Being snowed in, I've had time to do some reading!
I read in a book about Apache warfare, that the personal crest an individual warrior/hunter put on his arrow always went under the fletching. This leads one to the obvious theoretical reason: You cannot "counterfeit" your own crest on a found arrow without having to remove the fletching. If the crest was ahead of the fletching, it wouldn't be anything to sand it off or put your own over it and say, "Oh, no, this is my arrow!" The book states knowing the tribe the arrow belonged to was by differences in fletching styles usually. Sometimes, there were common crests on an arrow, ahead of the fletching. There was an Apache clan that identified their arrows by staining the entire arrow shaft red. But the individual owner's crest always went under the fletching.
I have the book "Greek Fire, Poison Arrows, and Scorpion Bombs: Biological and Chemical Weapons in the Ancient World" by Adreinne Mayor. There's also some great information in this book about fire arrows. Specifically, that cane arrows could be packed with flammable material and holes drilled near the head where the tow was wrapped. So that when the arrow hit, it carried with it some flammable payload besides the stuff wrapped at the head. This book theorizes that some of the cresting on Scythian arrows identified poison arrows in the quiver. In other words, a person could see by the cresting the arrow was poisoned, and thus be more careful with it.