I've been thinking about ways to predict, model and or quantify terminal performance on game. I'm not much of a statistician, nor am I great with test designs, and I'd like some input. I was thinking about utilizing a survey, with some very specific questions regarding entrance and exit locations, bullet impact velocity, and a GPS measured distance travelled, maybe rifling twist, not sure what other categories, but maybe you can help me with that.
Just a guess, but it would probably take a hell of a lot of input to get anything meaningful or statistically significant from this. Something on the order of several thousand samples.
Just figuring out what questions to ask to narrow down all the variables is pretty daunting though.
Among the things you'd need to specify:
Entrance and exit location-if sticking with broadside and quartering shots, I think you'd need a way to break up the torso into sectors based on horizontal distance betwen the point of shoulder and last rib and vertical distance between the bottom of chest and top of whiters.
Bones hit-you'd probably have to specify whether ribs, humerous, scapula or spine was hit on entrance and exit.
Also might have to specify whether the heart was hit.
Animal variables-while mammals all have roughly the same skeletal structures, there are probably constitutional differences among species, sex, and size which would affect terminal performance. Rather than required animals to be weighed, you might be able to get away with a chest girth measurement.
Projectile variables- Specifying Make, model, weight and caliber of the bullet probably isn't enough. Twist might matter, velocity will almost certainly matter, so knowing the exact distance to the animal will matter, as will knowing the muzzle velocity, which would likely rule out factory load users who haven't chronographed the load in their own rifle because, hey, maybe 100 fps does make a statistical difference?
Ideally, you'd be able to lump a lot of bullets into similar categories, like the bonded, tipped boattails, the tipped monometals, partition-style, etc to increase your bullet sample sizes. But due to jacket variations, bonding processes, material differences, that's probably not possible.
Shot reaction: don't how else to measure effectiveness other than to measure distance travelled from site of impact. And even then, how are you going to do that? A GPS measured distance will give a straight line distance, but can't account for any zig-zagging (which they don't do much of anyway in my experience.)
What else am I missing? Obviously, pulling off a meaningful study would be difficult due to the sampling size and variables involved. But I'm not certain this is impossible to pull off.