Originally posted by Furlow:
Originally posted by 5_Golden_Rings:
Originally posted by 5_Golden_Rings:
Originally posted by Furlow:
Originally posted by 5_Golden_Rings:
Originally posted by Furlow:
Originally posted by 5_Golden_Rings:
Originally posted by Furlow:
Originally posted by 5_Golden_Rings:
Originally posted by ritz126:
Originally posted by GoreGoreGore:
Passer rating has always been stupid. Every player always laughed about it. QBR is not perfect but it's a better metric. I'll trust the players and coaches when it comes to this.
which is more accurate
Daniel Jones is a top 10 QB or Daniel Jones is the 19th ranked QB
Derek Carr is a top 10 QB or the 16th ranked QB
Jacoby Brisset is a top 10 QB or the 18th ranked QB
Marcus Mariotta is a 11th ranked QB (LOL) or 17th?
Also the idea that Trever Lawrence andy dalton Goff and the other QBs above are ranked above Jimmy and even Cousins (who imo hasn't played amazingly but definitely well) is laughable it throws out the stat completely
they really need to rework that formula
Just need to wait for more data points.
No.
Yes. That's how this works. The main difference between passer rating and QBR—other than the fact that the former gives credit and blame to the QB that should go to the WRs—is that you can play a bad game and still have a high passer rating. For example, in the 2011-2012 NFCCG, Alex Smith had a 97.6 passer rating. And we were 1-13 on third down, and Alex was 9-23 for 84 yards on all but three passes. Passer rating is too limited to tell the truth about the quarterback.
More data points for a bad formula isn't going to help. You act like there aren't a s**t ton of examples of QB's with high QBR's who played terrible.
Have you heard the mathematical expression "returning to the mean?" Well sometimes you have statistical anomalies. Sometimes you have several in a row. But with enough time, the truth is revealed.
In any event, if we had a competition over who can find more bad games with high ratings, I'd win with passer rating over you with QBR.
No you wouldn't. Otherwise there wouldn't be so many more outliers with QBR than QB rating. That's not just this season, every season that QBR has been around it's like that. You just favor QB's who can run, as does ESPN because they're cool highlights, and fantasy football players because it racks up points. QBR is a reflection of the new generation of fan that cares about highlights and fantasy points.
Yes, I would.
1.) There aren't more outliers with QBR than QB rating. The top ten list is almost identical. More interestingly, if you take the top thirty* Total QBR and passer rating and see how the top scores and bottom scores correlate, they're almost exactly correlated (QBR is already normalized, so you have to normalize passer rating). So you must mean something else by "outlier" than a numerical outlier, because the two have identical correlation with one another when each rating is listed in numerical order.
If you mean players being more than a standard deviation above the mean, passer rating has six values that are more than a standard deviation above the mean, while Total QBR has only four values more than a standard deviation above the mean. As far as more than a standard deviation below the mean, passer rating has four below the mean, while Total QBR also has four more than one standard deviation below the mean. This indicates that passer rating inflates the difference between the good and not good more than Total QBR does (on this limited sample size), and that passer rating actually has more outliers.
*30 because I have 30 slots on my Excel sheet with the relevant formulas and don't feel like adding two or more rows.
If instead you just mean mediocre QBs at the top that you don't feel belong there, and good QBs near the bottom that you feel should be higher, well first of all, your OPINION on them is hardly gospel. And second, you'd obviously not be considering the fact that several young QBs took big steps this year—which is AS IS EXPECTED at some point in the careers of highly drafted QBs (such as Tua and Hurts). As for Geno Smith, his passer rating is hardly higher than last year, but his QBR is a lot higher. And absolutely no one disputes that he's having a career year. Which means QBR tells Geno Smith's story more accurately than passer rating. As for Derek Carr, he's been mediocre to slightly above average for several years. Guess what his 59.1 QBR is? Mediocre to slightly above average. Him being 10th in QBR doesn't mean he's the 10th best QB. It means some good QBs have had a couple of games where they struggled uncharacteristically, and as a result, the whole of all QB performance outside the very top performers has taken a hit.
Meanwhile, you have the likes of Andy Dalton as the 9th best QB based on passer rating, and as for QBR, the only strange one is Brissett (Cooper Rush went 4-1 in his starts. He "just won," and he "just won" for the same reason Jimmy "just wins." Rush had four games where he was managing the game and making the necessary plays, and one singular horrible game where he threw 3 interceptions, and for that one game, his four decent games are erased in passer rating.
