sbk92
2
- Messages
- 12,134
- Reaction score
- 6
Bob Sturm / Contributor
Sam Monson is a member of ProFootballFocus.com - a website that I really appreciate and enjoy. Every week, along with the staff of PFF, he grades every NFL player and play in great detail. The depth of their independent NFL research is admirable and I use the site frequently when doing my Cowboys and NFL analysis. I chatted with Monson recently and wanted you to have a chance to review his thoughts on their research and grading, as well as your Cowboys:
(All the untitled commentary is Bob Sturm)
8:51 Sam, just respond when you are ready. Thanks
8:52 [Comment From Sam Monson PFF: ]
OK I'm good to go
8:53 trust me. I have no idea how this software is going to work out. But, let's try it.
8:53 [Comment From Sam Monson PFF: ]
We've actually used it at PFF for some early staff meetings, worked ok
8:53 First, please tell me as much as you can about how you guys do what you do. I am fascinated by the project.
8:54 And I don't mean to be so vague. I assume it takes an incredible amount of man hours.
8:54 [Comment From Sam Monson PFF: ]
OK, well first off we have a lot of information on the site about what we do, but long story short we watch every game on a play by play basis and grade every player on each snap.
8:55 [Comment From Sam Monson PFF: ]
Each game takes something in the region of 10-15 man hours (depending on snaps and a whole host of other factors), and we divide it into two - Analysis and Player Participation.
8:55 This must be a sizable staff
8:55 [Comment From Sam Monson PFF: ]
Analysis is the play by play grading of everybody, and PP is recording who lined up where, and what they did (blitzing, dropping into coverage etc)
8:56 [Comment From Sam Monson PFF: ]
All of the analysis is done by 3 people combined. The main body of the PP is done by a different 3 people, and then we have a lot of volunteers helping out double checking PPs (when we double check PPs our accuracy goes up to between 99.9% and 100% as verified by a couple of NFL teams
8:57 What obviously makes what you guys do awesome is that difficult proposition of assigning blame or credit on a simple running play to 11 different guys. How is that done? For instance, what I run into is the issue of knowing the intent of each player on a given play. How do you reconcile that?
8:58 [Comment From Sam Monson PFF: ]
Well we start by looking at where the run went. Then you watch the linemen and see if the run was supposed to go somewhere else. So you will for example see a clear hole opening up, but maybe the RG is beaten by the DT, which forces the HB to cut away from that hole. Then we work out from the point of attack
8:59 And with only 3 graders, that satisfies my issues about consistency from game to game. Your feedback from the league is positive?
9:00 [Comment From Sam Monson PFF: ]
The main point though is that it's extremely hard to tell anything complex from a live viewing of a play without looking specifically at where you know the run is going. We do it after the game so we get the benefit of endless replays and pausing footage etc. That's why it takes so long
9:00 [Comment From Sam Monson PFF: ]
Yes, the 3 graders is a conscious effort to keep our grading consistent. We have a rigid set of guidelines and grading techniques, but even so it takes so long to train somebody up so that they are thinking the same that the fewer people we can keep analyzing the better.
9:01 [Comment From Sam Monson PFF: ]
As for feedback, pretty much all of our feedback from within NFL circles has been positive. Players and team personnel have both been impressed with the accuracy of it
9:02 For the sake of those who are just getting used to your site (like me), please explain briefly how your grading numbers should read. I see Doug Free at 17.9 and Marc Colombo at -34.1 and I understand some of that and wonder what I really need to process to see what you are telling me there.
9:03 besides the obvious, of course.
9:04 It seems that most players are far closer to 0 than those two extreme cases.
9:04 [Comment From Sam Monson PFF: ]
Well the first thing to note is that the figures aren''t comparable across positions. So a -30 for a QB isn't the same as a -30 for an OT. Essentially that figure is a cumulative total of that player's grades on every snap in a game, season etc that has a figure added or subtracted to it to make the league average for that position 0.
9:05 [Comment From Sam Monson PFF: ]
The basic theory we work from is that the 'average' play on any given snap gets a 0. Anything you would expect someone to do, no more, no less.
9:06 And does every positive result in a negative grade to his opposition?
9:07 [Comment From Sam Monson PFF: ]
Not always, but almost all of the time. Most blocks are simple adversarial encounters, so a positive is met with an equal and opposite negative, but there are exceptions.
9:09 OK, the first thing I might need explained from a Cowboys standpoint is the overall grade of the team on the opening page of your premium stats. Do I read that they are grading at 3rd overall in the NFL? 74.7 overall, I believe?
