There are 2 areas where I think the AH flight model favors light planes with oversized engines. Since yak3 is the most extreme case in this group, the effects are most evident.
The 1st is induced drag. Yak3 can maintain a climbrate that is close to optimal while flying at its lowest controllable speed (100 mph). Yak3 don’t actually stall at 100 mph (no flaps) - they fall off to the side due to torque after full rudder cannot overcome it. The stall buzzer will burn your headphones, you’ll have only sky in your forward view, and the plane still rockets skywards at about 4000 fpm. Spit 16 is quite similar btw, if you try to hold it at max power and lowest speed.
Now, some of this effect makes sense. At fixed G=1 (sustained climb), induced drag is a function of plane’s mass. Thus, light planes have flatter total drag curves below the min. drag speed. However, being able to sustain 4000 fpm climb on the edge of stall and controllability sounds a bit extreme. Emphasis on “sounds” because I did not go through the whole calculation nor do I have the data to do so.
The 2nd issue about slow speeds in AH is controll authority. At near stall speeds, control inputs in most planes are quite effective and have little adverse effects. For example, flying with the stall buzzer on and applying sudden full ailerons deflection (without adding pitch) does not cause one wing stall. Ailerons too are effective all the way down to the stall. Controls near stall are supposed to be a major issue for high power light planes that require a lot of controls inputs to keep flying at near stall. Yak3 does indeed cannot fly slower than 100 mph due to losing rudder authority, but the ailerons are fully effective.
If indeed these effects are a bit relaxed in AH, the light high powerloaded planes like yak3 will enjoy it more.
There are a lot of IFs above, I know.