Random points, stable twist. An Evidence Press briefing. Fireflies flashing together, pendulum clocks falling into step, generators on a power grid, neurons firing in rhythm: all are examples of synchronisation. The Kuramoto model is the standard mathematical picture of it. Each oscillator has a phase, a position around a circle, and it keeps nudging that phase towards the phases of its neighbours. Now scatter many identical oscillators at random over a flat torus, which you can picture as a square whose opposite edges are glued together, in any number of dimensions. Connect any two that are closer than a small distance. One stable outcome is that everyone agrees. Another is a twisted state: as you walk across the torus, the phase turns steadily round the circle a whole number of times, like a spiral wave frozen in place. On a perfectly regular grid, symmetry makes such patterns exact. With random positions, every oscillator feels a small unbalanced pull, and it was not clear that a stable twisted pattern survives. In 2025 Cirelli, Groisman, Huang and Vivas conjectured that it does. This paper proves their conjecture for every coupling in their class whose strength grows no faster than one over the distance. Once the number of points is large enough compared with the connection distance, then with high probability every twist up to a growing limit has a stable equilibrium close to the perfect pattern, with an explicit bound on how far the random positions bend it. So many different stable patterns coexist at once. The key tool is a new uniform estimate for the graph Laplacian. Couplings that blow up faster remain open, the constants are not explicit, and the numerical examples are only small illustrations. This is an anonymous, AI-assisted, unrefereed candidate, not peer reviewed or independently reproduced. Published on the fourth of November twenty twenty-six, its paper and evidence are linked from the Evidence Press release page. This briefing uses an AI-generated voice and is not additional scientific evidence.