I think Fabrico hit the pertinent points.
Consider that if the disk is just flung through outer space, there should be zero tendency to flip out of the plane (assuming you throw it straight). So you can assume that ALL the forces causing it to rotate out of the level are due to aerodynamic forces. If you know those forces, great, you can specify a maximum tilt angle, whip out the ol' dynamics textbook, and see what it takes to resist those forces. But in the more likely case where you don't know those forces, you also don't have any basis for figuring a stabilizing speed.
From past experience flinging Frisbees and other disks, you do need a fair bit of the weight near the rim to stabilize it adequately.