He posted this in the 3D forum, so that's why I biased my answer away from your more pattern oriented suggestion. I'm assuming he's near perfect neutral on the CG and elevator trim. If that's the case the plane should really drop equally inverted AND upright when chopping the throttle.

In my suggestion, observing up and down lines takes the gravity and CG variables out of the equation because if downline is truly vertical, he's got that right. All that remains is the downthrust variable and the engine angle is the largest factor while climbing once the others are set. Of course, he needs to view these from the side to observe pitch changes. Maybe I didn't make that clear.

I like your method for true pattern and park flyers, but even so the "trimmed for level flight" method assumes a certain speed if the CG, decalage and airfoils aren't optimum for aerobatics.

The 2 or 3 deg right thrust seems to be a good starting point. I like to check that when hovering to see if I keep needing to correct for right or left rudder. Again, a 3D answer vs your pattern oriented suggestion.

Many ways to skin a cat, I'm sure.