Writers of Pro Football Prospectus 2008

04 Apr 2008

Linemen Not Worth the Money They're Getting

What's the difference between elite production and average production along the offensive line? According to former lineman Ross Tucker, about fifteen plays over the course of a season. Are those plays worth the $21 million in guaranteed money the Jets paid Alan Faneca? Needless to say, Tucker doesn't think so. It's an interesting read.

Posted by: Sean McCormick on 04 Apr 2008

42 comments, Last at 10 Apr 2008, 2:44pm by crack

Comments

1
by Mr Shush (not verified) :: Fri, 04/04/2008 - 4:21pm

Ross Tucker is awesome . . . but also, I suspect, wrong. Does anyone believe Hutchinson hasn't made a huge difference for the Vikings?

My best guess is that the problem lies in the categorisation of individual plays. It's not that Hutchinson might make one more block in the running game per game. It's that every block he makes has a bigger impact - moves the defender back further, is more likely to knock him down, takes place a split-second earlier or whatever.

And the inclusion of kickers in the list of high impact positions is just daft, obviously.

2
by Kyle, Louisville (not verified) :: Fri, 04/04/2008 - 4:32pm

Agree with #1 and would like to add:

There might be a major difference when the OL is playing against elite defensive talent. Though Faneca and DeMulling would be roughly the same against the Bengal's defensive front, their performance could be worlds apart when put against the Patriot's DL.

3
by Erik (not verified) :: Fri, 04/04/2008 - 4:48pm

Football is often put up as a game of matchups. In one-on-one matchups, a player could dominate, give problems, tie, struggle, or be dominated.

If a matchup on the line ends up at the "dominate" or "be dominated" place (a strength of one team against the weakness of another), horrible things can happen. On the offensive line, this is tackles for loss, incomplete losses, sacks, interceptions and fumbles. At the opposite end, it can result in long runs and long passes as either running lanes are opened or coverages are blown.

An average starter will usually struggle against a star, but not be dominated. A below-average backup may well get dominated. Paying to always be in the tie-or-better position makes sense. Look what happened to the Raiders OL for passing -- they often are struggling or getting dominated at individual positions.

Also, there is synergy on a line. Art Shell and Gene Upshaw separately would not create the same concerns as those two together. Players adjacent to great players can play their own game with greater confidence, leading to the whole being more than the sum of the parts.

That said, I can't see a $35M a year offensive line every happening.

4
by MarkV (not verified) :: Fri, 04/04/2008 - 4:53pm

The most interesting thing he says is one he doesnt really support, which is that run blocking is basically interchangable. That seems wrong, but it is interesting to hear.

Also, it seems as though what he is saying would apply only toward guards, since centers have that intangible thing (snapping and line calls) going and he specifically excluded tackles. And honestly, it wouldnt suprise me much that there isnt much difference from Fanaca and Hutchinson to a random scrub.

5
by Mr Shush (not verified) :: Fri, 04/04/2008 - 5:24pm

"And honestly, it wouldnt suprise me much that there isnt much difference from Fanaca and Hutchinson to a random scrub."

I think a lot of Seahawk and Viking fans would beg to differ . . .

6
by Alex (not verified) :: Fri, 04/04/2008 - 5:26pm

I like this article, but I'm going to have to disagree with the reasoning in it. I mean, I really like this paragraph:

"I believe NFL teams put too much emphasis on a player's reputation and their desire to "upgrade" a certain position and not nearly enough time trying to quantify what that increased performance will provide in terms of value to their team. Football is not a purely statistical game -- there are intangible qualities a player can bring to a team in addition to his on-field production that need to be considered. That being said, there is room in the NFL for a more Moneyball-type quantitative analysis of what a given player's value will be to a franchise, especially at some of the positions like the line, where numbers are not a huge part of the equation."

I think it shows how FO's thinking is slowly permeating into the mainstream of the NFL. Still, I think he makes a pretty critical error here:

"Strictly by the numbers, that means that an upper echelon player would give up two fewer sacks and five fewer pressures, and probably make one more block in the running game in eight games over the course of the season. That means the difference between one of the highest paid guards in the league like Hutchinson and a journeyman interior backup like me is approximately 15 plays over the course of the season, give or take a few."

as noted in #1.

It’s not that Hutchinson might make one more block in the running game per game. It’s that every block he makes has a bigger impact

Exactly. I could do the same thing with QBs by pointing out that, given an average number of attempts, an elite QB would only complete about 25 more passes than an average QB. That doesn't mean that elite QBs are not much, much more valuable than average QBs, because elite QBs do more on every play.

And the inclusion of kickers in the list of high impact positions is just daft, obviously.

He's probably just thinking of game winning kicks. It's a very common error to attribute the entire win to the last person that got the team the lead, so I'm not too surprised by it.

7
by Joseph (not verified) :: Fri, 04/04/2008 - 5:52pm

I think the last paragraph(s) sum up WHY the Jets signed Faneca. I believe it was mentioned on FO as well. He will upgrade both the LT & C positions by his consistent play. Not to mention his work ethic, tips he will share with D'Brick & Mangold, etc.
Where I think Mr. Tucker's analysis IS right on is on the "2nd tier" free agents, the guys just below the superstars. Is the guy you signed for prob. too much (because plans A, B & C at that particular position signed with other teams) really THAT much better than a "street free agent"?For example, who will be the better CB for the Raiders--Hall or Fabian Washington? (I believe F.W. was the Raiders' #2 CB) Or even better--the Raiders' RB's last year--LaMont Jordan vs. Justin Fargas. The Saints' Reggie Bush vs. Aaron Stecker/Pierre Thomas. The list goes on. Another great comparison from last year--Matt Schaub vs. Sage Rosenfels, Texans' QB's.

