After many years in the lab where my tasks were to try and manage statistically chaotic topics like friction, wear, fatigue, and LOS controls, I have seen many strange statistical events, both good and bad. Very early on, I started using Bayesian stats to protect myself from false positives and false negatives. When your sample sizes are very low, those things creep into your life just to keep you humble....
I wasn't directly responsible for internal ballistics, but since my work depended on theirs, it was in my interest to help keep things under control.
It is nice when one sweep for charge, and another for depth gives us a successful load, but that doesn't happen for most contexts. When it does, the usual context is a very high quality bbl, very mature loading discipline, and the hands of a master or high master pilot.
My guess is you are bitten by undersampling, but that is a guess from the bleachers.
If this is a brand new bbl, then give yourself a chance to learn what it wants before you draw any conclusions.
Try and work from a single contiguous batch of prepped brass, primers, bullets, and powder lots to keep those variables minimized. I have seen folks get confused when their components or brass prep caused changes they didn't know about between sessions.
Go back and see if the previous (good or bad) results repeat with and without a cleaning after you check all your fasteners and scope, etc.
Scott's method is good for some contexts, but not for all. Is this a match gun with a known background or a brand new unproven rig, or something in between?
If your bbl has a known pet load, go back and test with that too and see if that load is acting normal.
If you are in a new bbl, that isn't helpful yet but is something good to keep in the debugging toolbox for times when something acts out of family as you test new recipes.
Maybe give us a better description of the rig? Are we talking about a heavy match bbl or a sporter?
Is your bbl brand new? Did you have to break into any new lots on components between sessions? Did you change your brass prep between the two sessions? Change tools? If the answer to all of these is no, then you will have to tray a larger sample to get confidence in the results. If the answer to any of those is yes, then it may or may not be the cause, you will have to test again anyway so test those changes when you can to see if they are significant contributors.
I would double check the fasteners and scope the bore after a good cleaning. Starting a new bbl or new recipe means you may need a new baseline to see when the rig needs a cleaning.
Pay attention to your brass prep details and don't give up just because things didn't go as easy as we see on Youtube. Most of us have to burn some energy to fight headwinds in our sports, so the quicker we learn the faster we climb.
