ESPNs QBR explained | FinHeaven - Miami Dolphins Forums

ESPNs QBR explained

Trying to use an efficiency stat in football just doesn't work for me.

Efficient to me is 23/32 300 yards a 2 touchdowns.

ESPN's QBR could give 2 different scores for this same stat line depending on factors usually out of the qbs hands. It's not baseball and basketball. It's football.
 
The problem is the subjectivity. Do you, or can you, trust a person who has inherent biases to look at everything through an objective lens and assign these numbers? If the rater grew up a Chicago Bears fan, can you trust him to be critical of the Bears and fair with the Packers, Lions, and Vikings? Additionally, how can you judge the success of a play if you aren't the one calling the plays? Some coaches call plays like dump offs to the flat a few times in the first half, despite knowing they will be minimally successful, in order to set up a play action or screen later on in the game. How does ESPN know the intentions of the play in cases like that? Those one yard dump offs probably count against a QB if looking at them, but result in a TD later when the defense bites on a playfake to that same side and a screen gets a TD to the other side.

For all the criticisms that the traditional QB rating gets, it is objective. It takes raw stats only, gives you the number based on those stats, and lets the person reading that information come to their own decisions. It's empirical and it's clean, which is the most you can hope for with stats.
 
The problem is the subjectivity. Do you, or can you, trust a person who has inherent biases to look at everything through an objective lens and assign these numbers? If the rater grew up a Chicago Bears fan, can you trust him to be critical of the Bears and fair with the Packers, Lions, and Vikings? Additionally, how can you judge the success of a play if you aren't the one calling the plays? Some coaches call plays like dump offs to the flat a few times in the first half, despite knowing they will be minimally successful, in order to set up a play action or screen later on in the game. How does ESPN know the intentions of the play in cases like that? Those one yard dump offs probably count against a QB if looking at them, but result in a TD later when the defense bites on a playfake to that same side and a screen gets a TD to the other side.

For all the criticisms that the traditional QB rating gets, it is objective. It takes raw stats only, gives you the number based on those stats, and lets the person reading that information come to their own decisions. It's empirical and it's clean, which is the most you can hope for with stats.

The part about play calls is spot on..
 
Anything that ESPN has it's hands in on is garbage and should be thrown out. I laugh at anyone who tries to use this **** rating method.
 
I've always said you take QBR QB Rating, Rushing Yards, PFF grades and whatever other metric you want to use to analyze things and that weighs out a 50% of the total value. The other 50% is the ol' eyeball test, tape, film breakdowns, analytics etc...the equation is as follows objective grade + subjective grade = Total Performance Grade.

The more you plug into each side of the equation the better your Total Performance Grade will reflect reality.
 
Once again, at no point do they actually give you the exact formula used for QBR. For instance, what is the weighting they put toward quarterbacks of teams that are going to appear on an upcoming Monday Night game they need to hype up? :) They don't tell you that.

There is also the absurdity I have repeatedly pointed out that the 'clutchness' part of the calculation TAKES POINTS AWAY FROM A QUARTERBACK IF HIS DEFENSE CHOKES AFTER HE MAKES A GOOD PLAY.

I'm all for people using advanced statistics to try to dig deeper and see which players might be under or overvalued in the grand scheme of things. The problem with ESPN's QBR... is that it's just not a good metric.
 
QBR explained - a convoluted number created by ESPN to self-promote its brand
 
When Sanchez had an "efficient" QBR with like 2 picks and 50% completion rate, I knew this stat was garbage
 
When Sanchez had an "efficient" QBR with like 2 picks and 50% completion rate, I knew this stat was garbage

Absolutely. One of the key points of advanced statistics in football is to try to better isolate a player's performance to minimize the impact of variables like his surrounding unit, the teams he faced, etc.

So for ESPN's QBR statistic to literally add and subtract points from the player's rating based on the performance of his defense, a factor over which he has absolutely no control whatsoever is completely insane to me.

ESPN ****ing sucks and they can't crash and burn quickly enough.


---

For what it's worth, I still believe the passer rating statistic is a pretty good metric. It doesn't necessarily tell you how good the quarterback is, but it tells you how well the quarterback's passing offense functioned. Outside of an over-emphasis on completion percentage (which I think is reaching the point of irrelevancy in the era of screen passes) it also correlates with winning football games.
 
ESPN thought to themselves that they had to compete with this advanced statistics revolution, so they introduced re****ed statistics with the QBR stupid **** revolution...
 
Absolutely. One of the key points of advanced statistics in football is to try to better isolate a player's performance to minimize the impact of variables like his surrounding unit, the teams he faced, etc.

So for ESPN's QBR statistic to literally add and subtract points from the player's rating based on the performance of his defense, a factor over which he has absolutely no control whatsoever is completely insane to me.

Whats explained in that link I have read before, years ago, and I have never seen them say anything remotely close to what you're stating and after clicking this new link and reading their explanation there is nowhere in there that implies this, either.

There is nothing in there that says the QB's rating will be affected by what his own defense does on the next drive - unless you mean how the other QB's scores are graded against his in relation. If Aaron Rodgers defense gives up a game winning TD in the 4th it won't adjust Aaron Rodgers QBR score other than it might bring the opposing QB's scores highers in relation to his - which makes perfect sense... Aaron Rodgers was in 1st and now he is in 2nd - so yea he does drop because the other QB just got the better score - if all other stat lines were equal - how does this not make sense?.


I think its very popular and "fun" to **** on ESPN for basically everything - I've never understood why everyone (here at least) hates on the network SO much - for literally everything. I consider the QBR rating system better than the old system. Not that it's brilliant, but it factors in things like sacks, penalties, defensive holds... it factors in an INT as a hailmary at the end of the first half is not the same value as a INT in the 4th in a tie game - that's advanced statistics that the old system doesn't have. I also like that the measure is based out of 100 and is in relation to every other QB in the league.

I know everyone here loves to hate on ESPN, but their QBR metric is actually not bad.
 
ESPN's QBR is a classic example of why they are in the position they are currently in. They got too big for themselves and thought they could just play sports God. No.
 
http://www.espn.com/blog/statsinfo/...-calculated-we-explain-our-quarterback-rating

My problem with stats is they do not tell you when something happened. Putting each play into context helps fix that. But I could easily believe after reading the explanation that the QBs teammates affect his score. So basically if your team stinks it is harder to get a higher score. I believe that too.


This is how they came up with it:

n29Pqd.gif
 
Back
Top Bottom