[Jeff Higgins]

Interesting, but an awful lot of specific to his discipline "engineering speak". He may have written this for other ballistics engineers, but he certainly did not write it for lay people, or even exceedingly advanced handloaders. At that level, this paper is actually pretty useless, save for one sentence in his conclusion: These results conclusively show that propellant selection can indeed affect dispersion.

Ammunition companies and handloaders have understood this simple fact for as long as we have been making firearms go "bang". Even back in the muzzle loader days, with black powder being the only available option, shooters understood that their choice of powders would affect the accuracy and velocity they could expect from their firearms.

Smokeless powders, with their far greater variety of burn rates, only served to complicate matters further. Their introduction increased the need for us handloaders to "work up a load" in any given firearm. The factories do not have that need, or that luxury, or however you want to view it - their loads must work well and be safe in all firearms chambered for any given cartridge. As such, factory loads will always be a compromise, with the only thing they can really quantify being safe pressures and adequate velocities from their test barrels.

"Working up a load" is the fun part for those of us who handload. Depending on what we will be doing with any given firearm and load, priorities will vary. For me and most of my guns, it's all about hunting performance, so I typically choose the bullet first, and will give up some accuracy and maybe even a wee bit of velocity to use that bullet. Next it has to chamber and extract, fired or unfired, without a hitch. I will typically try three or four different powders that several loading manuals agree might provide top velocities. Each powder selection is first loaded with the lowest listed charge weight, a "starting load", shot for accuracy over a chronograph so I get velocities as well. I creep up incrementally, one grain of charge weight at a time, shooting for group size and velocity until I find a combination that I like best. It may not always be the most accurate, it may not always have the top velocity, but it will always be the best compromise of many factors. These are hunting loads, so gilt edged accuracy is not as critical as in target loads.

Target loads are far more difficult and time consuming. Here, in addition to what I do to work up a hunting load, I add several bullets, even several bullet weights to the mix. Seating depth then becomes much more important, with small variations having a dramatic affect on accuracy, where hunting loads we just need to fit the magazine and chamber and extract easily. I usually test more powders as well, further adding to the time commitment. Things like chambering and extraction may not be as important depending on the discipline. Sometimes benchrest loads will have to be fired to be extracted without sticking the bullet in the throat and dumping powder down into the action, for example. "Over the course" NRA High Power loads cannot be assembled this way, but they do chamber much more tightly than pure hunting loads.

So, yes, it has been very well understood for, dare I say, at least a couple of centuries that powder selection will affect group size. In some rifles group size will vary dramatically, but that is often an indication that there might be something wrong with the rifle.