Ok. So I do get that. Its the .003 of extra bullet length is inside the case now. Which means increased pressure in that one cartridge, no? May be not enough to make a difference, but internal capacity shrinks on that one round, technically?
I also carry out seating depth testing in .003" increments, usually from ~.003" to .027" or .030" off the lands. I have rarely found it necessary to seat bullets into the lands. During such a seating depth test with jumped bullets, it is not at all uncommon to see anything from mediocre to completely unacceptable grouping across a good part of the test window, then all of a sudden you hit the "sweet spot" and the next couple or three seating depth test increments show one ragged hole. So even a very small change in seating depth can make a huge difference in precision. Although I guess you could refer to bullet holes out of the group during seating depth testing to be "fliers", in my mind the term flier means something very different than the typical shrinking and opening up of group spread observed during a seating depth test.
As Bart noted, what I was referring to earlier was the difference in the amount of bullet shank in the neck in a typical seating depth test, which will differ by whatever test increment has been chosen, even with bullets that have been sorted BTO. Although I was trying to convey the notion that the magnitude of velocity (pressure) change over a
total seating depth test range of .020" to .030" with jumped bullets is usually so small that a typical chronograph can't detect it, the concept is just as valid (if not more so) for a single .003" seating depth increment.
The key point here is that you can't detect a significant change in pressure or velocity for a .003" difference in the amount of bullet shank in a case neck without specialized equipment, and maybe not even with it. If you can't even reliably measure a difference in velocity/pressure, then it's unlikely that any beneficial effect gained by sorting bullets BTO is
caused by generating more uniform velocity/pressure. Therefore, it's likely due to something else. As I mentioned previously, I would speculate that it has more to do with friction/bullet release and/or improved external ballistics. However, there certainly could be other explanations. For example, I have heard a few different theories as to exactly what is happening when we tune in a load using seating depth, but I'm not convinced that even that process is fully and completely understood. Sometimes we do things in the reloading process without necessarily being able to fully grasp or explain why they work; they just do, so we do them. In those cases we just have to accept that we may not fully understand what is happening. In the the case of the OP's question however, I think it is unlikely that the exceedingly small change in the consistency of pressure gained by sorting of bullets by BTO is why people find it to be of benefit as a sorting technique.