Pos-t Mortem

Photo: Ed Yourdon. License

(Note: Check out this piece at our new location with a new look and feel… Blog-a-Bills….)(

We all had high hopes for Paul Posluszny when he was drafted by the Bills. He seemed the prototypical linebacker and a Buffalo guy; no-neck, ostensibly tough-as-nails, and he came from the same school as Shane Conlan (which seemed to matter to some at the time). But as his career progressed with the Bills, and injuries helped slow him down, the public perception about Pos flipped to the point where one would think he had run over some pedestrians while drunk on Chippewa.

Since the lockout ended, I’ve seen and/or heard people describe his problem as being unable to make tackles closer than 5, 7 or 10 yards to the line of scrimmage. Obviously, one must take those complaints with a grain of salt, and many were made with the intent of getting a laugh. But, as Bills fans, we’ve all heard this complaint before. Specifically, we heard it about London Fletcher, who is still a productive MLB years after the Bills let him go.

So, is this a case of Bills fans simply finding more to dislike about the man because of his (statistically proven) inability to produce the “big” play? Or is he actually making tackles that much farther down the field than his peers? And, even if he is, it’s better than watching Chris Kelsay do anything, isn’t it?

Fair warning: Much of this blog will be dedicated to looking at stats, and making conclusions based on them. This requires the use of more than what’s available on NFL.com or ESPN.com. Before we delve in to Pos’ performance, we will take a quick look at some of the metrics being used here.

While there aren’t many advanced metrics in the world of the NFL (at least not to the extent of the MLB or even the NBA), there are websites that have made leaps and bounds in being able to use advanced stats to judge some of these types of assertions. Specifically, we’ll be looking at the Football Outsiders Almanac 2010 from the fine folks at footballoutsiders.com, along with pro-football-reference.com, both excellent sources of advanced and detailed NFL stats.

The stats needed to make a determination on where a defender makes his tackles is easily found on footballoutsiders.com. Their advanced metrics include stats such as Stops (the number of plays which prevent success by the offense, with success being defined as 45% of needed yards on 1st down, 60% on 2nd, and 100% on 3rd or 4th); Defeats, which are the number of plays which stop the offense from getting a first down on either 3rd or 4th, and most interestingly to us; Average Yards, which is the average number of yards gained by the offense when the player is credited with making the play).

A stat we will use from pro-football-reference.com is their Approximate Value, or AV (CarAV over a carer). This number attempts to assign the value of a player based on traditional statistics, while also accounting for the quality of opponents and teammates, and a number of other mitigating factors.

Where Does Pos Make His Tackles?

The stats available on FOA go back 3 years. Below find Pos’ advanced general defensive stats over the past 3 years:

Year	Plays	Stops	Defeats	Yd/Play	Rank	Stop Rate	Rank
2010	155	73	21	5.1	71	47%	92
2009	115	56	18	5.9	97	49%	97
2008	116	63	14	5	59	54%	61

The stats above would seem to indicate the general consensus about Pos is accurate: He makes tackles far from the line of scrimmage. The average yards per play Pos was involved in making a tackle in was at 5.1, good enough for only 71st in the league (amongst front 7 players). This is actually an increase over his awful 2009 season, in which he essentially gave up 6 yards every time he made a tackle.

Stats available on the Bills team rushing defense may give us more insight into how that might have contributed to Pos’ performance (though he is undeniably a part of that performance). FOA has developed a stat, Adjusted Line Yards (ALY), which helps quantify the performance of both Offensive, and Defensive lines. These stats shed some insight into what Pos had to deal with upfront last year.

Buffalo was BY FAR the lowest ranked team in terms of ALY, implying that Buffalo’s defensive line was absolutely atrocious at stuffing the run. One would have to assume that this lack of any sort of penetration, stops or more by the DL played a role in where Pos was able to make his tackles. (We do not forget Kyle Williams and his contributions on the line, only that he was the only one making them).

Regardless of that caveat, there seems to be some quantifiable evidence suggesting Pos may actually make more tackles further down the field than his contemporaries. However, these numbers include numbers against the pass. I think we can agree that scheme plays a lot into what a player has to do in pass defense in the 3-4, so it might be unfair to judge RILBs against each other while including the passing stats. So, let’s take a deeper look at Pos, RILBs, and the run.

Pos and NFL RILBs vs the Run

While perusing the FOA database, it is not unusual to see that the vast majority of LBs see their AvYd stat vs the pass be significantly higher than it is against the run. This, of course, is not unexpected, as one would expect a completed pass play to go for more yards, on average, than a run (unless of course you are the 2008, 2009, 2010, and probably 2011, Buffalo Bills). With that in mind, here are Pos’ stats the last 3 yrs versus the run.

