Turning it into a numbers game is a big mistake, IMO.
Yeah, I could write an algorithm and do some mathematical modeling to plug numbers into to get a result, but I would have zero confidence in that model because you lose all the nuance that is team building, getting players to buy in, personality, chemistry, having enough types of weapons to use off the bench for situational basketball.
And, of course, there would need to be a ton of number smoothing based on era, rule changes, culture changes and game strategy changes. 40% FG% in 1965=/= 40% FG% in 2019 is an example.
Now, this is a completely subjective exercise, so there is no way of coming to a wrong decision(unless you are trying to game the system to your advantage by ranking teams you feel are your competition lower to get their overall score lower, thereby helping you). I just think the numbers should be used to enhance your opinion, not to rule your decision.