This study describes an effective protocol for naturalistic archaeological weapons experiments that improves cross-validation with controlled experiments and allows testing of multiple hypotheses. Stone-tipped atlatl darts and arrows were launched by skilled users against fresh carcasses, with high-speed cameras and radar guns capturing details of ballistic performance, impacts to bone and stone armatures, and other variables. The results pertaining to terminal ballistics in soft tissues are presented, with implications for what made ancient hunting projectiles effective and can be observed archaeologically. Fine-grained knappable stones seem to produce sharper armatures that can dramatically improve penetration, and presumably, lethality. Two commonly used metrics by archaeologists for estimating armature efficacy, tip cross-sectional area (TCSA), and perimeter (TCSP), are not among the significant variables for capturing penetration depth in soft tissues. However, armatures with larger TCSAs tend to be fitted to larger shafts that carry more energy and penetrate more deeply, providing one method for predicting wounding potential. The variability within weapon systems means that isolating efficacy to individual variables, such as tip cross-sectional size of stone armatures, can lead to erroneous interpretations.