Originally posted by swayze:
Originally posted by 49ers81:
Sorry it's not. It's just a stupid 'stat' made up to give people like Nina Kimes (is that her name?) something to talk about. Until it reaches the point where they become actual turnovers they have absolutely no more impact on a game than an incomplete pass would.
You can try and argue it anyway you want in the realm of if, and, or maybe but at the end of the day it's like saying, well today might be Wednesday, but it's not. It's Tuesday and no amount of phony projections about how it might really be Wednesday changes the basic fact of the thing.
The point of TWP% is to normalize a dataset that is highly variable based on numerous outside factors--whether or not a specific pass is actually intercepted is dependent on the defender's hands and concentration, the receiver breaking up the play, weather, "luck." We can't accurately judge the QB solely on the outcome because the outcome was out of his control; however, by normalizing TWP we get a clearer picture of the QB's decision making based on how consistently he is putting the ball in harm's way.
And yes, it's subjective--all statistics are even simply because we choose (consciously or not) which statistics to focus on. And yes, every completion, incompletion, sack, turnover, whether attributed to the QB in the box score or not, is impacted by many factors. The key question we have to ask ourselves is what are we measuring and why? The primary goal of football is not to generate statistics but to win games. In service to that we isolate statistics and try to correlate those stats with wins, so that we might better understand how to win. Which stats we choose to correlate should be flexible depending on what correlations we can identify.
It's not a question of whether or not the stat is subjective. Some guy at PFF with a degree in analytics sits around looking at film and decides whether or not it is a TWP or not. Whatever, it's no skin off my nose. My point is that until they turn into actual turnovers they are nothing more than an incomplete pass and have no greater or lesser impact on the game's outcome than that.
What you are arguing essentially is that somehow the 'analytic" aspect of it gives it some sort of greater authority than a coach looking at the film and saying, "well, we got away with one there."
You don't think that if a coach sees that happen often enough he isn't going to be aware that it might be an issue? Do you think that PFF "stats" telling him that his QB's TWP % is 2.8, or however it is they measure those things, is going to tell him something he doesn't know? Does the 2.8% somehow make it more meaningful? As I said, you are all welcome to come and revisit this issue when we are talking about actual turnovers.
Personally my whole take on football analytics is that it was developed by a bunch of math geeks trying to do for football what Bill James did for baseball. Not athletic enough to actually play the game themselves so they come up with this scam to make themselves feel like they can be "part of the game" and have some influence. Well that might work in baseball, within reason, I just think that football is a different sport. The game is so much more physical and a player's will to "win" in a particular circumstance is too difficult to quantify.
The perfect example is Jordan Willis. If you looked at his PFF stats you might come up with one picture but that picture might not account in anyway for what happened in the Green Bay game last year where his play on a field goal attempt let another guy in to block the attempt and where his blocked punt essentially won the game for them. You could argue that there wasn't a more impactful player on the team that night but I doubt that would show up in his PFF grade averaged out over a year.
There was a kick in the Seattle game that I thought was another great example. Al Michaels, doing what I am sure was some sort of promo plug, was saying just before one of Robbie's kicks, "Well, according to NextGen stats he should hit this kick 67% of the time". Shanked it to the right. Oops. I think it's mostly just meaningless nonsense that the league has been suckered into buying into that allows otherwise completely unqualified people to get on TV and "analyze" the game. That's my two cents on the topic and is all I have to say on this particular matter.