OK, here's a short test of induced drag and the power curve.
I measured the time-to-climb between 1000 and 4000 feet. For each plane I used the auto-climb set to the various IAS, starting from sea level, but taking time only starting from 1000 feet to let the speed settle. In all planes I used max power (WEP) except the Yak-3 that has none. The mean rate of climb is simply 3000 feet divided by time measured.
The auto-climb cloud not hold the yak at 110 mph, even though a player can hold it at speeds even slightly lower than 100 mph. So for the slowest speed in the Yak3 I just help it manually keeping the stall horn buzzing the whole time. The speed was slightly over 100, so I marked it at 105, but the climb time was probably longer than it could be because I was not fully stable and banked a little occasionally - the Yak climb angle is so steep that I see only skies in the front view...
Moss VI - 50% fuel, 170 rpg packege
190A8 - 25% fuel, 2*20mm package
Yak3 - 100% fuel
Here is the mean rate of climb:
And here is the rate of climb normalized to the rate at best climb speed:
The Yak-3 suffers very little energy loss due to induced drag on the back of the power curve. Even when flying with the stall buzzer yelling, it loses just a hair over 10% of its max climb rate (87%), while the climb rate of the Mossie and A8 plummets to 55% and 73% of their climb at best speed.
Whether or not this is true to reality, I still need to think about it - but I want to shoot at red icons, so I'll do the thinking later. However, these are the game facts and a confirmation to my feeling that the Yak-3 loses very little energy even when flying on the stall, while the mossie becomes a dump truck when flying on the back of the power curve.