If this is the case (either by accident or design), creating the two groups assists in my desire to remember as little as needed and provides an aid to quickly identifying the nature of the address.
The range naming convention doesn't make it out to the client end, of course, so out there, you have to remember where the address came from.
During my Linksys "blue box" years I developed that habit of leaving the DHCP range in the "middle" of the address space, following the lead of the default range. Then I would use the unallocated addresses below the DHCP for one purpose and the addresses above for another, effectively slicing my address pie into three pieces. Since the router was x.x.x.1, I just adopted the procedure of keeping all network hardware below the DHCP range (e.g., access points). Above the DHCP range I'd stick stuff like NAS devices. The only thing I had to remember was the DHCP range, and there I just extended the Linksys default DHCP range in one direction when necessary by the size of the original default range -- 50 addresses. It all made sense to me.
This all happened to work well when we first deployed Linksys VPN routers and it became impractical to rely on NetBIOS names. It was (or at least seemed) easier to remember where things were since it was almost certain that the only interesting piece of the pie to a road warrior was the piece above the DHCP range. A critical user device such as a fileserver would get the first address above the DHCP range.
I tell all this just to illustrate how even a single DHCP block can still leave the subnet's address space "organized" into three "easy-to-remember" blocks of usable addresses.
eBox changes things a little in that you can set up a static address at the server/router end rather than at the client end. I
love that feature!