How a Ball Bounces: A New Formula for Any Material express

Original: "Exact Phase-Space Analytical Solution for the Power-Law Damped Contact Oscillator"
· Y. T. Feng
arXiv:2603.27764 · 2026-03-29 · CC0 1.0 · 2 min
Scientists have found a way to precisely predict how high a ball will bounce, knowing only its material and impact speed.

A ball drops to the floor, squashes for an instant, and flies back up. How much energy it returns depends on the material: rubber bounces back to nearly the same height, while a lump of clay just splats and stays put.

Engineers used to calculate such bounces accurately only for two idealized cases: when the ball behaves like a spring (compression force grows proportionally to squish) or like a Hertzian steel ball (force grows faster). For the full variety of real materials (foam, plastic, living tissue) they had to rely on approximations.

The new work provides a general recipe. The authors noticed that if you replot the graph of ball velocity versus compression in special coordinates, any material turns into an ordinary spring with friction. You just need to choose the right friction coefficient in this imaginary system. Then the bounce is computed with one short formula that involves only two quantities: material stiffness and its "stickiness" (energy lost during compression).

The big surprise: the fraction of returned energy does not depend on how fast the ball hits the floor. A hard-thrown rubber ball bounces back with the same energy fraction as a gently tossed one. Previously, this was rigorously proven only for a spring; now it holds for any material.

The practical benefit is direct: computer games and engineering simulations can compute collisions faster and more accurately. Measure the two material constants once, and the machine will honestly predict the bounce at any speed.

The formula also tells you what time step to use in numerical calculations so the model doesn't "explode." The faster the ball flies, the smaller the steps must be, and this constraint is now computed instantly.

If you compress a rubber ball to half its diameter, the compression force for many materials grows not linearly but as the square of the compression. That's why the general formula uses an unusual change of variable, as if we're looking at the world through a magnifying glass that straightens curved graphs.

\alpha = \sqrt{2(p+1)}\cdot\frac{-\ln e}{\sqrt{\pi^2 + \ln^2 e}}
Here, α is a measure of energy loss in the material, p is the exponent in the compression law (p=1 for a spring, p=3/2 for Hertzian contact), and e is the fraction of velocity returned after impact (coefficient of restitution). The formula lets you find α from a measured e, then predict the bounce for any speed.
Scientists
Andrey Markov
Concepts
nonlinear dynamics Stochastic modeling