Forensic investigators commonly use simulants/backing materials to mount fabrics and/or garments on when recreating damage due to stab events. Such work may be conducted in support of an investigation to connect a particular knife to a stabbing event by comparing the severance morphology obtained in the laboratory to that observed in the incident. There does not appear to have been a comparison of the effect of simulant type on the morphology of severances in fabrics and simulants, nor on the variability of simulants. This work investigates three simulants (pork, gelatine, expanded polystyrene), two knife blades (carving, bread), and how severances in the simulants and an apparel fabric typically used to manufacture T-shirts (single jersey) were affected by (i) simulant type and (ii) blade type. Severances were formed using a laboratory impact apparatus to ensure a consistent impact velocity and hence impact energy independently of the other variables. The impact velocity was chosen so that the force measured was similar to that measured in human performance trials. Force–time and energy–time curves were analysed and severance morphology (y, z directions) investigated. Simulant type and knife type significantly affected the critical forensic measurements of severance length (y direction) in the fabric and ‘skin’ (Tuftane). The use of EPS resulted in the lowest variability in data, further the severances recorded in both the fabric and Tuftane more accurately reflected the dimensions of the impacting knives.