9:09 Or 3rd offensively, and top 10 defensively. forgive me
9:11 [Comment From Sam Monson PFF: ]
That's 3rd overall for Offense, and 10th for D. We said for a while this season that the Cowboys weren't actually playing that badly, but were just finding ways to lose games. They were by and large being undone by a couple of big plays a game while playing moderately well the rest of the time. And that was when they were losing. When they picked up under Garrett they continued to grade well.
9:11 So, from an analysis standpoint, you would call them a strong 6-10 versus a team that is miles away from contention?
9:13 [Comment From Sam Monson PFF: ]
They're certainly not far away from being able to challenge, we saw that with the way they finished the season under new coaching, even without Tony Romo (or even =Kitna!) at QB. Obviously there are some holes, but what team has a flawless roster?
9:14 OK. When I look at the games with my eyes, I see rather severe OL issues. Your grades indicate RT is a major weakness, but RG with Leonard Davis and C with Andre Gurode both received some level of positive marks. Give me your thoughts on the OL
9:19 [Comment From Sam Monson PFF: ]
We had an interesting debate amongst the staff about Gurode and the Pro-Bowl. As everyone knows, he still has major problems simply snapping the ball in shotgun, and those plays can be killer. But the bottom line is that even with those major negative plays factored in to our grades, he still ends up with a very strong grade over the season. He's capable of very bad games, but over the season he is a good blocker, albeit one with a penchant to snap the ball nowhere near the QB. Davis is a pretty strong run blocker, and good enough as a pass protector to hold up. When an offensive lineman is a better run blocker than they are a pass protector, they tend to look worse than they actually are in the eyes of people just watching the games, because sacks and pressures are easy to spot, you don't tend to notice good run blocks because you're following the runner through the hole.
9:19 Meanwhile, Doug Free appears to be considered the best run blocking tackle in the NFL. And, overall, very strong. It seems that PFF would have no issue paying Free his money as he approaches Free Agency.
9:21 [Comment From Sam Monson PFF: ]
We love Free as a run blocker, but teams pay left tackles to protect the QB's blind side, and Free was our 15th ranked OT in pass protection, allowing the QB to be knocked to the ground 12 times on the season. I guess it depends how much they think he can improve in that area (still being a relatively young starter), and how much they covet his run blocking
9:22 It seemed to me Colombo is still not right health wise, but in your league rankings, his 2010 was pretty much as bad as it was across the whole NFL at RT, right?
9:24 [Comment From Sam Monson PFF: ]
Yes, Colombo just wasn't at the races in 2010. I'm sure injuries have played a role, and with his age it's tough to see a road back for him, but in terms of play out on the field he was right there at the bottom of the rankings.
9:25 One last OL question. Over the years, Garrett has loved the G power game with a pulling Kosier as he pulls to the right. When grading where a run goes, how do we handle the pulling guards because his run block grades were quite low. My perception may be guilty of remembering 2007 and 2009 too much.
9:27 [Comment From Sam Monson PFF: ]
Pull blocks go down in a player's run block grade (as opposed to screen block or anything of that nature). Essentially we watch where he pulls to, and decide how he fared blocking the first player he attempts to block. If he pulls right and leads through the hole, does he get stood up by a linebacker? Does the LB get around him to make the stop, or does he clear the LB out of the hole and allow a successful run.
9:28 I am most interested in Miles Austin and your evaluation of him. His grade was negative for the season, with much of the red numbers being attributed to run blocking skills, although a neutral number in pass catching. His grade is even lower than Roy Williams. Shocking!
9:31 [Comment From Sam Monson PFF: ]
A lot of what we do allows people to put their own weighting on the numbers. Plenty of people don't care whether a receiver blocks well or not if he can get 100 yards a game and haul in a pair of touchdowns. We grade it and allow people to sort the rankings as they choose (click at the top of a column to sort by that column on the site). Austin didn't have a good year run blocking and Roy Williams for all his faults, isn't a bad blocker. That being said, if we had to choose one of them to keep, we wouldn't think twice about keeping Austin
9:32 One thing I love about PFF is your QB charts showing the success to certain parts of the field. It revealed what was obvious to most of us - Kitna, while admirable in certain areas, can only throw the ball to a limited space on the field. Between the numbers or shallow. I believe it shows that Romo is much more difficult to defend as DBs must respect the area he can attack. Very revealing.
9:33 [Comment From Sam Monson PFF: ]
There's a reason Kitna has never been a franchise QB. He can move the ball, and he can do enough to win games, but he's not the weapon a guy like Romo is and can be
9:34 This may change for each position, but what is the maximum a player can receive or lose on 1 play via your system?
