Indeed that's the principle I'm exploring now. Though I'm used to simulations where I'm not given specific units, only a slider and some vague arbitrary quantitative measurement, whereupon I simply take a guess and whack the car around a track to see how it feels. This is a rather different game, though, as the driving is all done through systematic algorithms.
Since we're on topic then, care to throw a bone?

I'm not so proud (or time rich, really) that I won't ask for a hint. I'd be interested to see what you can find for the Automation Test track, since I can tell you now I've a car with a time not yet submitted that is
significantly quicker than all previously posted (but I think it can improve, pending tests with a different frame), so I can tell you that the optimum downforce for that track would be quite a lot lower in general (I find if your car hits 300-320 on the long straight and coming out of the slingshot you may be in the ballpark), whereas the top speed for the Airfield is closer to 240-260 (if even that?)
For example, the engine for my Reborn Genesis Time Attack develops 794kw (I always tune engines for high end power). It weighs about 845kg (the frame is small and won't fit a giant 10.9L engine in it). It seems that the optimum downforce for such a car lies somewhere in between 500-700kg @ 200km/h, but any more and the time starts slowing down because I lose too much top end speed. Yet even with all that I'm still more than a second off that crazy 1:08 time! Supposing I were to try using a 1210kg car that develops 1175kw though... however I seem to have messed up the balance between front and rear downforce, which I'm assuming makes quite the difference. And one thing I'm not sure of, is if placement of the downforce fixtures matter, or whether lips automatically contribute to front downforce, and wings the rear...
Ultimately I'm less interested in taking the crown as I am seeing the ultimate time attack machine being built. Once that's done, I feel like at least I can stop obsessing over it a bit and start designing more aesthetic works

---
p.s. aha, back in the game! I'll finish this round of tuning sometime in the next 24 hours and then maybe we can consider sharing notes then.
p.p.s. disregard what I said about the comparative downforce between the tracks. Perhaps that's only because I wasn't working with enough sufficient downforce to begin with. I honestly don't know. You'll see what I mean when you see the times I've just submitted: they were done with a model of exactly the same tune (including a rather short top speed!).