| View previous topic :: View next topic |
| Author |
Message |
bulldog
Joined: 18 Nov 2008
Posts: 4411
Location: Salem
|
 Posted: Mon Jul 02, 2012 9:54 pm Post subject: State of the Franchise: The Defensive Line Edition |
 |
|
|
Here's yet another As The World Turns...I mean "State of the Franchise: Defensive Line Edition"
This group went through the biggest transition last year. 2 rookies fresh out of college, now starting. One DE that has never been a full time starter. Another DE that showed up 25 Lbs too heavy. One DT that played a 34 his whole career. Lots going on with no offseason. New coaches. New scheme. Oh the humanity.
I think some of these figures show why the FO did what they did this offseason. I also think it may open some eyes to how well the LB's did considering how poorly the DL was. It also makes the Secondary look amazing, I think.
There's alot of reading here. Please take the time to read it all. I promise it will be worth your time.
First things first. Quick reminders.
- Stops- The number of tackles (including sacks) that constitute a loss for the offense, unless otherwise notes. I.e- Run Stops- Stops on only run plays.
-PRP- Pass Rush Productivity - A grade based on total disruptions per snap
- Rankings are out of a total of 88 DT's and 67 DE's. So some of the rankings are a little lower than they would be for just starters. I.e.-
Rubin ranked 15th among DT's in Run Stop %, but 7th among DT's with 50% playing time.
- PFF grades are based on a grade of zero being average.
-quick note, all of these guys would have better grades had not been for the Houston game. The one game dropped Rubin's total PFF grade by 3.8 points in the run game. They took a beating in that game.
Ahtyba Rubin- The Anchor
Key Stats - Games 1-8 - Games 9-16 – Total - rank
snaps-503 - 442 -945 - 2nd
Tackles -34 - 25 - 59 - 1st
Assisted -4 - 7 -11 - 4th
Total Stops -24 - 21 -45 - 1st
Sacks -4 - 2 -6 - 4th
Hits -2 - 3 -5 - 14th
Hurries -6 - 11 -17 - 16th
PRP- 4.9 - 27th
Missed Tackles -2 - 2 -4 - 80th
Run Snaps-434-
Run Stops- 37- 15th/per snap
Penalties -1 - 0 -1
PFF grades
Total --0.1 - 9.0 -8.9 - 19th
Run -(-)1.4 - 3.8 - 2.4- 38th
Rush --0.8 - 2.9 -2.1 - 23rd
analysis- Our feature Defensive lineman. He really has a might motor, even with all of those snaps played. Really a great run defender, and a surprising pass rusher. He's no Warren Sapp, but he gets the job done when called on. This is a severely underrated player around the league.
Phil Taylor- Say Uncle
Key Stats - Games 1-8 - Games 9-16 – Total - rank
snaps-347 - 395 -742 - 16th
Tackles -17 - 24 -41 - 2nd
Assisted -3 - 4 -7 - 27th
Total Stops -14 - 13 -27 - 11th
Sacks -3 - 1 -4 - 12th
Hits -1 - 1 -2 - 36th
Hurries -6 - 4 -10 - 35th
PRP - 5.3 - 65th
Missed Tackles -0 - 2 -2 - 42nd
Run Snaps- 374 -
Run Stops -20 - 65th/per snap
Penalties -1 - 3 - 4
PFF grades
Total -0.2 - (-)7.6 -(-)7.4 - 77th
Run -(-)3.7 - 0.8 -(-)2.9 - 76th
Rush -2.7 - (-)5.6 - (-)2.9- 49th
analysis- Phil Taylor definitely fell into the "rookie wall" and never came out in the second half of the season. His play dropped off in almost every category. While his tackles went up, his total QB pressures went down, and penalties and missed tackles went up. On more snaps at that. He really needed this TC to refine his skills. This year may be a mulligan for him.
