I think this has to do with the order you purhcased a component. Earliest serial number stacks highest in the list -.-
Wow . . . you guys with all your math, and your technical comparisons, and the way you evaluate everything . . . makes me with my hands-on approach feel like a noob!
I do both. Things get more complicated after the quirkening though. Before you can at least count on things like heat, dps and weapon cooldown being identical accross mechs if you have the same DHS and setup, so all you need to test/tweak are focused on chassis variant specific things (hardpoint positions and speed and angles, etc), now they can be quite different. That makes tuning new mech config a bit more complex. This is especially true when it comes to heat and Ammo. You can chew up way more ammo with some 25% cooldown boost, and with that you also need more DHS to keep up with it if the chassis does not come with enough heat bonus.
This but no its easier in the lab imo as nowadays its always in a preset order so for example if you have 6 ERML they're in that order in the lab nowadays. contrast that to in the training grounds where on some mechs they're almost totally random.
In all honesty pre-quirks I just went with it in pug games unless it was a weapon load I really wasn't used to. When levelling mechs I'm almost always doing them in sets of 3+ and I want them to be distinctively varied. Running a different setup on every variant in a chassis means at times you're just not going to have the optimal build for your playstyle so you're trying to push yourself in a build. Post quirks I do test heat in the practice grounds as things vary so much mech to mech. Clan mechs I still don't bother as not much has changed. the one exception is where a mech is really struggling and I need to work out exactly where I can fire to not shut down - Mist Lynx's and Ice Ferret's were classic examples.