Lethal spring traps are widely used for killing small mammals in the UK. Many require government approval, based primarily on humaneness. However, mole traps and break-back traps for rats and mice are exempt; those available vary widely in price and apparent quality. The EU is considering implementing a Trapping Directive that would alter UK legislation, and a recent report advised the EU that trapping legislation should cover all trapped species and encourage improvement of traps. Mechanical trap performance is often used as an indicator of welfare impact. We examined the mechanical evidence for scope to improve the welfare standards of rat, mouse and mole spring traps. We measured mechanical performance among a range of rat, mouse and mole traps. Impact momentum values varied 6-8 fold, and clamping force values 4-5.5 fold, among traps for killing each species. There was considerable overlap in the performance of rat and mouse traps. Trap-opening angle and spring type were related to impact momentum and clamping force in traps for both species. There was no relationship between price and mechanical performance in traps for any species, except talpa mole traps. We are unable to judge the direct welfare impact of the traps tested, but rather the potential welfare threat associated with their exemption from approval. The wide variation in mechanical performance in traps for each species, overlap in performance between rat and mouse traps and increasing availability of weaker plastic rodent traps indicate considerable scope for improving the humaneness of spring traps for rats, mice and moles. We conclude that all such traps should be subject to the UK approval process. New welfare categories might improve trap standards further. Our results could also help improve rodent trap design and assist consumers in selecting more powerful traps. Many thousands of rats, mice and moles might benefit.