Brian Shaefering
Key Stats - Games 1-8 - Games 9-16 – Total - rank
snaps- 93- 142 -235
Tackles -10 - 16 -26
Assisted -0 - 1 - 1
Total Stops -6 - 8 -14
Sacks -0 - 0 - 0
Hits -1 - 1 - 2
Hurries -2 - 3 - 5
PRP - -
Missed Tackles- 0
Run Snaps- 81
Run Stops - 9
Penalties - 0
PFF grades
Total -1.3 - 1.5 -2.8
Run -0.1 - 1.9 - 2.0
Rush -1.0 - (-)0.9 - 0.1
analysis- This one surprised me. I think Sheafering is one of the worst players on the roster, but some of these numbers suggest that may not be so. His amount of stops in the run game is impressive considering how few snap he had. It also appears that his stats improved the second half of the season, but at closer look, not so much. His production-to-snap ratio in the first 8 games was 4.89. In the second half it was 4.9 He really doesn't offer much as a pass rusher though.
Scott Paxson
Key Stats - Games 1-8 - Games 9-16 – Total - rank
snaps-102 - 171 -273
Tackles -2 - 12 -14
Assisted -0 - 3 -3
Total Stops - 2- 10 - 12
Sacks -0 - 1 - 1
Hits -0 - 0 - 0
Hurries -0 - 2 - 2
PRP - -
Missed Tackles -0 - 0 - 0
Run Snaps- 131
Run Stops - 9- 38th/per snap
Penalties -0 - 0 - 0
PFF grades
Total -(-)7.3 - 5.2 -(-)2.1
Run -(-)4.9 - 5.1 - 0.2
Rush -(-)2.5 - 1.2 - (-)3.7
analysis- If you look at his first half stats, they are VERY underwhelming. However, his second half stats are a very impressive improvement. Paxson's production-to-snap ratio the first half of the season was 25.5. So one play every 25.5 snaps. However, in the second half jumped to 6.1. While not as good as Sheaferings, Paxson was with out a doubt the most improved player on the DL.
Defensive Ends
Jabaal Sheard
Key Stats - Games 1-8 - Games 9-16 – Total - rank
snaps-479 - 491 -970 - 4th
Tackles -23 - 14 -36 - 5th
Assisted -5 - 5 -10 - 5th
Total Stops -19 - 17 -36 - 4th
Sacks -3 - 6 -9 - 16th
Hits -2 - 2 -4 - 41st
Hurries -19 - 23 -42 - 5th
PRP – 9.2 - 21st
Missed Tackles -4 - 5 -9 - 67th
Run Snaps- 438 -
Run Stops -25 - 28th/per snap
Penalties -2 - 3 - 5
PFF grades
Total -8.8 - 2.5 -11.3 - 22nd
Run -0.4 - (-)2.2 -(-)1.8 - 50th
Rush -8.5 - 6.8 - 15.3 - 10th
analysis- He's definitely one hell of a pass rusher. To play that many snaps and still show production in the second half of the season is encouraging I think. He will most surely improve those sack numbers with more pressure from the other side, since he was almost to the Qb 42 times. He really needs to step up his game in the run defense. He's the weak link on the DL in that area now.
Jayme Mitchell- Out with the old…..
Key Stats -– Total - rank
snaps--738 - 20th
Tackles --26 - 20th
Assisted --7 - 22nd
Total Stops --20 - 33rd
Sacks --3 - 45th
Hits --7 - 22nd
Hurries --5 - 63rd
PRP – 3.2 - 68th
Missed Tackles --4 - 51st
Run Snaps- 333
Run Stops - 18- 31st/per snap
Penalties --
PFF grades
Total –(-)15.3 - 66th, second to last
Run –(-)0.4 - 44th
Rush –(-)18.8 - 67th, Last place. Next worst was (-)9.8 lol
analysis- Oh, there is sooo much suck here. He was literally below average in run defense, and The.Worst.Pass.Rusher there was last season. Any other player in the league would have been an upgrade here.