9:37 [Comment From Sam Monson PFF: ]
Essentially we grade from +2 to -2 in 0.5 increments, but the 2s are for crazy-good plays that win games essentially and are very hard to get. Certain positions can attain higher for compound plays. Marshawn Lynch's run against the Saints for example got a +2.5, and a defensive player beating his blocker immediately, making a sack, forced fumble, picking it up and running it in for a TD can get a +2.5
9:38 It appears looking at your grades for the defense that the secondary remains the weak link here. In particular Ball, Jenkins, and Newman.
9:40 [Comment From Sam Monson PFF: ]
Absolutely. This is a case where what you see with the naked eye in a live watching does tell the real story. Mike Jenkins had a hugely poor season after looking promising, and the secondary was a major issue
9:41 Anthony Spencer is controversial around these parts. Everyone wants a dominating pass rusher on the edge of the 3-4, and his sack totals just don't make the eyes pop. However, his play vs the run is notable. So what do you think of Spencer overall compared to players across the league at his spot?
9:42 [Comment From Sam Monson PFF: ]
Spencer had a bit of a down year rushing the passer this season, but last season he was excellent in that regard, despite the sack numbers. His play vs the run is very good, and though he may not be an All-Pro, he is a very good player, and a real weapon opposite Ware
9:43 Where are you on the Jay Ratliff debate. Because of his size, I really would like to see him DE because I believe in the NT run stuffer. Your run grades for him seem to back this up.
9:46 [Comment From Sam Monson PFF: ]
The thing about Ratliff is that he isn't a traditional 2-gap NT. He's a 1-gap player who is best shooting gaps and blowing plays up in the backfield. But he's shown in the past ('09 especially) that he can be a force from NT doing just that. This year wasn't a good season for him, especially vs the run, but we wouldn't write him off as a NT based just on 2010
9:46 Last question before I let you go: Top 2 or 3 things on the Cowboys that must be addressed this offseason?
9:48 [Comment From Sam Monson PFF: ]
We've talked about them so far: Secondary help, right tackle, more secondary help. Settling on a running back probably wouldn't hurt either
9:48 Thanks, Sam. I really enjoy your project and recommend it highly. Would love to do this again sometime soon.
9:48 [Comment From Sam Monson PFF: ]
No problem, a pleasure good sir.
Sam Monson is a member of ProFootballFocus.com - a website that I really appreciate and enjoy. Every week, along with the staff of PFF, he grades every NFL player and play in great detail. The depth of their independent NFL research is admirable and I use the site frequently when doing my Cowboys and NFL analysis. I chatted with Monson recently and wanted you to have a chance to review his thoughts on their research and grading, as well as your Cowboys:
(All the untitled commentary is Bob Sturm)
8:51 Sam, just respond when you are ready. Thanks
8:52 [Comment From Sam Monson PFF: ]
OK I'm good to go
8:53 trust me. I have no idea how this software is going to work out. But, let's try it.
8:53 [Comment From Sam Monson PFF: ]
We've actually used it at PFF for some early staff meetings, worked ok
8:53 First, please tell me as much as you can about how you guys do what you do. I am fascinated by the project.
8:54 And I don't mean to be so vague. I assume it takes an incredible amount of man hours.
8:54 [Comment From Sam Monson PFF: ]
OK, well first off we have a lot of information on the site about what we do, but long story short we watch every game on a play by play basis and grade every player on each snap.
8:55 [Comment From Sam Monson PFF: ]
Each game takes something in the region of 10-15 man hours (depending on snaps and a whole host of other factors), and we divide it into two - Analysis and Player Participation.
8:55 This must be a sizable staff
8:55 [Comment From Sam Monson PFF: ]
Analysis is the play by play grading of everybody, and PP is recording who lined up where, and what they did (blitzing, dropping into coverage etc)
8:56 [Comment From Sam Monson PFF: ]
All of the analysis is done by 3 people combined. The main body of the PP is done by a different 3 people, and then we have a lot of volunteers helping out double checking PPs (when we double check PPs our accuracy goes up to between 99.9% and 100% as verified by a couple of NFL teams
8:57 What obviously makes what you guys do awesome is that difficult proposition of assigning blame or credit on a simple running play to 11 different guys. How is that done? For instance, what I run into is the issue of knowing the intent of each player on a given play. How do you reconcile that?
