I see where you're coming from, but in the end that is exactly why trying to spreadsheet HPS doesn't work. At best we can prove something isn't like it is by stating an unlikely but preferable situation for job A, and showing that job B doesn't even perform that bad under those conditions. There are so many variables to work around all we can do is dumb it down to the point it can be discussed. And in the end, jobs performing differently in different scenarios should be a given. All we really need to do is tell SE when one job is always outclassing another, be it due to them being blatantly overpowered or none of the preferable scenarios ever occuring for job B, while they are for job A (which is exactly what is happening with the whole SCH vs AST debate now).