Frostee Rucker- ….In with the new
Key Stats -– Total - rank
snaps--480 - 47th
Tackles --37 - 5th
Assisted - 4 - 32nd
Total Stops -- 31 - 12th
Sacks -- 4 - 34th
Hits - 2 - 56th
Hurries - 5 - 63rd
PRP – 4.6 - 61st
Missed Tackles -- 1 - 10th
Run Snaps- 225 -
Run Stops -26 - 1st/per snap
Penalties - 0
PFF grades
Total – 3.4 - 36th
Run – 9.5 - 13th
Rush – (-)7.2 - 55th
analysis- His run stopping abilities are remarkable. This alone is a huge improvement to the DL. His pass rushing skills are very suspect, but even thats a huge upgrade over what was there. However, considering that Heckert said that Rucker will move inside in the Nickle packages, His PRP rating of 4.6 would rank 30th among ALL DT's. So there is some hope there. Being inside may fit his skill set better anyways.
Juqua Parker
Key Stats -– Total - rank
snaps-- 262 - 67th
Tackles - 7 - 62nd
Assisted -- 1 - 60th
Total Stops -- 9 - 62nd
Sacks -- 2 - 54th
Hits -- 4 - 41st
Hurries -- 14 - 45th
PRP – 11.2 - 8th
Missed Tackles - 0 - 1st
Run Snaps- 103
Run Stops - 5- 37th/per snap
Penalties -- 2
PFF grades
Total – 7.3 - 28th
Run – 0.6 - 37th
Rush – 8.1 - 23rd
analysis- Parker doesn't add much in terms of a run stopper. He does bring some pass rush though. Really, he's quite good at getting to the passer on a per snap basis. Getting some rest for our DL was clearly a priority. Since we didn't really have a backup LDE last year, this is a big upgrade.
Marcus Benard- Unsung Hero
analysis Benard played on 11 passing snaps, rushing every time. He generated 1 sack and 1 hurry. So 2 pressures in 11 rushes is a good number. He played a total of 26 snaps, pulling in two more stops.
In 2010, Benard was ranked the 8th best pass rusher for 34 LB's. He had a total of 32 total pressures on 225 rushing snaps. On top of that, his grade from both the left and right side were almost identical.
------------------------------------------------------------------------------
The Conclusion- Rubin was really the only player that played at a decent level last year. Taylor and Sheard played like rookies. Thought, they did show some promise. Our backup DT's show some promise, but the FO must think they didn't show enough since they drafted two DT's. Getting Benard back will be a big deal. Also. Rucker and Parker will be much appreciated players.
I think I've shown how not so good this unit was last year. Then looking back at the other "state of the franchise' threads, I think it shows how much more impressive some of the other positions were. The LB's, by most people's standards, were very average last year, minus DQ. However, take into account how poorly the line was in front of them, maybe they were a little better than average...
This unit has no way to go but up. The addition of Rucker should dramatically improve the run D and improve flexibility. Parker brings another element to rushing the passer. Combine his arrival with the return of Benard and we should be at least 3 deep at premier pass rushers. Rubin is steady as can be. The other DT spot needs sorted out. I feel that between the improvement of Paxson and the rookies, we should be ok there.
Up next, the Offensive Line Edition.
-Other "State of the Franchise" threads
Linebackers
http://www.footballsfuture.com/phpBB2/viewtopic.php?t=493656&start=0
Secondary
http://www.footballsfuture.com/phpBB2/viewtopic.php?t=492652
Receivers
http://www.footballsfuture.com/phpBB2/viewtopic.php?t=493522 |
|
| Back to top |
|
 |
bruceb
Joined: 15 Dec 2006
Posts: 5339
Location: Rocky River, OH
|
| Posted: Mon Jul 02, 2012 10:49 pm Post subject: |
|
|
|
This and the entire series of threads you have posted are nothing short of remarkable.
Tremendous, data-based and otherwise informative contributions to the forum, imho. Thanks!
This one does put a spotlight on where improvement was needed most on D, and explains FA and the draft.