8:58 [Comment From Sam Monson PFF: ]
Well we start by looking at where the run went. Then you watch the linemen and see if the run was supposed to go somewhere else. So you will for example see a clear hole opening up, but maybe the RG is beaten by the DT, which forces the HB to cut away from that hole. Then we work out from the point of attack
8:59 And with only 3 graders, that satisfies my issues about consistency from game to game. Your feedback from the league is positive?
9:00 [Comment From Sam Monson PFF: ]
The main point though is that it's extremely hard to tell anything complex from a live viewing of a play without looking specifically at where you know the run is going. We do it after the game so we get the benefit of endless replays and pausing footage etc. That's why it takes so long
9:00 [Comment From Sam Monson PFF: ]
Yes, the 3 graders is a conscious effort to keep our grading consistent. We have a rigid set of guidelines and grading techniques, but even so it takes so long to train somebody up so that they are thinking the same that the fewer people we can keep analyzing the better.
9:01 [Comment From Sam Monson PFF: ]
As for feedback, pretty much all of our feedback from within NFL circles has been positive. Players and team personnel have both been impressed with the accuracy of it
9:02 For the sake of those who are just getting used to your site (like me), please explain briefly how your grading numbers should read. I see Doug Free at 17.9 and Marc Colombo at -34.1 and I understand some of that and wonder what I really need to process to see what you are telling me there.
9:03 besides the obvious, of course.
9:04 It seems that most players are far closer to 0 than those two extreme cases.
9:04 [Comment From Sam Monson PFF: ]
Well the first thing to note is that the figures aren''t comparable across positions. So a -30 for a QB isn't the same as a -30 for an OT. Essentially that figure is a cumulative total of that player's grades on every snap in a game, season etc that has a figure added or subtracted to it to make the league average for that position 0.
9:05 [Comment From Sam Monson PFF: ]
The basic theory we work from is that the 'average' play on any given snap gets a 0. Anything you would expect someone to do, no more, no less.
9:06 And does every positive result in a negative grade to his opposition?
9:07 [Comment From Sam Monson PFF: ]
Not always, but almost all of the time. Most blocks are simple adversarial encounters, so a positive is met with an equal and opposite negative, but there are exceptions.
9:09 OK, the first thing I might need explained from a Cowboys standpoint is the overall grade of the team on the opening page of your premium stats. Do I read that they are grading at 3rd overall in the NFL? 74.7 overall, I believe?
9:09 Or 3rd offensively, and top 10 defensively. forgive me
9:11 [Comment From Sam Monson PFF: ]
That's 3rd overall for Offense, and 10th for D. We said for a while this season that the Cowboys weren't actually playing that badly, but were just finding ways to lose games. They were by and large being undone by a couple of big plays a game while playing moderately well the rest of the time. And that was when they were losing. When they picked up under Garrett they continued to grade well.
9:11 So, from an analysis standpoint, you would call them a strong 6-10 versus a team that is miles away from contention?
9:13 [Comment From Sam Monson PFF: ]
They're certainly not far away from being able to challenge, we saw that with the way they finished the season under new coaching, even without Tony Romo (or even =Kitna!) at QB. Obviously there are some holes, but what team has a flawless roster?
9:14 OK. When I look at the games with my eyes, I see rather severe OL issues. Your grades indicate RT is a major weakness, but RG with Leonard Davis and C with Andre Gurode both received some level of positive marks. Give me your thoughts on the OL
9:19 [Comment From Sam Monson PFF: ]
We had an interesting debate amongst the staff about Gurode and the Pro-Bowl. As everyone knows, he still has major problems simply snapping the ball in shotgun, and those plays can be killer. But the bottom line is that even with those major negative plays factored in to our grades, he still ends up with a very strong grade over the season. He's capable of very bad games, but over the season he is a good blocker, albeit one with a penchant to snap the ball nowhere near the QB. Davis is a pretty strong run blocker, and good enough as a pass protector to hold up. When an offensive lineman is a better run blocker than they are a pass protector, they tend to look worse than they actually are in the eyes of people just watching the games, because sacks and pressures are easy to spot, you don't tend to notice good run blocks because you're following the runner through the hole.
9:19 Meanwhile, Doug Free appears to be considered the best run blocking tackle in the NFL. And, overall, very strong. It seems that PFF would have no issue paying Free his money as he approaches Free Agency.
9:21 [Comment From Sam Monson PFF: ]
We love Free as a run blocker, but teams pay left tackles to protect the QB's blind side, and Free was our 15th ranked OT in pass protection, allowing the QB to be knocked to the ground 12 times on the season. I guess it depends how much they think he can improve in that area (still being a relatively young starter), and how much they covet his run blocking
9:22 It seemed to me Colombo is still not right health wise, but in your league rankings, his 2010 was pretty much as bad as it was across the whole NFL at RT, right?
