Luke
"The points of interest for line rake are the cg of the airplane and the line guide."
Agreed, and since the line will have a rearward "Bow" or "rake" during flight due to line drag, where the line exits the wing relative to CG will determine if the model tracks the circle properly.
If our model speed is fairly constant, and it usually is during a powered speed flight, the distance measured from CG to where the line exits the wing will differ depending on the wing span of the model. On a short wing, the location where the line exits the wing relative to CG will be less rearward than with a longer wing, assuming the line rake is the same.
That is the only reason I can see for the inboard wing measurement.
Relative to your comment on the ever changing CG due to fuel burn during flight. I'm sure most of us, I do anyway, build the line guide on a speed model big enough to be able to insert a looped line end through the guide to attach to the control mechanism. Due to making the line guide a certain width for the given loop size of the wire, said wire will ride against either the front, or rear edge of the line guide during flight depending on how accurately the line guide was positioned relative to the CG. If the guide is positioned near, or at optimum for the model, then the width of the line guide should accommodate small CG variances without causing excessive yaw issues.
Obviously, if you use a line guide just big enough to insert the diameter of wire you're using, and then build the line end afterwards so that the line always stays attached to the controls, then any CG variances would be more critical to yaw.
"And relative to your comment about the program working for you. How you know?"
It may not be the last word in determining the line guide location, but it works better for me than what I was doing...
Les