Montague has given you very accurate information.

For any flying wing to acheive positive pitch stability (a must) two things must be present. First CG must be further forward than a tailed aircraft and the second there must be aerodynamic reflex. The reflex can be acheived in the airfoil itself or can be done simply by rigging the elevons with an up reflex in neautral.

The amount of reflex and forward CG is directly related to the sweep of the wing. A straight wing flying plank will require the most most forward CG and up reflex and a delta will require the least.

The 10 to 15 percent starting point is indeed a good starting point for a straight wing. 30% percent will without a doubt be uncontrollable. Now I think there is some confusion about MAC (mean aerodynamic chord) going on here. On a straight wing its a no brainer 25% MAC is indeed a quarter of the way back from the leading edge but if there is any taper (leading, trailing edge, or both) or sweep then the 25% percent point is not going to be a quarter of the way back at the fuselage. There is a simple way to figure MAC graphically and someone I am sure can help you with that in the aerodynamics forum.

The last wing I did was for some Public RC shows and was simply a conventional forty size trainer with the tail wacked off and reattached without the elevator and rigged with a release mechanism the CG was move forward to around 13% and an additional servo added to the ailerons to enable elevon with the ailerons reflexed up twenty degrees. For the shows the tail was dropped on T/o and the flight was completed without. Was quite a show stopper.