Incidence angles: I have the theory that the elevator trim makes a massive difference to the canard airfoil and angle.
First of all, as I think Don would say, get the CoG right, so that you load the canard slightly more. Then use a main wing airfoil that's close to symetrical so that zero lift is zero AoA, with no down-pitching moment. Then, with a flat bottomed airfoil for the canard, give it an incidence of no more than 1 deg relative to the wing. In straight and level flight, as the plane slows down, the canard will stall first and, as it speeds up, it will start to lift first (which is also important). I believe you will end up trimming the elevator with the panel deflected down a little (depending on speed). This will give the canard airfoil a camber, and a cambered airfoil has lift when the AoA is less than zero, doesn't it? The canard in this configuration will have an incidence of more than 1 deg relative to the wing. Before take-off on the knee-trembling maiden flight, trim the elevator this way.
I am not at all sure that my physics is right in the above paragraph, but can say for certain that my planes have flown well, excluding pilot error
Here is the plan I used for my simplest Starship..
PS This crossed with Don's post, above. But I don't think anything he wrote makes me want to change any of it!