Quote:
Originally Posted by Dam8610
I say it's subjective because their main statistic that they're famous for relies on them "grading" each play for each player, with no objective criteria given for how a player earns each level of grade on a given play. That means they're going to inherently be prone to giving players they like more positive grades than players they don't like. Also each individual grader might grade the same play for the same player differently. For the particular accuracy metric you shared, the way I could easily see biases creeping into that is that they've defined target areas and what makes a bad throw, an okay throw, a good throw, and a great throw. But what if the ball placement is in between target areas? How does that get graded? Dependence on the grader and their personal biases is once again high. It's easy to imagine one of their graders dinging a QB like Peyton Manning early in his career hard for doing something you described positively with good reason. Low throws and underthrows are bad in their grading system, but a QB doing it to cut down on INTs is doing a good thing, we seem to agree on that. If PFF's grader just despised underthrown footballs or that QB for some reason, it's very easy for him or her to ding his accuracy heavily in that grading system. I don't have proof of their biases in action other than they have no objective criteria by which their numbers are easily reproducible.
I think you and I mostly agree on Lance. The biggest area of disagreement seems to be accuracy. Of course level of competition is a concern, but productivity mitigates that somewhat. My biggest concern is that he tries to play like Cam Newton but looks more like RG3. He needs to use his mobility to escape pressure and keep plays alive, but scrambling should be the last resort. That's my biggest reason for being hesitant to put him with the top QBs in the class, and as I said earlier, I don't know that Ballard and Reich will want to stake their careers on this kid.
|
I see what you are saying. As technology increases, high-resolution cameras with high frame rates, the ability to pause and examine, I see this as the natural evolution of analyzation. We do this every game our selves when we see a well-thrown ball vs a bad throw. You asked about balls thrown in-between zones which is a good point, and it probably depends on which side the ball is more on. But the zones aren't hit/miss it's accurate/less accurate/inaccurate etc. If it was either-or, that is where a bias could hurt you more, but the levels of increment minimize the impact of a bias.
Remember they are:
- absolving quarterbacks from getting downgraded for throwaways, spikes, batted passes and plays in which they're hit while throwing.
- Did the QB put the slant route on the front number for optimum yards-after-the-catch opportunity? Did he hit him with an accurate pass on his frame? Did he leave it in a catchable spot, but in a less-than-desirable YAC location?
- are also careful to add proper context to passes that appear to be off-target but are thrown away from the leverage of the defense. Such passes get an “away from coverage” designation that falls into the proper bucket of accurate passes.
Manning would have fallen into the last two categories there, which he was doing on purpose. I would assume they would know that if the announcer knew it and talked about it in-game. So Manning would fall into that area of accurate passes as well.
Finally, the whole argument depends on the presence of bias in the system, but there is no evidence of any bias. Just because there is a possibility for something to be corrupted doesn't mean it is. For example, the president and the GOP has lawsuits in every swing state trying to remove mail-in ballots arguing that there is corruption. They have not won a single case though because there is no actual evidence that there is mass mail-in voter corruption (that isn't caught) to discard the system, just their theories. (and before this turns into a political debate for some of you, yes I know the examples the pres said in the debate the other night, all those have been explained and you can look it up.) Bias would ruin the results if the same person did all the evaluations for one QB and no one else and no one checked the results. OR, the entire organization is biased against that QB and everyone working on that QB skewed the results. I find either unlikely because PFF is staking their reputation on the accuracy of their analyzations. If someone found bias it would taint everything the company does. Therefore they have a vested interest in being impartial. I would change my mind if something did actually come out about it. But I can't assume it is happening because it could happen.
Yup we largely agree in Lance, except for the accuracy issue.