theogt;5018735 said:
All 22 film is publicly available. If you have examples of errors in their coverage statistics, please share. But I don't see there being much room for meaningful amounts of errors in the statistics I'm referencing, even without the all 22.
The timing of the release of the All-22 film means that we still cannot use it as our primary source for our grading. We have deadlines to meet for providing data to our clients before the All-22 film is expected to be released on Gamepass/Game Rewind. However we will be using it as part of our review process each week.
https://www.profootballfocus.com/blog/2012/09/06/week-1-cowboys-giants/
Even had they used their All-22 coverage stats are flawed to say the least. It's like using a compass to measure distance.
Safeties play the vast majority in zone. Their assignation of a player for a target is based on who is closest to the target when the pass gets caught or falls incomplete. As the last line of defense, any time a coverage was blown or a guy just gets completely smoked the blame goes to the safety.
For example if a CB whiffs on a jam or you have a guy like Dan Conner or Bradie James covering the underneath, if the S ends up closer when the ball is caught then it's on his butt.
Baseball defensive stats, as hellishly flawed as they are, do better. Correlation does not imply causation and this doesn't really even correlate to anything consistently meaningful. It also has zero controls as well. Things like coverage, QB accuracy/decisionmaking, route combinations, etc, are in no way considered.
pff coverage stats suck. I remember it fooled you into thinking Hamlin was an elite coverage safety.