization of the time interval when theoretical predictions match the experimental readings, exactly.