_________________
Everything happens exactly the way it is supposed to happen...otherwise, it would happen some other way |
|
| Back to top |
|
|
bulldog
Joined: 18 Nov 2008
Posts: 4411
Location: Salem
|
| Posted: Tue Jul 03, 2012 5:40 am Post subject: |
|
|
|
| bruceb wrote: | This and the entire series of threads you have posted are nothing short of remarkable.
Tremendous, data-based and otherwise informative contributions to the forum, imho. Thanks!
This one does put a spotlight on where improvement was needed most on D, and explains FA and the draft. |
Thank you. I really appreciate it. |
|
| Back to top |
|
|
mistakebytehlak
Joined: 20 Nov 2005
Posts: 38876
Location: Brooklyn, NY
|
| Posted: Tue Jul 03, 2012 9:07 am Post subject: |
|
|
|
| Quote: | 2) What Do We Grade?
Throughout the course of the season (regular season and playoffs) we grade every single offensive, defensive and special teams snap. We log data such as the point of attack of a running play, the location a pass was thrown and hang time of kicks and punts before moving on to the player-performance analysis.
.
A typical line of analysis will describe an offensive and defensive player being graded for a one-on-one confrontation. This will include their names and grades as well as a comment describing the play. So for example, a match-up between a right guard and left defensive tackle could result in the following comment:
.
“The RG drove the DLT down the line of scrimmage opening a wide hole off his outside hip for the running back (##) to pick up the first down on 3rd & 3.”
.
This type of notation serves a few purposes. First, it captures detail for grading, a concise comment that can be referenced back to individual players for further analysis at a later date. Also, due to each play having a unique ID, it also creates a clear and accessible audit trail for all analysis. |
| Quote: | • YOU DO NOT HAVE TO APPORTION BLAME ON EVERY PLAY — On each play there is often a “winner.” One unit, be it the offense or the defense, will usually get the better of a play by varying degrees. This, however, does not entail that one or more individuals on the losing unit are to blame.
For example, if an offense is stopped on 3rd-and-3 on a running play for 2 yards, that would constitute a failure for the offensive unit. But each member of the offense may very well have carried out his assignment properly.
Say the defense sets up overloaded against the run. Every defender except one is successfully blocked. A lone, unblocked defender makes a strong tackle to stop the back short of the marker.
In that instance, no one individual is at fault for the play failing. The defense simply had the right play called. Sometimes plays are designed badly, sometimes coaches don’t adjust. This site is looking at individual player performance, not that of coaches and not necessarily how individual player performance correlates with team performance. |
without knowing the play called and without having the all 22 (not available)--- the way they grade is incredibly wrong--- they assign blame without knowing whats going on. additionally, if a free pass rusher comes, how do we not know it was the center's fault? how do we not know that the dude went all Lavar Arrington and freelanced? Additionally, since htey grade individual vs individual, how do they grade zone coverage or the zone blocking scheme?
additionally, quite subjective to decide what gets a .5, 1. 1.5 and a 2.0. these valuations should somehow be given by nature.
we're not there yet.
_________________
 |
|
| Back to top |
|
|
ReggieCamp
Joined: 06 Dec 2006
Posts: 8713
Location: Canonsburg, PA
|
| Posted: Tue Jul 03, 2012 11:42 am Post subject: |
|
|
|
^^^
They do provide some nice metrics that are not easy to find elsewhere, but yes, some of their stats are subjective. They have to be, really. Even if they firm up their criteria, there will always be a subjective part of it.
It will be interesting to see how they change things once they have access to the all-22 film.
