Now this Pandoras Can of Worms seems to have an answer, what would it do to manouverability ???
The basic problem is deciding "what matters operationally" and "how can we plug that into what matters in the specific air combat code of WTIP"?
This is not an easy thing to do.
First, "what matters operationally" is not always the same thing. In fact,
you need to begin with the elementary knowledge that durability, firepower, speed, maneuverability and altitude ALL COMBINED matter LESS THAN one quite different factor ALONE matters: surprise. Both offensively and defensively, surprise is an order of magnitude more important than the sum or product of ALL other factors. So this is a marginal matter - it probably matters more in the game world than in reality - but this game is very sophisticated - and it DOES have lots of die rolls, some of which can be interpreted as meaning surprise was achieved.
Second, more often than not, speed matters more than maneuverability does. Thus, a plane with an altitude (actually energy) advantage can buy some speed, and use that to make an attack (or escape) run it could not normally make, demonstrating that success can really be a function of speed. In WITP code, according to material disclosed in this forum, while there are many factors in the air combat routine, the dominant one is called "maneuverability". Statistical analysis of the "maneuverability rating" of stock aircraft shows this is, for the simple, single engine case, either maximum speed divided by 10, or that same factor combined with a small amount (10%) of rate of climb data. [At first I was upset with this. But careful analysis caused me to learn that it works, and not just for WWII era aircraft, to a remarkable degree. Eventually I increased the fraction to 20%, but retained the original practice.]
Now one can take a wholly different approach - and all my life I have done - following some more complex simulations. Maneuverability is usually defined in terms of power loading, wing loading and turning rates.
Many also add in rate of climb. The very best maneuverability ratings actually change at different altitudes: my own system gives you a "rate of climb correction factor" (and similar adjustments), and recalculates maneuverability at each altitude (in increments of 10 meters or about 33 feet). [Rate of climb is, by definition, either 100 feet per minute, or 30 meters per minute - they are virtually the same value - at the service ceiling - and zero under stable forward flight conditions at the absolute ceiling. The ROC correction factor is the amount ROC decreases every level until you reach service ceiling, at which point the ROCCF is doubled, until you reach an altitude at which ROC = 0.] In a similar way, speed is modified by a speed correction factor - except that speed INCREASES until the aircraft reaches its optimum operating altitude, then it DECREASES between that altitude and service celing. Figuring out what the speed or rate of climb is for any given altitude is quite different than what we need here: we are allowed only ONE value in ONE field. And figuring out how to combine loading, ROC, speed, turn rates, etc - is very very tricky - since there is no one thing that always dominates - and since the amount of any given thing really varies with lots of factors.
The first step in simplification is to believe in the code. It considers altitude - so leave that alone. It gives higher planes an advantage or disadvantage for altitude. It almost certainly considers surprise - so leave it alone.
The second step in simplification is to understand what we can do in this model: give each plane one average value. Theory aside, we can't do more than that. That forces us to decide "what matters most, most of the time" and, maybe, "what else matters, and in what proportion to what matters most?" Here the genius of the design shines: speed matters most. Using speed as the dominant factor was the right design decision.
The third step in simplification is to understand what data we have, and can get. IF you must use something in your calculation that is not already in the data set - THEN you must be able to get that factor in equal validity for all planes - even those that never flew and cannot be measured. Rate of turn might be very nice to use - but (a) it is almost never recorded in standard references; (b) it is not measured for many aircraft in the data set. Gathering the data for rate of turn will turn generating this one factor into a task that would take a man year - more or less - to achieve - and might not be very valid - since so many estimates would be required.
OK - that is the process simplified - make your choices.
