Equal source weights remove clear shuffle support for an echo-screen excess in 79 selected CHIME bursts

We compare a simple echo screen for 79 two-component CHIME bursts from 14 repeaters with different-event pairs from the same source. At three quoted errors, 36 real pairs pass (45.57%), against a 28.98% mean different-event fraction with event weights. Equal source weights give 39.45% and 38.67%, with a conditional shuffle upper tail fraction of 0.324. Removing FRB 20220912A removes clear shuffle support for the event-weighted excess. Instrument controls leave 37 bursts and give a time-group shuffle tail fraction of 0.0747. Shared fit errors, intrinsic emission, selected components and different observing conditions limit interpretation. The background is a compatibility fraction, not a lens false discovery rate. No lens or planet is found. Original code, aggregate evidence, source hashes, a figure and external regeneration steps are included. Raw catalog data and record-level outputs are excluded.

astronomy · CHIME · fast radio bursts · echo screen · source weighting · conditional shuffle · gravitational lensing