_________________
Cleveland Browns: We put the fun in dysfunction.
 |
|
| Back to top |
|
|
bulldog
Joined: 18 Nov 2008
Posts: 4411
Location: Salem
|
| Posted: Tue Jul 03, 2012 12:17 pm Post subject: |
|
|
|
| mistakebytehlak wrote: | without knowing the play called and without having the all 22 (not available)--- the way they grade is incredibly wrong--- they assign blame without knowing whats going on. additionally, if a free pass rusher comes, how do we not know it was the center's fault? how do we not know that the dude went all Lavar Arrington and freelanced? Additionally, since htey grade individual vs individual, how do they grade zone coverage or the zone blocking scheme?
additionally, quite subjective to decide what gets a .5, 1. 1.5 and a 2.0. these valuations should somehow be given by nature.
we're not there yet. |
You don't like PFF, I know.
The stats are debatedable. Show me a stat outside of catches, TD's and INT's that isnt.....
The telling story in those figures are the production per snap. Thats what I hold highly. Not so much their individual grades. There's no messing up on a per snap basis. |
|
| Back to top |
|
|
mistakebytehlak
Joined: 20 Nov 2005
Posts: 38876
Location: Brooklyn, NY
|
| Posted: Tue Jul 03, 2012 3:39 pm Post subject: |
|
|
|
| bulldog wrote: | | mistakebytehlak wrote: | without knowing the play called and without having the all 22 (not available)--- the way they grade is incredibly wrong--- they assign blame without knowing whats going on. additionally, if a free pass rusher comes, how do we not know it was the center's fault? how do we not know that the dude went all Lavar Arrington and freelanced? Additionally, since htey grade individual vs individual, how do they grade zone coverage or the zone blocking scheme?
additionally, quite subjective to decide what gets a .5, 1. 1.5 and a 2.0. these valuations should somehow be given by nature.
we're not there yet. |
You don't like PFF, I know.
The stats are debatedable. Show me a stat outside of catches, TD's and INT's that isnt.....
The telling story in those figures are the production per snap. Thats what I hold highly. Not so much their individual grades. There's no messing up on a per snap basis. |
a true statistic- one that is a function of the data isn't debatable. the only debate is what it truly means.
if the individual grades are suspect, the per snap grades are suspect.
the dudes at footballoutsiders use valid statistical measures of regression. brian burke at advancednflstats is a bona fide mathematical genius. any statistic he puts out, i trust.
we dont know the credentials of the guys at PFF and their ranking system is suspect at best.
this could be why schaefering looks like a good player...
_________________
|
|
| Back to top |
|
|
mistakebytehlak
Joined: 20 Nov 2005
Posts: 38876
Location: Brooklyn, NY
|
| Posted: Tue Jul 03, 2012 3:40 pm Post subject: |
|
|
|
| ReggieCamp wrote: | ^^^
They do provide some nice metrics that are not easy to find elsewhere, but yes, some of their stats are subjective. They have to be, really. Even if they firm up their criteria, there will always be a subjective part of it.
It will be interesting to see how they change things once they have access to the all-22 film. |
thats fine if we need to have some subjectivity--- ONLY IF--- they knew what they were talking about, but they clearly state that they are not NFL scouts.
_________________
|
|
| Back to top |
|
|
bulldog
Joined: 18 Nov 2008
Posts: 4411
Location: Salem
|
| Posted: Tue Jul 03, 2012 6:21 pm Post subject: |
|
|
|
| mistakebytehlak wrote: | | bulldog wrote: | | mistakebytehlak wrote: | without knowing the play called and without having the all 22 (not available)--- the way they grade is incredibly wrong--- they assign blame without knowing whats going on. additionally, if a free pass rusher comes, how do we not know it was the center's fault? how do we not know that the dude went all Lavar Arrington and freelanced? Additionally, since htey grade individual vs individual, how do they grade zone coverage or the zone blocking scheme?
additionally, quite subjective to decide what gets a .5, 1. 1.5 and a 2.0. these valuations should somehow be given by nature.
we're not there yet. |
You don't like PFF, I know.
The stats are debatedable. Show me a stat outside of catches, TD's and INT's that isnt.....
