Trajectory deflections of pistol bullets from four different firearms, fired through soft tissue simulants under two different incidence and exit angles were studied. The data from this study can be used in reconstructions of shooting incidents where human soft tissues (not bones) were perforated with pistol bullets and assumptions must be made about bullet deflection in order to correctly reconstruct trajectories. The results demonstrate that deflection was influenced by the length of the "wound channel" through the simulants. In short, deflection was small to virtually absent with 5 and 10cm channel lengths. With channel lengths of 15, 20 and 25cm, there was a clear increase in deflection and/or a more erratic deflection behaviour with most shots. The data also suggest an influence of the angle of incidence and/or exit on both the direction and the magnitude of the deflection.