The Brian Hoyer thread...

[bob2the2nd]

I notice you keep avoiding my question...perhaps it's because the answer doesn't jive with your agenda.

Part of my faith in their system revolves around who uses their numbers. Top media, teams, players. There must be something behind it if people closest to the game find it useful. bob, if you had a proprietary formula that is your bread and butter would you post it out there for free so anyone could use it? Doubtful. They outline their methodology extensively on their site.

It's not perfect, as they freely admit. What it provides is an idea of how well someone performed throughout the game. If you want to argue that a -3.0 is not 10% worse than a -2.7, I'll grant you that argument. Don't get too hung up on the specifics. I use it to validate what I saw on the field...and it usually does that.

how do you know "top media, teams, players" use it?

 

[CosmoKramer]

@jlj3184So when you show me X website, that charges for its services, with absolutely no methodology, and total lack of openness. youll excuse when i believe they are full of shit.

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

It's been posted before, but after reading your post I can't help but re-post it just for you.

The methodology is pretty easy bud, they watch every play and assign every single player a grade. It's not exactly easy, you're grading individuals in a complex team sport after all, but most intelligent fans have accepted PFF's numbers as a better gauge of performance than basic stats.

 

[jlj3184]

how do you know "top media, teams, players" use it?

You are beginning to become insufferable on this issue. Let's just move on.

 

[AZ_]

This has seemingly been ignored the first time, so I'll post it again:

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

It's not an overly complicated system.

As to whether or not PFF is flat out lying when saying NFL teams use their methodology, I'd imagine they wouldn't get very far using the NFL name and expecting to continue as a legitimate source.

But keep pretending everyone else is insufferable.

 

[bob2the2nd]

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

It's been posted before, but after reading your post I can't help but re-post it just for you.

The methodology is pretty easy bud, they watch every play and assign every single player a grade. It's not exactly easy, you're grading individuals in a complex team sport after all, but most intelligent fans have accepted PFF's numbers as a better gauge of performance than basic stats.

sorry cosmo but general bullshit with absolutely no real explanation doestn cut it for me. where are their formulas. where are the explanation of guidelines that their analysts use. and thats the point, you have no idea, because they dont tell you.

i dont care that "most intelligent fans" have accepted their numbers, because for all i know they are full of shit. Until they release their methodology, and their formulas for analysis i couldnt care less. Its nothing more than a wizard behind a curtain, making shit up.

 

[bob2the2nd]

You are beginning to become insufferable on this issue. Let's just move on.

so, you dont know. instead youre just willing to take some random website X's self description as a reason to trust them?

ive got some great ocean front property in kansas for you. I cant tell you where it is but its dirt cheap.

 

[AZ_]

sorry cosmo but general bullshit with absolutely no real explanation doestn cut it for me. where are their formulas. where are the explanation of guidelines that their analysts use. and thats the point, you have no idea, because they dont tell you

Since you once again decided not to read and just continue with this nonsense...

The Basics

Each grade given is between +2.0 and -2.0, with 0.5 increments and an average of 0.0. A positive intervention in the game earns a positive grading and vice-versa. Very (very) few plays draw a +/-2.0 rating.

Another way of looking at the “0” grade is viewing it as the “expected” grade for an NFL player. We expect an NFL quarterback to accurately throw a 6-yard curl route against off-coverage on 1st-and-10, and whether or not his WR gets tackled for a 6-yard gain or breaks a tackle for a 70-yard touchdown, the quarterback’s grade remains the same.

 

[The Voice]

so, you dont know. instead youre just willing to take some random website X's self description as a reason to trust them?

ive got some great ocean front property in kansas for you. I cant tell you where it is but its dirt cheap.

If they're charging for the service, why the hell would they give you the formula?

I'm not sure coke is better the Pepsi, give me the ingredients and step by step instructions how to make the syrup.


Sent from my iPhone using Tapatalk at your local pub.

 

[bob2the2nd]

Since you once again decided not to read and just continue with this nonsense...

boobie i have no idea why you quoted this. it literally gives almost no insight into how they evaluate, other than on a 6 yard curl their analyst would give a QB a 0.

what if its 7 yards? what if its 5? what if the receiver drops the ball?

my point is and continues to be this is a bias way of evaluating with no published methods. Which means none of us have any idea of if they are randomly making these numbers up or not

 

[bob2the2nd]

If they're charging for the service, why the hell would they give you the formula?

I'm not sure coke is better the Pepsi, give me the ingredients and step by step instructions how to make the syrup.


Sent from my iPhone using Tapatalk at your local pub.

thats my point the Voice. they charge for their service to hide their methodology to make money and take advantage of people.

it could all just be, totally made up.

 

[The Oi]

@bob2the2nd getting flambozzled in here.