The telling story in those figures are the production per snap. Thats what I hold highly. Not so much their individual grades. There's no messing up on a per snap basis. |
a true statistic- one that is a function of the data isn't debatable. the only debate is what it truly means.
if the individual grades are suspect, the per snap grades are suspect.
the dudes at footballoutsiders use valid statistical measures of regression. brian burke at advancednflstats is a bona fide mathematical genius. any statistic he puts out, i trust.
we dont know the credentials of the guys at PFF and their ranking system is suspect at best.
this could be why schaefering looks like a good player... |
All three of these sites have debatable data. Period. Truth is, I could poke holes in all three of their grading systems.
You've posted your stuff from advancedNFLstats.com on here before. I don't buy into their Win Probability crap. But I didn't give you grief b/c I think its great to be able to look at others opinions to get different perspectives.
Football Outsiders shows a general stat for each play based on field position and situation, but not player position. (Nickle CB, slot WR, pass rush specialist, run stop specialist) They don't even have a section for LB's or the secondary. At least not that I'm aware of.
I've devoted hours to each thread. If you don't like it, Fine. I get it. You don't care for PFF. Some other's enjoy reading it. As they have with your stuff you've posted in the past.
I'm not defending the stats, grades, whatever. They are there to be discussed and to bring different light to a situation. If you have nothing constructive to say about anything, then just let it rest. |
|
| Back to top |
|
|
mistakebytehlak
Joined: 20 Nov 2005
Posts: 38876
Location: Brooklyn, NY
|
| Posted: Tue Jul 03, 2012 7:31 pm Post subject: |
|
|
|
Since when did discussion and absolute agreement become synonyms? Discussion includes criticism.
Why don't you buy into the WPA? Statisticians have analyzed down and distance and found the average win probability added from those plays from many, many, many previous games and situations. the expectations are added. this is commonplace.
FO don't have those on their website. I know they dont differentiate between LG, C, and RG because they are statistically insignificant--- i.e. they actually use rigorous statistics to ensure what they put out is fine.
Some articles they have nickel CB etc--- it may very well be in their almanac.
If you don't like that people don't agree with PFF- debate the value of their statistics.
_________________
|
|
| Back to top |
|
|
bulldog
Joined: 18 Nov 2008
Posts: 4411
Location: Salem
|
| Posted: Tue Jul 03, 2012 8:34 pm Post subject: |
|
|
|
| mistakebytehlak wrote: | | Why don't you buy into the WPA? Statisticians have analyzed down and distance and found the average win probability added from those plays from many, many, many previous games and situations. the expectations are added. this is commonplace. |
I don't like it b/c its just like vegas odds. A hand at blackjack, if you will. There's WAY to many variables in football to rely on an average. A percentage of winning. PFF focus's almost exclusively on individual talent. Which is where more of my interest is. So I may be more biased b/c of that. Its more about the football player, less about the game or the team.
| Quote: | | FO don't have those on their website. I know they dont differentiate between LG, C, and RG because they are statistically insignificant--- i.e. they actually use rigorous statistics to ensure what they put out is fine. |
What statistics? You said yourself a stat is something that isn't debatable. So what stat? They base a "win" off a book where 45% on first down is a win. 60% on second down is a win. and a first down on 3rd or 4th down is a win.
I was wrong when i said that a "stop" is a loss for the offense. Its actually this. | Quote: | What constitutes a “stop” or “offensive failure?”
It’s when an offense fails to get:
40 percent of the required yards for another first down on first down
60 percent of the required yards for a first down on second down
A first down on third or fourth down |
Look familiar??
This could explain why Sheafering's numbers look decent. A defense can "win" on first and second down, but still "lose" on third down. A player's numbers may not reflect that.
| Quote: | | Some articles they have nickel CB etc--- it may very well be in their almanac. |
PFF has whole sections for these things. Farther explaining a players strength and weaknesses.
| Quote: | | If you don't like that people don't agree with PFF- debate the value of their statistics. |
Their grading system is debatable, as is everyone else's. Thats one of the reason I don't make remarks on PFF's grades. I do put a lot of stock into their signature stats. For example, they will show a QB's yards IN AIR, with and without drops. A Qb's accuracy minus drops, spikes, and throw aways. How many times a QB was blitzed, how accurate said QB was when blitzed, with and without drops.
