Yep!
One of the reasons there were thousands of gunsmiths all over the place, was that production guns were mostly crap! By todays production standards, in any case.
If you look into the manufacturing technology used in Winchester during essentially their 'glory days' most of the machinery was set up to make a single part or even a single cut on a part, using a master part set on to the machine to set the depths of cut.
There were very few actual measuring tools in use on the production floor, as the operators were there to feed the machines and pull levers, not to make changes above their pay grade. A few more skilled workers did the set-ups.
Looking at the drawings that the tool room used to make their masters from, for the Winchester 1885 High Wall rifles, tolerances of plus or minus 5 thousandths were pretty common, as were tolerances of several degrees. This has been one of the eternal thorns in the arses of the guys trying to reverse engineer the parts by copying a single example, as they have to guess what the dimension was supposed to be, based on the likelihoods for it's purpose, and the dimension of the finished part in hand.
In the case of measuring tools and repeatability, accuracy and precision, there are several large and dry books about the subject. Foundations of Mechanical Accuracy, by Moore is a pretty good one of you wish to read up on the stuff.
Testing any measuring tool on only full even increments of it's thread, rack, etc., opens you up to becoming a victim of cyclical errors that you have avoided. What I mean is, if you only check every even Inch, you do not know that there is not a fault with the rack or screw thread that manifests at a different portion of the reading, if that makes any sense. It gets covered a lot when discussing calibration and ensuring accuracy. An example would be if one tooth was bent on the pinion of a dial indicator. As the pointer goes around and around, there will be an area where it reads correctly, and a smaller area where the measurement is not reading accurately. The dimensions used to calibrate tools in a calibration center, are chosen to search those errors out.
Like I said before though, if you are not making parts for NASA and mailing them in to be measured there by other tools than the ones used for making them, you pretty much need to be reasonably sure that the parts you deal with are consistent and measure the same each time with the same tools. Having a few Standards around (accurately made items of known dimension) goes a long ways towards that peace of mind. measuring anything, several times in a row, and coming up with the same measurement, is a pretty solid way to add some confidence to your repertoire. It gets you used to the feel needed to get the best out of what you are doing, and most of the real world applications of measuring tools in our uses for stuff like reloading, the differences between very expensive tools and merely OK ones, are largely academic.
Cheers
Trev
