S197 wrote: ↑
Mon Dec 07, 2020 12:21 am
J. Kapp 11 wrote: ↑
Sun Dec 06, 2020 9:41 pm
You really like that ESPN QBR.
There are other stats out there. As brought up in this post, an important one is a QB's play in the 4th quarter. Wonder what his QBR is in that frame, since from what I understand QBR accounts for garbage time. My guess is that ESPN won't provide that because it doesn't support their never-ending narrative that Cousins is a choker.
EDIT: And for those of you who love PFF, Kirk Cousins is the 5th-rated QB in the entire NFL, according to them. So who's right. ESPN's nebulous formula, or PFF's "evaluators"?
It’s not ESPN’s QBR, it’s just a QBR. Someone posted the formula from the wiki in another thread. This thread, which I didn’t start, is about his QBR so why wouldn’t I bring it up?
I think there's some confusion as to what we're talking about.
Here's what I understand:
QBR, short for Total Quarterback Rating, is a metric developed by ESPN in 2011.
They say, "Unlike other measures of quarterback performance, it incorporates all of a quarterback’s contributions to winning, including how he impacts the game on passes, rushes, turnovers and penalties. Also, since QBR is built from the play level, it accounts for a team’s level of success or failure on every play to provide the proper context and then allocates credit to the quarterback and his teammate to produce a clearer measure of quarterback efficiency."
In a rough sense, the Total QBR is supposed to measure QB efficiency, as opposed to statistical value. If a guy gets a 75, it supposedly means that he played well enough for his team to have a 75% chance of winning. Therefore, a perfect rating is obviously 100. An average rating is 50.
None of us has never seen a formula for Total QBR, and there's a good reason for that. It's proprietary. It belongs to ESPN, and they're not going to share it (and thus let somebody else copy it — or improve it). According to their website, they determine an Expected Points Added (EPA) by the quarterback for each play. That EPA can be positive or negative. They then add it all up to get the final rating for the game. How they derive EPA is also proprietary.
ESPN itself admits that the credit given to the quarterback on each play is subjective. So for example, on the 62-yard catch-and-run touchdown from Case Keenum to Adam Thielen against the Rams in 2017, ESPN likely wouldn't give Keenum as much credit for that play as they would if he'd thrown it farther downfield, or from a collapsing pocket under duress. They'd say it was mostly Thielen. The problem with that assertion is that after-the-game investigation revealed that Keenum audibled to that play, giving Thielen a chance to go to the house if he could break one tackle. In other words, it's subjective, and not always based in knowledge of what's actually going on.
Thus ESPN hedges: "Although QBR is not always a perfect reflection of a quarterback’s performance, it does solve most of the problems of traditional stats."
Passer Rating is the official statistic used by the NFL since 1973.
It is based on attempts, completions, yards, TDs and INTs. There is no subjectivity. Plug in the numbers and get the Passer Rating. A perfect score is a weird 158.3.
Obviously there are issues with Passer Rating. For example, it doesn't differentiate between stats earned during garbage time vs. clutch situations.
But there are plenty of issues with Total QBR. It's a system that is a) secret, and b) subjective. Why is it that we are skeptical about PFF's rating system, but somehow we're OK with ESPN's? They're both based on subjective decisions made by very few or one person for whom qualifications are unknown. At least with Passer Rating, I know how the score was determined. Meanwhile, I find it hard to believe that a subjective system isn't influenced by personal bias. Call up any game involving Aaron Rodgers, and you'll see what I mean. Recently they lost to Indianapolis, and ESPN gave Rodgers a significantly higher rating than Philip Rivers, who had nearly identical numbers.
EDIT: Perfect example. ESPN gave Cousins a 25.7 rating for yesterday's game. That's bias. ESPN says, "That means on completed passes, the EPA is divided among the quarterback, his receivers and the offensive line based on how far the ball travels in the air, what percentage of the yards were gained after the catch (compared to how many yards after catch are expected) and whether the quarterback was under pressure
." Under pressure! Any of us who watched that game know that Cousins was getting buried out there, play after play. And some of the throws he made under that pressure were absolute dimes.
Yes, his pick-six probably hurt his rating in a big way. But even that wasn't totally his fault. Dalvin Cook never looked for the ball. I could see his rating ending up 50 or 60. But 25.7?
Meanwhile, they gave Mike Glennon a 38.3. Think about this. Mike Glennon, who threw a red-zone pick, a pick in overtime, lost a fumble, and took a sack for a safety (ESPN says "taking sacks" hurts QBR ... how about taking a safety?) ended up with a better QBR than Kirk Cousins. It's ridiculous on its face.