The angle of attack is not determined by the incidence relative to an arbitrary line drawn on the plans. Incidence is easy to measure. Direct measurement of angle of attack is best done in a wind tunnel. Direct measurement of the angle of attack of the wing of a model in flight has rarely been done and requires special instrumentation.

A plane in level flight at a constant air speed takes on a pitch attitude which produces the necessary angle of attack. In this state of dynamic equilibrium, the sum of all the moments is zero and the sum of all the vector forces is zero. A simple case would be where the lift acts at the CG, the down load on the tail times its moment arm is equal and opposite to the pitching moment of the wing, the thrust line passes through the CG, the lift is equal to the weight and the thrust is equal to the drag. For the lift to be equal to the weight at a certain airspeed, the wing must be at a certain angle of attack and the plane assumes a pitch attitude by the flight trimming process that produces the necessary angle of attack relative to the direction of flight relative to the air mass it is flying in. If the plane changes airspeed and assumes a new state of dynamic equilibrium, it will assume a new pitch attitude to produce a new angle of attack such that the lift produced is equal to the weight. The new value of thrust will produce a new airspeed such that the new drag is the same magnitude as the new thrust. The new pitching moment of the wing matches the new down load on the tail.

Therefore, the "ideal angle of attack" is determined by the throttle position and the elevator trim position for a given CG. The plane then takes on a pitch attitude which produces the "ideal angle of attack." The trick is to arrive at the airspeed, CG location and elevator trim that produces the "ideal angle of attack" for that model. The "ideal CG location" will result in zero down load on the tail at the associated airspeed.

When flying in a level turn the trick is to fly at the angle of bank that puts the wing at the "ideal angle of attack" for that airspeed and lift force. The desired angle of bank can be determined by flying circular laps at full throttle and various angles of bank to see which angle of bank produces the fastest lap time.

A paper or computer analysis of a particular design configuration can give you an idea of how it will perform and the configuration can be changed to see if the performance improves. In this way the designer can evolve a design that is more refined. The main problem is that the coefficients of parasitic drag for exposed cylinder heads, control horns, etc. are not generally known because they haven't been measured.

The engineers at the University of Stutgart did some measurements and found that the exposed control linkages for two flaps and two ailerons on an F3B model were about 10% of the total drag in high speed flight!!! The message is that attention to detail in streamlining can have very benefical results.

BTW, I chose the S6063 in my example because it seemed to fit the application and the polars from windtunnel measurements were available at the reynolds number of interest. The particular wind tunnel test model was not an accurate representation of the airfoil because it had a slight reflex in the trailing edge of the model. The mean line camber, lift and drag coefficients were a bit lower than the true S6063. Since Martin Heperle's web site is shut down, the MH33 polars generated by X-foil were not available to me.