Jump to content

Why ProFootballFocus is not an authority.


FireChan

Recommended Posts

So I generally like PFF. I think they are good at their jobs and are relatively accurate with their predictions. But, this afternoon I saw perhaps the most egregious ranking I have seen on their site after week 7.

 

https://www.profootb...ders-in-review/

 

Sammy Watkins -0.2

 

Mike Evans +1.0

 

What?

 

Evans is actually ranked higher in play than Sammy.

 

Sammy has 35 receptions for 433 yards, with 4 TD's. 7 20+ yard catches.

 

Evans has 21 catches for 258 yards, with 2 TD's. 4 20+ yard catches.

 

I'd also like to point out that Evans has gotten a majority of his catches in 40 point blowouts.

 

If anyone could attempt to explain this to me, that'd be great.

Edited by FireChan
Link to comment
Share on other sites

PFF is buyer beware. From what I understand most of their "experts" are just internet junkies hired to watch game film and "grade" the players based upon one-on-one matchups, without any context of the play call or situation.

 

I'm not saying their stuff is complete junk, but it needs to be taken with a grain of salt.

Link to comment
Share on other sites

PFF is buyer beware. From what I understand most of their "experts" are just internet junkies hired to watch game film and "grade" the players based upon one-on-one matchups, without any context of the play call or situation.

 

I'm not saying their stuff is complete junk, but it needs to be taken with a grain of salt.

 

OK, then I will: their stuff is complete junk.

 

GO BILLS!!!

Link to comment
Share on other sites

Sammy had a real bad game where he dropped three passes. His grade was -3+, which offsets his grade from this past week or Detroit. I don't think anything is particularly off with that grade yet and going forward I expect it to trend positive. But of course it is imperfect. The minutia of the game they are trying to capture numerically can never be exactly right.

Link to comment
Share on other sites

Sammy had a real bad game where he dropped three passes. His grade was -3+, which offsets his grade from this past week or Detroit. I don't think anything is particularly off with that grade yet and going forward I expect it to trend positive. But of course it is imperfect. The minutia of the game they are trying to capture numerically can never be exactly right.

 

So three dropped passes outweighs 2 100 yard games, with 3 TD's, including the game winner? That's not just imperfect, that's ridiculous.

Edited by FireChan
Link to comment
Share on other sites

PFF is useful in certain respects; the way they track snap counts, subpackages, etc is very informative.

 

Their grading, however, is very subjective and based on a LOT of guesswork.

 

Maybe they'll get better at their jobs now that Chris Collinsworth is part owner; we shall see.

Link to comment
Share on other sites

PFF is useful in certain respects; the way they track snap counts, subpackages, etc is very informative.

 

Their grading, however, is very subjective and based on a LOT of guesswork.

 

Maybe they'll get better at their jobs now that Chris Collinsworth is part owner; we shall see.

 

Yeah, Collinsworth is never in his life at all subjective in his opinions or analysis.

Link to comment
Share on other sites

Analytics!

 

Exactly.

 

PFF deals in analytic statistical analysis. Nothing more. It is what it is. They can provide insight and also mislead if used exclusively to make a judgement. I'm not sure why anyone would put complete trust in ANY statistic or, for that matter, choose to ignore all stat analyses because they sometimes seem to come to strange conclusions.

 

For what it's worth, I believe PFF tweaks their formulas, from time to time, to try and get their stats to more closely resemble "objective" reality (if they don't, they should). The best advice I can give is, don't rely on any one source, or metric, to make decisions.

Link to comment
Share on other sites

Exactly.

 

PFF deals in analytic statistical analysis. Nothing more. It is what it is. They can provide insight and also mislead if used exclusively to make a judgement. I'm not sure why anyone would put complete trust in ANY statistic or, for that matter, choose to ignore all stat analyses because they sometimes seem to come to strange conclusions.

 

For what it's worth, I believe PFF tweaks their formulas, from time to time, to try and get their stats to more closely resemble "objective" reality (if they don't, they should). The best advice I can give is, don't rely on any one source, or metric, to make decisions.

 

Some on TBD use PFF's rankings exclusively to "appeal to authority."

Edited by FireChan
Link to comment
Share on other sites

If you ever tried to watch a game and do the analysis yourself about whether a guy did his job or not, whether it was a blitz or not, whether it was a drop or not, whether it was a QB pressure or not, it's an impossible task. There are no facts or right or wrong answers. It's enormously subjective on a lot of plays.

Link to comment
Share on other sites

I'll admit to being a stats geek. But I know stats have their limitations. Take baseball. New offensive metrics are superb, and GMs would be idiots not to rely on them. Advanced defensive metrics? Not so much. They're useful, but even the best sabermetricians will use a blend of new defensive metrics plus scouting reports -- that is, what your eyes tell you. That's how I see advanced football stats. PFF tells me that Richardson and Pears are two of the worst guards in the NFL this year. I saw that. But it also tells me Seantrel is one of the worst tackles. I missed that; I'll watch more closely -- I think I've been distracted by how awful Pears has been. For traditional "skill" positions I rely much more heavily on what I see (and what the All 22 shows) and the traditional stats -- things like YAC for receivers. Bottom line: PFF's stats look "advanced," but they're really just counting stats (a + if a lineman wins a one-on-one, a - if he loses, all in someone's subjective impression) -- in baseball, that's positively medieval and you'd be laughed off boards like this for citing to them.

Link to comment
Share on other sites

×
×
  • Create New...