RedSkyFarm,
It doesn't always matter. If you read the article
Bugholes from a Bi-pod, the shooter there said he tried weighing .308 win cases and could never detect a difference in actual results, though I believe he did sort by headstamp. The reason is complicated. Brass lots come off mixed tooling. Even premium brass maker Lapua mixes brass that comes off different forming die sets that are being run simultaneously. No two die sets are a perfect match, but they do pretty well on capacity. What seems to vary more are the heads. Well, if one head is a little wider in diameter than another, that affects case weight, but doesn't affect powder capacity at all, and this is a lot of the weight variation you see.
If you start mixing headstamps, that's another matter. Not only can the head designs vary more, but they don't start out designed for the exact same capacity. But this brings in other problem factors like differences in brass alloy that can affect how fast they work-harden from resizing and how hard the neck grips the bullet. So I never mix headstamps just to keep the variables out of it.
Back in the 1960's, Wm. C. Davis, Jr. did some testing that showed .30-06 brass of different makes had capacity variation such that you needed to alter charge weight about 1 grain for every 16 grains difference in brass weight. But it's approximate. I measured case water overflow capacity and weight of some mixed headstamp .308 once and found that weight among those different headstamps predicted case water overflow capacity with an accuracy of ±20% of the predicted difference. So Davis's rule is going to have that same precision limitation.
Because of that last point on precision, I recommend you sort by headstamp at a minimum. If you are using military brass, you can also sort by year until you are satisfied they are all about the same. Even if you should find cases where weight corresponds to difference in internal capacity, it will take about ±1.6 grains brass weight to cause the same change in barrel time That ±0.1 grains of powder does. It's about the same effect changing ambient temperature 3°F has on a non-temperature compensated powder. So if you are in a situation where the temperature changes 10°F during the course of the shooting day, you would have to have brass ±4.8 grains (a 9.6 grain spread) to see it cause the same issue
IF the capacity is changing proportionally with the weight. But if you sort by headstamp that doesn't usually seem to happen much.