They show what a QB's passer rating was throwing to certain WR. How about a WR's or TE's drop rate....drops compared to catches. They show which WR's did the best as deep threat (+20 yards) and in the slot
For the OL, they base their pass block efficiency on how many pressure said player allows on X amount of snaps. The same for individual OL.
Jared Allen led the league in sacks last year. Does that mean he's the best pass rusher? or did he just have the most opportunity? Allen rushed the passer 567 times. Racking up 24 sacks, 8 hits, and 34 hurries. Trent Cole on the other hand only had 11 sacks, 12 hits, and 44 hurries. Cole did that on only 355 pass rushing attempts. Thus their Pass Rush Productivity rating.
There's more, but I'm not going to put it all out there.
PFF takes into account one thing different. Opportunity. The other 2 sites don't. At least not that they show.
AdvancedNFLstats shows that JPP had 82 tackles and 27 TFL.... How many of those came on screen passes or his great hustle getting down field? PFF separates those plays out.
Does Football outsiders discount QB scrambles for their OL grades? PFF does.
Like I've said, everything outside of catches, INT's and TD's is debatable. All three of these sites have a different grading system. But doesn't it all start at or around the same place? they just all have different way of showing it. |
|
| Back to top |
|
|
bungleodeon
Joined: 19 Mar 2009
Posts: 7888
|
| Posted: Tue Jul 03, 2012 8:54 pm Post subject: |
|
|
|
| mistakebytehlak wrote: | Discussion includes criticism.
|
Milk shot out of my nose laughing at the irony.
_________________
Adopt-a-Brown: Jabaal Sheard - Stats:
2011: 55 Tackles - 8.5 Sacks - 5 FF - 1 FR - 2 PD - 7 STF
2012: 54 Tackles - 7 Sacks - 1 FF - 3 PD - 2 STF |
|
| Back to top |
|
|
mistakebytehlak
Joined: 20 Nov 2005
Posts: 38876
Location: Brooklyn, NY
|
| Posted: Wed Jul 04, 2012 1:06 am Post subject: |
|
|
|
| bungleodeon wrote: | | mistakebytehlak wrote: | Discussion includes criticism.
|
Milk shot out of my nose laughing at the irony. |
_________________
|
|
| Back to top |
|
|
mistakebytehlak
Joined: 20 Nov 2005
Posts: 38876
Location: Brooklyn, NY
|
| Posted: Wed Jul 04, 2012 1:15 am Post subject: |
|
|
|
| bulldog wrote: | | mistakebytehlak wrote: | | Why don't you buy into the WPA? Statisticians have analyzed down and distance and found the average win probability added from those plays from many, many, many previous games and situations. the expectations are added. this is commonplace. |
I don't like it b/c its just like vegas odds. A hand at blackjack, if you will. There's WAY to many variables in football to rely on an average. A percentage of winning. PFF focus's almost exclusively on individual talent. Which is where more of my interest is. So I may be more biased b/c of that. Its more about the football player, less about the game or the team.
| Quote: | | FO don't have those on their website. I know they dont differentiate between LG, C, and RG because they are statistically insignificant--- i.e. they actually use rigorous statistics to ensure what they put out is fine. |
What statistics? You said yourself a stat is something that isn't debatable. So what stat? They base a "win" off a book where 45% on first down is a win. 60% on second down is a win. and a first down on 3rd or 4th down is a win.
I was wrong when i said that a "stop" is a loss for the offense. Its actually this. | Quote: | What constitutes a “stop” or “offensive failure?”
It’s when an offense fails to get:
40 percent of the required yards for another first down on first down
60 percent of the required yards for a first down on second down
A first down on third or fourth down |
Look familiar??
