ESPN laughably still pushing 'Total QBR'

[Warner2BruceTD]

Typical Monday morning, ESPN on as background noise, checking some emails and getting some work done. Glace up, and notice on the scroll "Total QBR leaders: Week 1". I had forgotten this existed. Nobody talks about it or references it.

It appears that ESPN's first attempt to make up a statistic has failed. Does anybody care about 'Total QBR'?

 

[FedEx227]

No, nobody does, it's a weird stat that worse of all, they don't tell you how to calculate. Any stat that doesn't have a formula that is visible or at least tangible should be thrown out. You basically say, here's this stat, no it's perfect now go use it. Stupid.

Even then football analytics still lag very far behind and rightfully so. Like basketball there's way more team involved than baseball so attributing individual metrics is really hard to do and in most cases pretty silly.

 

[Warner2BruceTD]

That is where footballoutsiders, pro football focus, etc fall on their face. You can not boil down individual performance in football with a neat, concise number. If you try hard enough, you can poke holes in most baseball metrics, and like you say, that's (for the most part) a sport dependent on individual performance. So doing it for football makes little sense.

 

[FedEx227]

Basketball metrics only really work IMO with 5-man units and finding what the best combo of guys is, but individual stats are just so polarizing because LeBron James can turn Daniel Gibson into "one of the most efficient players" in the league when we all know that's not true.

 

[jms493]

i dont even know what total QBR is.......:shifty:

 

[Warner2BruceTD]

i dont even know what total QBR is.......:shifty:

Well, nobody does. That's part of the problem.

I don't know if they ever exposed the formula, but when I looked for it last year, it was impossible to find.

 

[Juggernaut]

I haven't even heard of qbr until this thread, a few minutes of skimming some articles and its obvious this formula is flawed.

I'll unignore the Total QBR long enough to point out that I have found justification for ignoring it: Under the Total QBR, Tim Tebow performed better on Sunday than Aaron Rodgers.

Rodgers, whose Packers won at the Georgia Dome, completed 26 of 39 passes for 396 yards and two touchdowns. His Total QBR was 82.1.

Tebow, whose Broncos lost at home to the Chargers, completed four of 10 passes for 79 yards and a touchdown. And he ran the ball six times for 38 yards and a touchdown. And his Total QBR was 83.2.

http://deadspin.com/5848582/tim-tebow-gives-us-our-first-total-qbr-kerfluffle

People just need to watch the games to really see how good certain players are.

 

[FedEx227]

Well, nobody does. That's part of the problem.

I don't know if they ever exposed the formula, but when I looked for it last year, it was impossible to find.

They never did. Meaning it's probably stupid. Why hide it unless you know there are flaws?

 

[Champ]

According to this who were the worst QBs this weekend? Weeden and Vick?

 

[Obst]

According to this who were the worst QBs this weekend? Weeden and Vick?

Tannehill and Weeden were the two worst.

http://espn.go.com/nfl/story/_/id/6833215/explaining-statistics-total-quarterback-rating

Here is ESPN trying to explain what QBR is

 

[Villain]

http://espn.go.com/nfl/qbr

Top 5:
Matt Ryan
Mark Sanchez
Peyton Manning
RG3
Tom Brady

Bottom 5:
Russel Wilson
Vick
Ryan Fitzpatrick
Ryan Tannehill
Brandon Weedon

 

[FedEx227]

You know what a great way to explain what a stat is... showing us what goes into it. I'm guessing this guy who wrote the article, given his background, would be fine giving out the formula but ESPN wants to keep it under wraps so everyone doesn't steal it, of course.

 

[Senser81]

Tannehill and Weeden were the two worst.

http://espn.go.com/nfl/story/_/id/6833215/explaining-statistics-total-quarterback-rating

Here is ESPN trying to explain what QBR is

And you don't need a stat to tell you that Tannehill and Weeden suck. Which is another aspect of these overblown FootballOutisders statistical machinations...they are great at taking basic stats, putting them through some equation, then telling us what we already know while saying “Why isn’t anyone else getting this?”.

FootballOutsiders had an article which said the Chicago Bears had the best defense in the NFL last year despite being 14th in points and 17th in yards allowed. I asked how they came to that conclusion, and they said its because their statistics are "weighted", so they take into account the strength of the opposition. I think the Bears would have to be playing the 1998 Vikings every week for their stats to be "weighted" that much, so I asked them to show me the specific formula/equation they were using. "Oh, its too complex to explain". Got it.

 

[Rudi]

Any QB related stat that ever has Mark Sanchez in the top 5 is irrelevant. ESPN gunna ESPN.

 

[FedEx227]

To be fair, it's also one game, so Sanchez is top 5 FOR THIS WEEK, that doesn't necessarily mean he's going to continue that pace.

 

[Mogriffjr]

Any QB related stat that ever has Mark Sanchez in the top 5 is irrelevant. ESPN gunna ESPN.

he balled hard yesterday, stfu...

 

[Villain]

Any QB related stat that ever has Mark Sanchez in the top 5 is irrelevant. ESPN gunna ESPN.

19-27, 266 yards, 3 TD, 1 INT.

:kanye:

 

[Saluki]

Total QBR isn't any more ridiculous than the qb rating system that's been in place for years

Sent from my HTC myTouch 4g using Tapatalk 2

 

[JeremyHight]

I think if they just added a running and sack component to regular QB Rating, it could be improved dramatically. However, I think using just one stat to say a guy is good or not will never work.

 

[dell71]

Well, nobody does. That's part of the problem.

I don't know if they ever exposed the formula, but when I looked for it last year, it was impossible to find.

I watched the specials and read a number of articles on this last year when they were rolling it out. I've come to the conclusion that they didn't release the exact formula because there is no exact formula. The closest thing to explaining it is that the qb is given a quantifiable grade on every single play that isn't a hand-off. Plays on 3rd down and/or in the 4th qtr are worth more. At the end of the game, they add up all the plays and/or average them out to come up with that #.

Big big problem #1: it sounds really subjective. For instance and in theory, a perfect throw will result in a higher grade on a given play than a bad throw even if both result in completions with a whole range of grades possible depending on how good or bad the throw is. Well, that means that espn stat geek #1 might give someone a pretty good grade on one play while, stat geek #2 gives someone in another game a different score on an identical play. Like someone in an article I read said, we just have to trust that each play is scored correctly.

Big big problem #2: Coming up with a grade requires watching every play. Very few people have access to all the plays and no one not employed by espn has time to do this. Since no one else can figure it out, it will never catch on. It's simply espn tooting its own horn.

BTW, this isn't the first stat they've created. The much more widely used PER in basketball is credited to one of their guys.

Total QBR isn't any more ridiculous than the qb rating system that's been in place for years

Sent from my HTC myTouch 4g using Tapatalk 2

I disagree wholeheartedly for one simple reason: whether you love it or hate it any schmuck with a calculator can figure it out. If you're lazy, there are a number of websites that allow you to plug in the raw numbers and spit out the rating. It's something we can easily gauge. Sure, it's flawed but we know what we're getting.

Interesting article:

http://sportsillustrated.cnn.com/2011/writers/kerry_byrne/08/03/defending-qb-rating/index.html

If you don't read the whole thing the basic takeaway is this paragraph:

Few, if any, are more indicative of wins and losses than passer rating. Teams that posted a higher passer rating went 203-53 (.793) in 2010 and an incredible 151-29 (.839) after Week 5.