circumstantial data from one game to filter out and ignoring any other related circumstantial considerations in any of the others. For what you're doing to be relevant and accurate, you have to apply a consistent and balanced filter to the whole data set.
If you have the time or inclination to do so, I'm sure we'd all be interested in seeing it. But, only applying one element of it to one game is not statistically sound at all. It's just basic cherry-picking.