Interesting. In my case, Rusto blue, and black metallic on two different 4" rockets, and Duplicolor green metallic on a 54mm rocket (probably 2 or so coats on all of them); painting was done a various times in the last 5 years or so as well. BRB RDF in the nose cone (I only put trackers in the nose away from other electronics for many years - no point in risking it), with antenna pointing up into the NC. This is only an RDF tracker, and is low power, 16mW @ 433MHz, using a home-made YAGI. Flights to ~10k' at most, and recovery at a max distance of a bit under a mile. Its hard to judge however - at 1 mile distance the simple highs and lows of the field can easily block signal; and I don't bother checking during flight, so have no idea whether I get signal at 10k' AGL. What I can say is that while I may not always get signal at the pad after the rocket is down if the distance is > 0.5 miles, when I get closer it has not been a problem; and as one expects, signal gets to be too strong as when you get near.
As you say, it will depend on the size of metal particles in the paint(assuming they are not in electrical contact with each other) and transmitter frequency. I also wonder if decoding packet radio requires a stronger signal than an RDF. You've got me interested enough I might try a test this weekend comparing signal strength with the tracker inside and outside of the NC.....