 

[jlj3184]

so, you dont know. instead youre just willing to take some random website X's self description as a reason to trust them?

ive got some great ocean front property in kansas for you. I cant tell you where it is but its dirt cheap.

I like how you keep ignoring my original question and posting non-sequiturs. I've bolded it below, so you can find it for sure this time.

Do these two plays represent equal Quarterback play?
Play 1: QB takes snap, 7-step drop, evades two rushers, throws 40-yard pass to WR over the outside shoulder in the corner of the endzone while beating double coverage.
Play 2: QB takes snap, 3-step drop, perfect pocket, throws 3-yard slant pass to WR who breaks 3 tackles and runs 37 yards to the end zone.

Making snide remarks and belittling people is not a good look. Do I trust PFF when they say that NFL teams use their stats? Yes. Do I trust them when they say players and agents use their stats? Yes. Do I know for a fact media members use their stats? Yes. Do I have receipts to prove the first two? No, but the fact that someone is willing to stake their reputation by making a claim like that is enough for me.

I can't prove 9 out of 10 dentists recommend Crest, either.

 

[bob2the2nd]

I like how you keep ignoring my original question and posting non-sequiturs. I've bolded it below, so you can find it for sure this time.

Do these two plays represent equal Quarterback play?
Play 1: QB takes snap, 7-step drop, evades two rushers, throws 40-yard pass to WR over the outside shoulder in the corner of the endzone while beating double coverage.
Play 2: QB takes snap, 3-step drop, perfect pocket, throws 3-yard slant pass to WR who breaks 3 tackles and runs 37 yards to the end zone.

Making snide remarks and belittling people is not a good look. Do I trust PFF when they say that NFL teams use their stats? Yes. Do I trust them when they say players and agents use their stats? Yes. Do I know for a fact media members use their stats? Yes. Do I have receipts to prove the first two? No, but the fact that someone is willing to stake their reputation by making a claim like that is enough for me.

I can't prove 9 out of 10 dentists recommend Crest, either.

no. they shouldnt. but tell me using numbers how you would rank these two plays. then tell me, using numbers, how PFF would rank these two plays? Then tell me what if the WR breaks only 1 tackle and scores, what if the WR is unguarded and scores. Is that more credit for the QB for finding the player? or more credit for the WR not being covered?

and a question for you. why do you trust a web site that doestn clearly explain how they evaluate, to the T. but yet charges for their services?


and of course 9 out of 10 dentists recommend crest, you just have the ask the right ones. (no, im not joking)

 

[The Voice]

thats my point the Voice. they charge for their service to hide their methodology to make money and take advantage of people.

it could all just be, totally made up.

It's proprietary information. If they posted their formula's then who the hell would subscribe?

I want the exact recipe to make coca cola, just so I can make sure it's the best.

Sent from my iPhone using Tapatalk at your local pub.

 

[jlj3184]

no. they shouldnt. but tell me using numbers how you would rank these two plays. then tell me, using numbers, how PFF would rank these two plays? Then tell me what if the WR breaks only 1 tackle and scores, what if the WR is unguarded and scores. Is that more credit for the QB for finding the player? or more credit for the WR not being covered?

and a question for you. why do you trust a web site that doestn clearly explain how they evaluate, to the T. but yet charges for their services?


and of course 9 out of 10 dentists recommend crest, you just have the ask the right ones. (no, im not joking)

PFF would (probably) grade the first play as a +1 or +1.5 and the second as a 0. The WR would probably get a 0 on the first play and a +1 or +1.5 on the second. A play like Beckham's last night would get a +2.0, while Eli would probably get a +.5 or +1. Is there going to be variance, absolutely. Is the system perfect, no.

You know what QB Rating a QB would have based on the two above plays? 158.3 for the first and 158.3 for the second. That is a HUGE flaw in the system, and why people view PFF as more comprehensive and a better reflection of individual play. This is the point we've been trying to make when you cite things like 320 yards and a TD as the metric of a QB having a good game. If you want to argue he should have had a -5 instead of -5.5 from last week, that's fine. That is called variance. I would probably grade plays slightly different than you would. Over 70+ snaps, there is going to be some discrepancy in our numbers. There are plenty of imperfections in PFF, as they'll freely admit. When you understand that and take it for what it is, it allows a complete picture of an individual's game. ESPN QBR does a slightly better job accounting for game flow, but doesn't account for the noticeable difference in the amount of skill required like PFF does. It's not perfect, but it's the best out there that us "normal" folks can get our hands on.

I trust them because I've done my research and read testimonials from people whose livelihood depend on this sort of analysis who call it the best out there. I have reservations about some of their stuff (my job is inherently quantitative), but I understand the need to add qualitative measures to something as complex as player analysis in football.