Interesting paper. My first thought at a glance is that it would really benefit from a bit more detailed kinematic modelling, for example considering the approx

Interesting paper. My first thought at a glance is that it would really benefit from a bit more detailed kinematic modelling, for example considering the approximately quadratic-in-velocity drag force (very important when talking about differing spear designs!) and computing approximate coefficients of drag. The main issue is that the force that muscles can generate is a function of velocity (higher contraction velocity, lower force) and hence the total specific impulse imparted to a spear is limited, but the power-time curve a human arm might be able to generate is complex and bounded within some physiological limitations. I'm sure you could go to town with this data (well, if you're an aerodynamicist, at least). Spears and the ilk are very nice shapes mathematically to play with.