Either .50 or .54 will work for a Revolutionary period rifle.
However, .50 is probably closer to the average Rev War caliber than is .54, IF we restrict our sample to guns that we know haven't been used hard (i.e., those that were brought to Europe and ended up in collections) and period sources (George Hanger, most notably). Distribution is something like two in the .60 caliber range (Schreit and Lion & Lamb rifle), .52 (one Oerter rifle), a .42, and four or five in the .45-.49 range, IIRC. Interestingly, the two really big ones are a rifle that was built just after the French and Indian War and had been in use for at least 14 years prior to being taken to England, and one that had been modified for military use with a cut down barrel, iron ramrod, and a bayonet.
I believe that the .55 caliber as average Pre-Rev/Revolutionary period caliber figure is derived from just averaging the rifles in Shumway's Rifles of Colonial America, and doesn't take into account age or condition. Plus, at least three rifles, RCA 42, 43, and 140 turn out to be .02-03" smaller than listed, and we ought to consider the possibility that Shumway similarly overestimated others he examined.
I do think that some of the surviving big-bored rifles were made that way - they'd be un-managably heavy if much smaller in the bore - but others I do think have been enlarged over the years, as their barrels are pretty thin or they look pretty unpleasant to shoot in the caliber they are now. I think it quite possible that the majority of rifles were originally in the .45-50 range, with a smaller number of big-bores made specifically with war, bear or buffalo in mind, and that usage and refreshing in period, the early 20th century habit of reboring originals to restore them to shooting condition, and the tendency of overestimate bores/overlook funneling at the muzzle have all contrived to push the perceived average up a bit while blurring any noticeable distinction between big and average bores. That is my theory at the moment, anyway...