Year	Plays	Stops	Dfts	Yd/Play	Rank	Stop Rate	Rank
2010	107	53	10	4	80	50%	95
2009	76	40	8	4.4	94	53%	97
2008	62	43	6	2.9	22	69%	25

Clearly, there is much left to be wanting here. His stop rate has declined steadily since 2008, indicating he is making fewer plays when it matters . His yd/play vs the run has improved from last season, but is still not what one would expect from an average-to-middling RILB, let alone a top-flight RILB being paid so much by Jacksonville he could probably buy a small-to-medium-sized island in the South Pacific now.

We can all agree that these numbers in a vacuum mean relatively little. In order to make a more apples-to-apples comparison, here is a look at the 2010 Yd/Play stats for each starting RILB in the NFL:

Player	Yds/Run
Lawrence Timmons	2.5
Mario Haggan	3
Kevin Burnett	3
Chris Gocong	3
Brandon Spikes	3.1
Derrick Johnson	3.6
David Harris	3.7
Jameel McClain	3.7
Keith Brooking	3.7
Rocky McIntosh	3.7
Ray Lewis	3.8
Karlos Dansby	3.8
Paul Posluszny	4
Desmond Bishop	4.2
Patrick Willis	4.7
Daryl Washington	5.3

This is where things get particularly damning for those (such as me) who were attempting to make the argument that Pos is nowhere near as bad as many of the “haters” indicated he was on his way out of Buffalo.

Again, just for the sake of an apples-to-apples comparison, these are starting RILBs. Out of 16 players who qualify under that criteria, Pos ranked 13th. We all know that 4 yds/rush is what a team hopes to get when it runs the ball. These numbers imply that Pos is no better at making them miss that mark than just 3 other starters at his position in the league.

It’s tempting to make the assertion that Pos may have improved in his 2nd year in a 3-4 at ILB, especially since he himself indicated the 4-3 in Jacksonville was one of the reasons he left, but there are a number of players ahead of him in the above table that also made that switch for the last year, and did so much more adequately.

So, Was he as Bad as we Thought he Was?

Speaking as someone who spent a fair amount of time defending Pos as an adequate inside linebacker, statistically it would appear that he actually does have trouble making tackles close to the LOS. Again, I would caution that there are many other factors that play into that number, but none that would seem to consistently explain his poor performance in making plays near the line.

That said, I wouldn’t say it’s completely fair to pillory Pos. Was he worth $7MM a year? The numbers and observations would seem to indicate he is not, and at no point would I have ever made the argument that he was worth that money. But the man does NOT miss tackles, and was always surrounded with subpar talent, so I think some of the criticism he received was unfair.

But value is the name of the game in the NFL these days, and it’s not there at $7MM. It might not even have been there for the $4MM the Bills ended up paying Barnett.

Speaking of Barnett….

At this point, if you’ve made it this far, you’re probably wondering how much better or worse Nick Barnett is when compared to Pos (though I don’t think one was necessarily signed to replace the other).

Interestingly enough, FOA performs a statistical analysis to identify players who had similar levels of productivity/impact on their team within position groups. For the 3-year period we have been looking at, the player who is the closest analogue to Pos (according to FOA and based on similarity in stats) is none other than…..Nick Barnett.

That little bit of trivia aside, here is a look at some of Barnett’s numbers against the run the last 3 years….

Year	Plays	Stops	Dfts	Yd/Play	Rank	Stop Rate	Rank
2010	12	8	2	3.1	n/a	67%	n/a
2009	64	44	13	3.3	36	69%	36
2008	36	23	4	3.8	50	64%	50

Obviously, Barnett’s performance last year was limited due to injury (which is not an inconsequential fact). However, no matter how you slice it, Barnett’s numbers against the run have been better than Pos’, and not by an insignificant number. Not once in the last 3 years has Barnett been over 4.0 yards/run play, while Pos’ last two years were either at 4, or well above.

Bye Bye Pos

I have to admit: When I set out to write this post, I honestly felt that too much was being made of where Pos made his tackles, and that we were falling victim to “London Fletcher Syndrome” again. However, in looking at the big picture with the metrics we have available, it seems that there is little-to-no case to be made that Pos is an above-average ILB, let alone worth $7MM/year.

Enjoy, Jacksonville. After picking up Trent from the Bills last year and Pos this year, I don’t think we’ll see many more Buffalo retreads down there anytime soon.