I wonder how long of a race series it would take to for "faster" marbles with better density/inertial imperfections to win, and for the result not to be just brownian motion. Ending up in the inner/outer lane of the first curve off the ramp seems significant, but random.
I've been wondering that too. There's bound to be differences in weight/smoothness just because of manufacturing, but it's not generally enough to make one team totally dominant.
This is definitely a good enough excuse to keep spreadsheets of stats for this wonderful sport. Probably not a good enough excuse to try to measure variation of imperfections in a bag of marbles. There are some principals about inertia of rolling disks and spheres that I learned in a dynamics class and can't remember right now.
I support the use of spreadsheets on most if not all occasions. I'll be curious to see what you come up with :D