I agreed Dave's point earlier, but at a minimum the 51B needs to be in the mix.
The P-51D was perhaps a better overall weapon system with more firepower and ammo, built in safety mods for the gear door uplock, beefed up tail vertical spar, bubble canopy, 1000 pound external rack capabilitiy - but at the cost of 10 percent weight increase over the P-51B and same Hp (engine)in summer 1944..
The 51B/C upgraded engines and stayed with the P-51D's 1650-7 and 150 octane fuel to simplify logistics... but that complicates this model slightly as the USAAF did not perform flight tests to get dash speed comparisons with the lighter airframe of the B. The 51B/C also fought against all the 109s and 190s that fought against the D's. To be anal - one could solve for Thrust of P-51D at it's rated top speed Hp and altitude and Plug it into P-51B Thrust.
I do agree the boundary condition of equal ammo/fuel load for One set, and T.O. Combat Load without internal fuselage tank fuel and without drop tanks as the other extreme.
Back to the model. I was waiting for an explanation regarding 'how' this exercise was going to proceed but offer the following - and hope that Marcel stays awake
The reason we have to get Hp converted to Thrust, and assume Thrust for that Hp/Altitude/Boost is constant, is that the freebody force diagram in the horizontal axis has Thrust opposite Drag (Induced and Parasite(everything other than wing)). The Parasite Drag Coefficient (Cdwet) is given for multiple sources in Lednicer's oft quoted reports, as well as Fw 190D-9 and A-8 and Spit IX.
To Continue:
When velocity is changed, the Total Drag is changed, but the Thrust remains the same at peak Hp/Boost and RPM. At max speed - no acceleration, below that you can accelerate from a lower boost/rpm back to max thrust and accelerate until the increase in velcity increase drag to equally balance the Thrust. Every engine and propeller and gearing set up converts Hp to thrust slightly differently and every propeller is a compromise design - so there really should not be an assumption that Hp is analogous to Thrust for any of these a/c COMPARISONS
At THAT maximum Thrust (we haven't solved for it yet) for That altitude and density, the Velocity is found from the Flight Test Speed at That Altitude. For that altitude and standard temp/pressure, or whatever you want to use for temp and pressure, you now have density. As noted below you need to normalize TAS to surface tem/pressure conditions at that test location
You look up the aspect ratio's, the tip chord ratio's, the Cdwet for Parasite drag at zero lift (say Cdwet) for each of the a/c.
That enables us now to plug velocity, density, and airfoil efficiency factors of Aspect Ratio and Tip/Chrord ratio to get the Induced Drag Coefficient. We have the Parasite Drag - now we can solve the equation for maximum Thrust.
T=Drag induced + Drag parasite .....at Maximum Velocity
At that altitude, that hp, that rpm/boost, for level flight at maximum speed in a flight test. Normalize each of the flight test densities for the altitude by the surface temperature and pressure of the tests to normalize True Airspeed for each of the recorded flight tests conducted in different locatopns (i.e Augsburg, GY versus Palmdale, CA).
As Soren said once, this isn't rocket science, and it doesn't need to be as long as we realize this is a perfromance 'speculation' based on rigid wing, airfoil, no consideration for factoring pitching and rolling loads on the airframe, etc. We are not factoring in the induced drag for Rudder or Elevator trim
Now, on to turn/G load/Velocity plots
As the bank of the turn increases from zero/level flight, the Load factor increases with Cosine of the Angle of Bank.. The turn requires increasing the angle of attack and rudder deflection to carve the turn horizontally. The actual lift and drag vectors change with angle of attack of the free stream velocity similar to an initial climb but we won't 'do that' - as we begin to complicate the model by incorporating real life effects and start to factor in elevator and rudder loads and pitching/yawing moments which must be accounted for when solving for the Free Body Force Diagram for Manuevering Flight
As the angle of attack increases, the Thrust remains constant from our solution above, but the velocity will move to a lower value based on the angle of bank (but not linear with the change in bank). The relative angle of attack from free stream angle of the wing/body system increases as the stick is pulled back in the turn. The Lift Coefficient increases as the Angle of Attack increases to sustain the vertical Lift vector required to balance the Weight to keep same altitude in the new increased bank angle.
At each altitude and constant thrust, there is one maximum bank angle and velocity which will still sustain a component of Vertical Lift (equal and opposite to the Weight vector - which is greater than the Lift Vector in level flight) while still holding altitude constant... This Vertical Lift Vector is Different from the Normal Lift Vctor to the Wing
As the angle of attack increases from the local angle of attack and CL setting for maximum speed - the drag increases beyond Thrust and the a/c slows down, as the bank increases, the local angle of attack increases to sustain a higher CL to maintain altitude and the resultant Drag again is higher than the available Thrust and it slows down more.
Going at a steeper bank angle from this point while maintaining constant altitude continuously causes the local Angle of Attack to increase - taking it to, and then past, CLmax for the wing - and into a stall.
The net - is that no aircraft can sustain a bank angle at its rated top speed and still maintain it's altitude. If it enters a turn at less than max power and speed, throttles up and increases the bank, there will be a speed lower than its max level speed at which it can sustain THAT bank angle and load factor.
Solving for the above will get us to a point where we can talk. I am a little busy at the moment... but I can check the math.