This could explain why Sheafering's numbers look decent. A defense can "win" on first and second down, but still "lose" on third down. A player's numbers may not reflect that.
| Quote: | | Some articles they have nickel CB etc--- it may very well be in their almanac. |
PFF has whole sections for these things. Farther explaining a players strength and weaknesses.
| Quote: | | If you don't like that people don't agree with PFF- debate the value of their statistics. |
Their grading system is debatable, as is everyone else's. Thats one of the reason I don't make remarks on PFF's grades. I do put a lot of stock into their signature stats. For example, they will show a QB's yards IN AIR, with and without drops. A Qb's accuracy minus drops, spikes, and throw aways. How many times a QB was blitzed, how accurate said QB was when blitzed, with and without drops.
They show what a QB's passer rating was throwing to certain WR. How about a WR's or TE's drop rate....drops compared to catches. They show which WR's did the best as deep threat (+20 yards) and in the slot
For the OL, they base their pass block efficiency on how many pressure said player allows on X amount of snaps. The same for individual OL.
Jared Allen led the league in sacks last year. Does that mean he's the best pass rusher? or did he just have the most opportunity? Allen rushed the passer 567 times. Racking up 24 sacks, 8 hits, and 34 hurries. Trent Cole on the other hand only had 11 sacks, 12 hits, and 44 hurries. Cole did that on only 355 pass rushing attempts. Thus their Pass Rush Productivity rating.
There's more, but I'm not going to put it all out there.
PFF takes into account one thing different. Opportunity. The other 2 sites don't. At least not that they show.
AdvancedNFLstats shows that JPP had 82 tackles and 27 TFL.... How many of those came on screen passes or his great hustle getting down field? PFF separates those plays out.
Does Football outsiders discount QB scrambles for their OL grades? PFF does.
Like I've said, everything outside of catches, INT's and TD's is debatable. All three of these sites have a different grading system. But doesn't it all start at or around the same place? they just all have different way of showing it. |
Expectations are exactly like vegas odds? what does this even mean? The sample average is very underrated. Its an MVUE estimator--- i.e. its the best estimator, as long as the average makes sense. This average makes absolutely PERFECT sense.
| Quote: | Teams are ranked according to Adjusted Line Yards. Based on regression analysis, the Adjusted Line Yards formula takes all running back carries and assigns responsibility to the offensive line based on the following percentages:
Losses: 120% value
0-4 Yards: 100% value
5-10 Yards: 50% value
11+ Yards: 0% value |
http://footballoutsiders.com/stats/ol
as long as their estimates are met- their statistics were regression coefficients which are BLUE and MVUE. so, again- extremely valid.
WPA takes into consideration winning on first down and losing on third down- since burke used so many trials in determining the win probabilities for each down and distance, the variance on the WPA should be practically nothing--- therefore his estimates should be extremely accurate, and should weigh 1st and 3rd fantastic. much better than the subjective t-score given by PFF (not even z-score. no chance that statistical abomination knows the population standard deviation) that SHOULD be unskewed since they made it--- but it isnt.
I don't care if PFF rates nickel CBs. They don't have the all 22. pretty much individual stuff for CBs and safeties are worthless, unless you can tell if they are in all man, and know that they should have been in all man from the get go. or if you can somehow definitively tell who was at fault in the cover 2. in the cover 2 when the CB releases, there will CERTAINLY be subjectivity.
You're right- they all have different ways of showing it.
two of them use regression and rigorous statistics.
one uses subjective t-scores.
_________________
|
|
| Back to top |
|
|
Dawgpoun8017
Joined: 14 Jan 2009
Posts: 9279
Location: Waterloo,NY
|
| Posted: Wed Jul 04, 2012 1:57 am Post subject: |
|
|
|
Too many math terms being thrown out.........
_________________
2013 Joe Blackburn HOF Award Recipient
"Every man dies, not every man really lives."
 |
|
| Back to top |
|
|
|
|
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
|
|
|
Powered by phpBB © 2001, 2005 phpBB Group
|
FAQ
Register
Profile
Log in to check your private messages
Log in
