No, I don't think you've understood the crux of the criticism unless you factor in all the points. However, the fact that you consider them insignificant is a non-argument. It it is a rhetorical dismissal, but not an argument, that they could be used to discard any statistical evidence. This is simply not the case.
The fact is, the two points you note do invalidate the study. If you can't see that, I think you are being emotionally defensive of your own post. To stretch the logic - if Brady were playing against Division II, points/game would be meaningless - we understand that clearly. But, then you have conferences that are stronger in some years than others. However, even if 1 is granted as a non-factor, #2 is a factor because of the significantly small # of games you are factoring in. If you had 10 games against 10 common opponents, that would be a better factor. Or, if you had the last two years, even better.
Ultimately, no legitimate statistician in the world would take anything significant from your analysis. I'd have been laughed out of class if I had presented something like that in my old Econometrics classes.
These are equivalent to the arguments that people make about some natural ingredient that will prevent or help cure cancer. There is anecdotal evidence but it does not rise to validity b/c of the lack of consistent statistical data over a large number of people.
And, you also failed to factor in the distinctions between how Auburn played against LSU versus other quarterbacks. Florida and Auburn both had great defenses, playing at home, who rose to the occasion in highly emotional games.
It's a nice try, but your info makes no argument from a statistical viewpoint. 3 games against uncommon opponents with defenses difficult to compare across conferences do not make a valid argument.
It's purely anecdotal. It's interesting. It may or may not say something. But, ultimately not something one can give much credence to.
Film study of each player in these and other games and the game situation play is what will really tell the true story. Statistical evidence can give us robust data - much closer to a good statistic.
