The reconstructed sequence first, then each driver’s own account played through the identical model — same masses, same impact geometry, same road surface. The only thing that changes is the speeds each account asserts. Where an account cannot put the vehicles where they were found, the gap is drawn and measured on screen.
Schematic rendering at 25 fps. Vehicles and the scene are drawn to scale but are deliberately not photo-realistic: the sequence illustrates the analysis, it does not stand in for a photographic record. 1800 frames, 72 seconds.
| Account | Speeds asserted (A / B) | A rest error | B rest error | Result |
|---|---|---|---|---|
| The reconstructed solution | 52 / 40 km/h | 0.1 m | 0.1 m | physically consistent |
| The account of the driver of vehicle A | 50 / 75 km/h | 1.0 m | 15.8 m | not physically consistent |
| The account of the driver of vehicle B | 52 / 40 km/h | 0.1 m | 0.1 m | physically consistent |
Tolerance applied: 2.4 m and 24 degrees, being the consistency threshold widened by the recorded uncertainty in the documented rest positions themselves.