In this post of the same thread I addressed the concern about crit rate.

Otherwise, yes, it is deliberately simplistic. If we include things like mob positioning, encounter mechanics and GCD optimization then a theoretical simulation won't cut it. We would need practical feedback in the form of parse results.

Ultimately, if ID increases your DPS to any significant degree, do it. If it doesn't, don't. I think it's too risky for what little you get back.