The Mathews Question

Good call CJ...stewart got some runs late but CAM never ever got going.  By the way TAnnehill with a huge run on the read option on a big 4th down today agaist the Bares.

I tell you, on that play where Clay just missed sacking Cam and then chased him down the field to avoid a big play, the difference in effort between he and Nick Perry was mind boggling. Perry is rushing from the left and as soon as Newton leaves the pocket he just stopped playing. CMIII is on the ground and gets up to chase down the QB. I hope that play is reviewed over and over again in meetings.

Here's Pro Football Focus' grade of Matthews from yesterday.

 

Clay Matthews, OLB, +4.9

Breakdown: After nearly half a season’s worth of average play, Matthews broke out in a big way against a couple of overmatched tackles. Recording a sack and team-high six total pressures, the USC product looks to have regained his form after an injury-marred 2013 season.

Signature Play: It wasn’t all pass rushing for the multi-dimensional linebacker. At 11:04 in the opening quarter, Matthews shot off the line unblocked to take down Jonathan Stewart for a three yard loss.

Not directed toward you MichiganPacker, but how the F does this site come up with their grades?   I can't find the algorithm anywhere.  It's like all of a sudden this site is the authority on who can play and who can't, yet we know nothing about them.  

 

At least with baseball advanced stats, it's transparent. 

A +4.9 may or may not be better than a +2.5. But it is most likely, but not a certainty to be, better than a -3.9.

 

That is all you really need to know.

Can't wrap my head around how they grade QB's, but AJ Hawk scores poorly game in game out, so they must have something right.  

Actually, Hawk hasn't scored poorly from game to game. IIRC, he had some > +2 grades earlier this year. 

 

**edit**

Those were games when Jones was getting lots of PT.

 

 

Example, Dick Rodgers.   He had a great game blocking.   Saw several huge blocks watching live, yet he has a crap score.  Same with HaHa, sure he had one missed tackle in the open field, but had a really good game overall and has crap grades.

 

I don't get it. 

If we just look at the secondary grades, you would think they had an average day.   Reality is they played outstanding.   Heyward gets the best grade because he had a ball thrown right to him.  I think T-Buk could have made that play. 

 

Dix (-0.2)
Tramon +0.2
Burnett (-1.9)
House +0.6
Hyde +0.3
Heyward +1.8
Richardson +1.5

Ok... Beating a dead horse, but here is what I found.

 

4) The “Rules” of Grading

Because of the nature of the roles, each position is graded in a slightly different way and the definitions for each run on for many pages. Although we’re not going to publish our 30+ page document on how we do this, not least because that’s our IP, below are a few of the key principles in our grading methodology:

https://www.profootballfocus.com/about/grading/

 

AKA, We are not going to tell you who we are or how we grade, but trust us, we know what we are doing. 

 

Fk off. 

 

Aaron Rodgers was only a +0.6 yesterday?

Holy schit. SIX???

Brainded is absolutely brain dead...I mean, absolutely right. If they are using a secretive, proprietary format, it doesn't hold much value.

the bread and butter of PFF is their signature stats. No arguing with those numbers. 

 

The grades aren't really much more than an idea of how well a certain player played on a per play basis. Keep in mind, in order to get a positive grade a player has to do something extraordinary. If they do nothing more than their job (what is expected) on a certain play they get a neutral (0.0) grade. 

 

At the end of the year, the aggregate grades for each player are a good indicator of who played well that year. 

Here is one of their problems, they don't know what the players job is on any given play. PFF does not know the plays called, audibles, defensive call, or adjustments while the play is occurring.

 

Rodgers with a +0.6 and Peppers with a -1.1 yesterday are prime examples of their inability to properly grade a player. According to their grading, Rodgers performed just slightly better than expectations and Peppers performed below expectations - almost Hawk like it seems. 

 

Think about the amount of staff it must take to get this done.   Lets say there are 60 snaps in a football game.   Now, on each of those snaps there are 22 players that need to be evaluated.   Lets say an avg play lasts 5 seconds, but to really evaluate a player you probably need to spend more than 5 secs so lets say 30 secs to evaluate a player on a given play.  Ya know, you might want to watch twice and consider things like down, distance, play call and so on. 

 

So here is what he have.  ((22 players x 30 seconds) x 60 snaps) = 39600 seconds or 11 hours.

 

So they would need a staff of about 15 (15 games a week?) working 11 hours, non stop,  on Sunday night to get these grades out by Monday morning.    Either that or double the staff to cut it down to 5.5 hours non stop.   Of course, the other option is they are not spending that much time. 

 

 

 

90 percent of what you see on the current site has been completed by either Ben or I - Neil Hornsby, PFF founder