In other words, like I said, that formula overvalues chuckers from bad teams and undervalues great players who accept lesser roles on great teams.

No one here is dismissing the whole idea that usage / role matters. But your specific formula is probably not the way to go.

An inefficient chucker can also hurt the team by taking away shots. If you build an offense around Mike Beasley and let him have a gazillion of shots, then role players can only improve your offense so much.

So yep, there has to be a distinction between a useful and a harmful chucker. And talking about the formula, it should value "useful chucking" higher than "harmful chucking".

But at this point, we are kind of re-inventing the wheel here.

Offensive win shares (basketball reference) are basically calculated on the same principles we are talking about. They account for offensive rating, usage, they apply a parameter for "harmful chucking", and then they reward "useful chucking". Their formula is more sophisticated than that, it breaks it all into small pieces, but it's based on the same ideas:

http://www.basketball-reference.com/about/ws.html
Btw, in their formula, "useful chucking" starts at OffRtg = 0.92 * league average = a bit over 97.

It would be cool if they had OWS / 48 rating.

It's elementary math. You divide all results by 20. So the 20 in your formula simply makes the number smaller. It doesn't change which player looks better.

I simply suggested you a way to double check (multiply your end results by 20). But it's obvious.