Much of the "science" is based on an understanding of marketing types.

If marketing materials spec a speaker at say 55-24khz -3db I'm going to assume that's under ideal conditions, in an ideal room, at ideal volumes, and with the audio engineer's tongue hanging from the left side of his mouth instead of the right, and want to try to back off at least 10hz from that lower limit. I'm also going to assume that they are using -3db instead of +/-3db for a reason, like perhaps a nasty bump someplace - probably near the upper or lower limit of the frequency response.
80hz is a nice round figure. Low enough that the transition from the speaker to the sub and back is pretty transparent (if the sub is setup properly) and high enough to take some load off the receiver and avoid the lower limits of the speaker. Taking some load off the receiver can be enough reason by itself. My new center is rated as close to dead flat to 40hz as anything you'll run into but I'm still crossing it over at 60hz just to take some load off the receiver and may eventually bump that to 80hz.