If the wings are a swept style and they do not have any washout then I sure hope they are using a self stabilizing airfoil. But by using such an airfoil you're shooting your self in the foot since on a swept planform you can push the more forward part of the wing to higher lift coefficients and the higher lift is actually part of the stabilzing force setup. And yes, when I talk washout I'm meaning twist. Although for a flying wing "twist" is tecnhically more accurate since the term "washout" is something done on regular airplane wings to reduce tip stalling rather than make the wing more stable. It's the same thing but the name indicates why the twist is added. Clear as mud?
And I still say that when it's loaded that you're going to be lucky to get it to 10 mph let alone 20. And consider too that there's still that issue about the "effective" wing area and thus the effective wing loading and the resulting flying speed.
Without seeing the planform and airfoil choice it's hard to say much more than that. But either way you can't input 1000 sq inches operatintg at high lift coefficients and get a true picture of the real world situation here. Flying wings need to give up something in the lift department in some manner in order to be stable in pitch. So realistically your minimum flying speed will be higher than you have calculated. Again if you could show us the numbers you used we can indicate if you're being unrealistic in some area.
A power launcher is full of risks and how it works would depend a lot on the use of guide rails or other factors. I still say that if you really don't want landing gear on the model then the best way to take off is to use a big wire frame takeoff dolly. That way there's no surprises for the pilot like you're bound to get from a handlaunch of this large a model.