(10-27-2020, 10:39 PM)NeptuneBeachBum Wrote:

(10-27-2020, 10:16 PM)mikesez Wrote: Bortles also threw 28 yard touchdown passes that Allen Robinson would have to stop and jump for.
Does QBR tell any difference between a touchdown pass like that and the one Minshew threw on Sunday?

Nope.  But neither does Passer Rating, nor Adjusted Net Yards per Passing Attempt (ANY/A).  To the best of my knowledge, we were debating QBR vs. Passer Rating vs. ANY/A.  I guess you realize that argument isn't going very well right now and you want to change the subject?

You're missing the point.  I said every stat has its limitations.  But if you have to pick one to evaluate QB play, QBR gives you the best information because it provides context that Passer Rating and ANY/A do not

Because you emphasized, twice, how difficult and beautiful that pass from Minshew was.  Why emphasize something if it doesn't matter to the point you're making?


I'm down with subjective grading, watch the play and decide for yourself if you should grade Bortles high for a long pass or grade Robinson high for chasing down an inaccurate duck. It's just an opinion, but it can be interesting.
I'm down with anything I can compute for myself with a simple play by play.  If you can do it off the stat line, even better.
What I'm not here for is a number I cant verify, that they're going to pass off as an objective fact.  That's nonsense, to me.  I'm not here for the computer saying, this play had a bigger impact on the outcome game than this other play, and you attributing that to the QB rather than fate.  The QB didn't know the computer readout before the snap.  The coach didn't know that before the play call.  What they did know is what the defense was showing, and which of their guys were looking tired. And the QB knows what the play call is, even though the QBR algorithm does not. If we are going to wait the value of a play based on something that happened before the play was made, it should be things that the person making the play would be aware of. Time remaining, sure. Score, sure. Down and distance, sure. But make it predictable.  Don't weight it on some statistical analysis based on the outcome of other, similar games.  But why bother trying to find and reward those "key plays"?  As I said, if the QB is playing the entire game, all that should come out in the wash anyways.

---