Re: optimum heeling angle
Changes in speed due to changes in wetted area are microscopic. These are too small to be detected when sailing in open water even with precision instruments. They cannot be detected by anyone sailing.
Change in waterline length due to heeling sounds good but again this is too small to have any noticeable effect.
Most of the time sailing flatter makes you go faster because the sails are at a more optimal angle and thus producing the maximum amount of forward thrust. Sheeting them in too tight is generally what causes excessive heel.
I should point out that you will never get a simple answer to heeling even if you exclude non-related items. Sometimes (especially when racing) the goal is best track time rather than maximum speed. In other words, I'm here and the marker is there. I want to get around that marker in the least amount of time. That doesn't necessary mean that I can do it by sailing the fastest or by pointing the highest. For a given course with a given wind speed and direction there is going to be an optimum sailing profile. If I'm lucky then that profile is fairly flat and moving away from optimum won't change things much. If I'm unlucky then the profile is sharply peaked and moving away from optimum will make a big difference.
Rules of thumb don't work for boats. A boat will heel to a certain degree based on its stability profile and sail configuration. If you change the sail shape it changes the heeling profile. Again, if conditions are fixed with steady wind and steady course then one particular heeling angle will be optimum but this will not be the same for other boats or boats with different sail plans and will not be the same for different courses, different wind speeds, and different wind directions.
The change in hull performance due to heeling comes almost entirely from the change in canoe shape (the canoe is the part of the boat below the waterline). The rear of my boat is elliptical so heeling has very little effect. However, the bow comes to a vee and when you heel the front becomes asymmetric. On some boats this increases weather helm, on others it causes the boat to crab into the wind. It is changes like this that can effect sailing track and overall efficiency. For example, if the boat turns more into the wind then you have to steer out which makes the rudder lift towards the wind. This can be more efficient than having the boat turn away from the wind and having to correct with rudder because then the rudder is lifting away from the wind which puts more load on the keel.
There is no rule of thumb here though because different hull shapes and different stability profiles and rudder configurations will act differently. Usually when you have a tall, skinny keel it works really well and with great efficiency when you are moving at a normal speed through the water. However, at low speeds these keels may not develop enough lift so sometimes heeling the boat will add some countering force because one side of the vee is pushing more water away than the other. Then too sometimes heeling will change the track angle of the hull and this again can effect the sailing efficiency. This happens because often hulls are not rotationally symmetrical and as you tilt the hull the line of buoyancy will shift left or right. In other words, heeling can sometimes cause the boat to yaw slightly.
One reference was to light puffy conditions. If your sails are heavy and the wind is light and puffy then sometimes what happens is that a puff comes along and fills the sails but then dies off before the boat starts moving and the sails collapse again. Sometimes under these conditions it can help to sit on the downwind side and let the sails sag due to gravity. That way when a puff comes along the sail is already bagged out so the wind has time to begin moving the boat instead of just puffing out the sail. This should be common sense because light wind sails are normally made of very light material so that they fill very easily.
You are basically in the same position as the Wright brothers were in after their first glider model didn't work and they realized that Lilienthal's tables were wrong. They had no choice but to build a wind tunnel and spend hundreds of hours testing airfoil sections.
In the local lake where I've been sailing the wind is too variable because it is blocked and altered by the surrounding trees. Trying to make anything optimal is a waste of time. However, if you have a good open area with steady wind then you can test things yourself just by sailing a fixed course with a stopwatch. Eventually you'll build up a profile of how your boat handles in different conditions. If you are lucky then the profile will be fairly flat and have a lot of tolerance. If the profile is not flat then you pretty much have no choice but to just try things over and over to build up a more detailed profile.