"It seems most people claim that a tight fitting patch equates to better accuracy. Some hammer theirs down it's so tight. That's too tight for me but then I don't shoot in matches."
{Slightly off topic] I just finished testing this very thing in my Lyman .50cal. cap lock, 1:60 twist on Friday. A few days prior to testing, I used a nominally .490" RB (.488" in fact) + a .018" (compressed) pillow ticking patch and couldn't hit my targets well. (Btw, that combination seated almost too easily.)
Last Friday, I decided to bench rest the rifle @ 25 yd. using a variety of RB diameters from .488" -> .497" (cast from Lyman, RCBS, and Lee molds) as well as .018" -> .020" patch thicknesses. Suffice it to say that the .488" RB + .020" patch shot better than the same dia. RB with the thinner patch. Additionally, the .492" (as cast) RCBS RB + a thicker, .020" patch shot best of all, followed by the Lee .497" (as cast) RB and a thinner, .014" patch. The larger ball + thinner patch, however, was very difficult to start (needed a mallet on my short starter). That larger ball with a .018" was even harder to start and seat, but a tad less accurate than that ball with a .014" patch. OTOH, neither of these combinations was quite as accurate as the .492" ball + .020" patch, which required moderate effort, i.e., no mallet to start and/or seat. (It would work very well for a follow up shot when hunting, for ex.)
In short, I couldn't demonstrate that a thicker patch always was more accurate any more than I could show that a larger ball was always the better choice. One needs to experiment with a variety of RB diameters and patch thicknesses to hit upon a repeatably accurate combination, factoring in how difficult it is to start said pairing of course.