OCD is catching up with me. I'm sorting 2000 bullets.
Only heard a few sentences on this topic - never done it before.
So all I do is put two comps on my caliper, line them up as best I can, designate a golden bullet to be zero length, start measuring, and group them into +.002", -.005", etc.
Simple as that? Nothing else to it?
Thanks!
Regardless of the component, sorting is a process that can pay dividends IF the item being sorted has sufficient variance to warrant sorting. In the case of bullets, what that typically means is that commercial production bullets will likely benefit from some kind of sorting process, whereas with high end custom bullets it may not be necessary.
If you're going to attempt sorting bullets, the best place to start is by taking measurements from 5 or 10, to as many as 20 randomly selected bullets from a single Lot, including OAL, BTO, and bearing surface measurements. Once you have these measurements in hand, look to see which one exhibits the greatest variance as a potential sorting guide. For example, the many Lot #s of Berger bullets I have used over the years seem to have very consistent BTO and bearing surface dimensions within a single Lot. Thus, sorting by BTO or BS would not be productive for those particular Lot #s of bullets.
On the other hand, OAL for the same bullets typically varies about .015" to .020" or so within a given Lot #. There are a couple pieces of information that can be gleaned from this observation. The first is that length variance within the Lot #s of Bergers I have been using largely resides in the nose region of the bullets. This is important for a couple reasons. First, seating depth consistency is dependent on the distance between two critical contact points; i.e. the point out toward the bullet meplat where the seating die stem contacts/pushes the bullet, and the point where the caliper insert seats when we measure CBTO of a loaded round. Both of these points lie within the nose region of the bullet, so length variance in that region can affect seating depth consistency. I also point bullets, which means that producing consistent points requires minimal OAL variance. For these reasons, I sort bullets into overall length groups of .0015". That way, each length group will differ from the adjacent groups by .002" overall, but there should only be .0015" length variance within a single length group. This is sufficient to ensure very uniform points without having to change the pointing die micrometer when pointing bullets from within a single length group. Thus, with the specific bullets I'm using, sorting by OAL improves consistency of both seating depth and bullet pointing.
For the OP - in order to determine how sorting might best benefit your needs, you need to have some feel for the dimensional consistency of the bullets you're using, as I mentioned above. It would be better to determine how the bullets you're using stack up before arbitrarily selecting some parameter by which to sort them. If that seems like too much work, you can always start out sorting solely by OAL. Because all three bullet regions (nose, bearing surface, boattail) each contribute to OAL, sorting by OAL can help improve consistency regardless of where the greatest length variance lies within the bullet. However, sorting by OAL would only be an indirect method for specifically improving length variance within the BTO region, for example. So knowing where the greatest variance lies within the bullets you're using is usually the place to start.
Another important question is how do you expect to benefit from sorting? In other words, what exactly are you trying to accomplish by sorting bullets? In addition, how can you test to determine whether the sorting process you choose actually provides some benefit? For example, I have tested otherwise identical loads on numerous occasions where the only difference was using either bullets straight from the box, or bullets I had length-sorted and pointed. From these tests I know conclusively that length-sorted and pointed bullets generate tighter groups at distance (i.e. 300+ yd), and improve the bullet BC by about 3 to 6% as evidenced by the reduction in scope elevation adjustment required to keep the group center point at the same height on the target at a given distance. At 300 yd it is fairly easy to distinguish these differences (group size and elevation adjustment), whereas I cannot readily tell the difference between the same load with sorted/unsorted bullets at something like 100 yd. The groups simply aren't large enough at that distance to clearly reveal differences between the two in my hands. So it's important to know what you hope to achieve by the sorting process, as well as how you're going to test to determine whether there is actually a quantifiable benefit from whatever process you choose.