9:24 [Comment From Sam Monson PFF: ]
Yes, Colombo just wasn't at the races in 2010. I'm sure injuries have played a role, and with his age it's tough to see a road back for him, but in terms of play out on the field he was right there at the bottom of the rankings.
9:25 One last OL question. Over the years, Garrett has loved the G power game with a pulling Kosier as he pulls to the right. When grading where a run goes, how do we handle the pulling guards because his run block grades were quite low. My perception may be guilty of remembering 2007 and 2009 too much.
9:27 [Comment From Sam Monson PFF: ]
Pull blocks go down in a player's run block grade (as opposed to screen block or anything of that nature). Essentially we watch where he pulls to, and decide how he fared blocking the first player he attempts to block. If he pulls right and leads through the hole, does he get stood up by a linebacker? Does the LB get around him to make the stop, or does he clear the LB out of the hole and allow a successful run.
9:28 I am most interested in Miles Austin and your evaluation of him. His grade was negative for the season, with much of the red numbers being attributed to run blocking skills, although a neutral number in pass catching. His grade is even lower than Roy Williams. Shocking!
9:31 [Comment From Sam Monson PFF: ]
A lot of what we do allows people to put their own weighting on the numbers. Plenty of people don't care whether a receiver blocks well or not if he can get 100 yards a game and haul in a pair of touchdowns. We grade it and allow people to sort the rankings as they choose (click at the top of a column to sort by that column on the site). Austin didn't have a good year run blocking and Roy Williams for all his faults, isn't a bad blocker. That being said, if we had to choose one of them to keep, we wouldn't think twice about keeping Austin
9:32 One thing I love about PFF is your QB charts showing the success to certain parts of the field. It revealed what was obvious to most of us - Kitna, while admirable in certain areas, can only throw the ball to a limited space on the field. Between the numbers or shallow. I believe it shows that Romo is much more difficult to defend as DBs must respect the area he can attack. Very revealing.
9:33 [Comment From Sam Monson PFF: ]
There's a reason Kitna has never been a franchise QB. He can move the ball, and he can do enough to win games, but he's not the weapon a guy like Romo is and can be
9:34 This may change for each position, but what is the maximum a player can receive or lose on 1 play via your system?
9:37 [Comment From Sam Monson PFF: ]
Essentially we grade from +2 to -2 in 0.5 increments, but the 2s are for crazy-good plays that win games essentially and are very hard to get. Certain positions can attain higher for compound plays. Marshawn Lynch's run against the Saints for example got a +2.5, and a defensive player beating his blocker immediately, making a sack, forced fumble, picking it up and running it in for a TD can get a +2.5
9:38 It appears looking at your grades for the defense that the secondary remains the weak link here. In particular Ball, Jenkins, and Newman.
9:40 [Comment From Sam Monson PFF: ]
Absolutely. This is a case where what you see with the naked eye in a live watching does tell the real story. Mike Jenkins had a hugely poor season after looking promising, and the secondary was a major issue
9:41 Anthony Spencer is controversial around these parts. Everyone wants a dominating pass rusher on the edge of the 3-4, and his sack totals just don't make the eyes pop. However, his play vs the run is notable. So what do you think of Spencer overall compared to players across the league at his spot?
9:42 [Comment From Sam Monson PFF: ]
Spencer had a bit of a down year rushing the passer this season, but last season he was excellent in that regard, despite the sack numbers. His play vs the run is very good, and though he may not be an All-Pro, he is a very good player, and a real weapon opposite Ware
9:43 Where are you on the Jay Ratliff debate. Because of his size, I really would like to see him DE because I believe in the NT run stuffer. Your run grades for him seem to back this up.
9:46 [Comment From Sam Monson PFF: ]
The thing about Ratliff is that he isn't a traditional 2-gap NT. He's a 1-gap player who is best shooting gaps and blowing plays up in the backfield. But he's shown in the past ('09 especially) that he can be a force from NT doing just that. This year wasn't a good season for him, especially vs the run, but we wouldn't write him off as a NT based just on 2010
9:46 Last question before I let you go: Top 2 or 3 things on the Cowboys that must be addressed this offseason?
9:48 [Comment From Sam Monson PFF: ]
We've talked about them so far: Secondary help, right tackle, more secondary help. Settling on a running back probably wouldn't hurt either
9:48 Thanks, Sam. I really enjoy your project and recommend it highly. Would love to do this again sometime soon.
9:48 [Comment From Sam Monson PFF: ]
No problem, a pleasure good sir.