Of the top 10 guys in QBR, only two of them have a passer rating ranking that is different by 5 or more rankings. Of the top 10 guys in passer rating, FOUR guys have a passer rating more than 5 rankings different than their QBR ranking. If one of these ratings systems has more outliers, it's passer rating. At least for the top ranked guys.
As for the bottom 10 guys, QBR has 3 guys with the worst QBR who rank more than 5 spots differently in passer rating. Passer rating also has only 3 guys with rankings in QBR more than 5 spots different.
.
.
And lastly, with respect to Jimmy, his QBR on the season is presently 50.4. That's very slightly above average for the year. He's played in 9 games. He's played one absolutely horrid game, two other bad games (Atlanta and Kansas City—they were bad games, sorry fan bois), three pretty good games (Seattle, Panthers, Rams and Chargers), two excellent games (Rams and Cardinals). On balance, the Falcons and Chiefs games cancel out two of the "pretty good" games, and the Broncos game was so bad it cancels out one excellent game as well as another pretty good game. That leaves you with a little better than "pretty good" for the year. So his QBR for the year is maybe 5 points lower than it should be. And that factor is easily accounted for by remembering how many huge plays the weapons he has makes, which as a whole is simply superior to everyone else in terms of action after the ball is in their hands.
His passer rating is high because passer rating is not capable of understanding how truly horrible his Broncos game was, and it's that game right there that is the albatross keeping his QBR barely at average (and rightfully so). However, with more games, the weight of that Broncos game will diminish, and a more accurate representation of what Jimmy has done this year will be represented.
Oh, and it's not running yards by a QB that makes QBR different from passer rating. I've already explained this to you twice. It's that QBR takes into account CONTEXT of the stat. It takes into account whether the pass was a throwaway or spike, whether it was dropped, whether it was short or long, and so on.
It heavily weights running (you add the passing and running metrics together for total QBR). Hence why QB's like Fields, Daniel Jones, Lamar Jackson, Mariota, Brissett, etc. are ahead of Jimmy G. If you want to at least appear unbiased, then only use the passing metric for QB's. That puts Jimmy ahead of all of those besides Brissett. Essentially, it removes the outliers and more closely resembles passer rating. That's if you still think that QB's should pass the ball, rather than run it.
QBR does not "heavily weigh rushing." What it does is weigh SCORING, that is, and expected points added, and EPA favors passing over rushing any day of the weak, because passing is how you score points. Hence why only ONE of the QBs you listed gets more points from their legs than their arms.
Regarding passing, Jimmy is only ranked 15th in expected points from passing. That is as much why he's not higher than his "lack of rushing."
But starting with Brissett, he literally has a higher expected points added from passing than Jimmy. He is literally helping his team score more points with his arm than Jimmy. There is no point in even bringing up his rushing contribution, as he is helping his team more with his arm than Jimmy is, at 33.9 compared to 30.5. He does contribute 9.9 expected points from rushes as well, while Jimmy loses about 1.9 points per game with his legs.
But disregarding him, as I said, ONLY ONE of the QBs you listed would help their team more by running more than passing, and that's Fields. All the others have comparable passing expected points to Jimmy, being only slightly behind him in expected points added from passing, whilst also adding a good ten to twenty expected points from the run.
Fields: HE is the one who is definitely up there because of rushing. His expected points added on pass is 4.6, while his expected points added on rushes is 38.8.
Jones' expected points on pass is 28.5, only 1.8 below Jimmy. His rush EPA is 18.2, meaning his passing is over 150% more important than his rushing. So yes, Jones' rushing helps his QBR, but it only helps his QBR because it helps the Giants win. Nevertheless, his passing is 1.5 times more important.
Lamar Jackson also has almost the same contribution in expected points from passing, at 27.2, whilst also bringing the rushing threat to the tune of over 17 expected points per game.
Mariota's expected points from passing is 29.5. He's within one point of Jimmy as a pure passer in terms of adding points, and is also nearly ten times more useful as a runner. Mariota is essentially Jimmy this year, except he can run as well.
.
.
In fact, if you rank the QBs by expected points added from rushes, literally only one QB helps his team more by running than he does by passing, and that's Fields. Jimmy is behind Kirk Cousins in expected points added with passes, but the two are virtually the same (Cousins at 30.5, Jimmy at 30.3).
Oh, and Mac Jones is almost identical in expected points added passing and rushing. An interesting oddity.
.
.
