Shots with two different calibres (0.32 Auto and 9 mm Luger) were fired through phantoms that simulated human torsos, mounted on undercarriages with witness panels. The perforated phantoms were scanned with computed tomography (Siemens) using 80 kV and 140 kV and a slice thickness of 1 mm. The intracorporeal trajectories in the phantoms were compared to the known extracorporeal trajectories, derived from the perforations in witness panels. The discrepancy between the intracorporeal and extracorporeal trajectories, denoted as the absolute angle, was calculated for the trajectories before (front) and after (rear) the phantoms. Mean absolute angles at the front were lower than at the rear (2.27° vs. 4.54°) and the difference was statistically significant (p < 0.001). The results of the study imply that the line between the entrance and the exit wound in a scanned victim can be extended to the extracorporeal bullet trajectory leading towards the entrance wound. The absolute angles presented in this study give an impression of the expected errors with the two calibres. This can be helpful in shooting investigations to assess the position of the shooter from entrance and exit wounds in a scanned victim.