8
by 53Deep (not verified) :: Fri, 04/04/2008 - 6:05pm

It's not true. The difference between an average lineman and an All-Pro is probably closer to five to ten plays A GAME.

The difference between "a scrub" and an elite player is more like ten to fifteen plays a game.

By the end of the summer I should have enough data from the 2007 season to back this up.

9
by Pat (not verified) :: Fri, 04/04/2008 - 6:11pm

That said, I can’t see a $35M a year offensive line every happening.

Are you sure? Before the Redskins restructured a few deals, the Skins were in line to have a ~$30M/year offensive line this year. Now they're "down" to $20M.

10
by Will Allen (not verified) :: Fri, 04/04/2008 - 6:23pm

A claim that a market is massively inefficient, with nearly every participant spending their money nonsensically, should be viewed very skeptically.

11
by Bobman (not verified) :: Fri, 04/04/2008 - 6:51pm

This is pretty good stuff, but I too think he whiffs on a few calls.

Running game blocking fungible? I think the UPGRADE in Indy from Pro-Bowler Tarrick Glenn ro Rookie Tony Ugoh at LT was pretty pronounced. Now they may have tweaked other things along the line to help Ugoh, but he seems to be an upgrade over a very good veteran LT in the running game, and maybe just a minor downgrade if at all in the passing game. In the real estate appraisal biz, that's close to a "matched pair analysis" in which everything else is the same except for the LT. It really focuses things.

Also, I suspect (but he sure knows better than I) that because it's a game of inches, a superior blocker will keep a DT 12 inches back from the LOS and control him decently, but an inferior blocker will contain the DT right on the LOS. Now in both cases, the DT is contained and does not get penetration, but his position in one instance may alter the RB's pass route, or make the RB stay in in case of a jail break, or put the slightest pressure on the QB--make him take one step deeper, and therefore throw off the timing/placement of throws, or make him flush a half second sooner.

It's not the actual measurable hits, it's the reliability, the comfort the QB has. On film, both DTs did not penetrate and neither got a pressure or sack, but one being just one foot closer might have had an impact on the actions and minds of five playes on the other team. And if it happens on 30 pass attempts per game, that's a lot. No pressures and no sacks, but.... say the QB goes 20/30 that game. On those 10 incompletes, how many were attributable to his RB not getting out into a pattern because he stayed in to block a second longer than he needed to because the DT looked like he was getting close? How many because the QB took one step to the right and then threw a pass just one foot off target? Say out of those 10 incompletes, four are because of this "average" blocker. Then put him against an elite DE and see it get worse.

Tucker knows a lot more about this than I do, but I can certainly imagine a world where an elite blocker's superiority is measured in mere inches, and STILL it's very significant.

12
by Stuart Fraser :: Fri, 04/04/2008 - 9:03pm

11 - ah, but there's the rub, isn't it? It depends on who the QB is. If it's Tom Brady, he's going to step into the throw anyway, make the completion and quite possibly take a hit. If it's JP Losman, he'll probably not step into the throw as a result and then throw incomplete three feet over everyone.

(If it's Ben Roethlisberger, he'll hold the ball another two seconds, allowing the DT to beat the blocker, after which Roethlisberger will spin away from the DT who only manages to get a hand on him, and complete to Heath Miller for a first down. But I digress).

Likewise, if the TE is Jason Witten or the aforementioned Miller they probably chipped the problem rusher on the way out into their pattern, so the average tackle or guard can do the job of a much better player. If it was Jerramy Stevens, not so much.

Football isn't really about individual matchups, and you can always compensate for a weakness with a strength elsewhere. And it may be that being strong on the interior line isn't worth it if it means pricey free agent contracts.

I don't know that I agree with Tucker, but the only one of the supposedly well-run teams that have been participants in the make-guards-rich spree has been Dallas, and they've always been the risk takers of that set. Which might be taken to indicate he's not entirely barking up the wrong tree.

13
by Fred (not verified) :: Fri, 04/04/2008 - 10:41pm

I don't understand how he can say the difference between an average starter and a superstar isn't that big.

Everyone else remembers the Week 4 Giants-Eagles game where Osi abused Winston Justice for 6 sacks. Week 14 when Tr..... sorry... Willi....wait he changed back to Tra Thomas was recovered and started at LT Osi only mustered one sack. In fact, according to the dallas news (The first site google popped up with the stats) Thomas only allowed 7 sacks in the 15 games he started.

Yeah, this is only a singular situation but it shows the difference between a 5 Million Dollar A Year Veteran and some 2nd Round draft Pick.

If it costs 21 million guaranteed to get the best player on the market at the position then so be it.

14
by mm (not verified) :: Fri, 04/04/2008 - 10:53pm

You guys missed the follow-up article he published just today (click for link), where he claims left guard is easier to play than right guard, and thus guys who play left guard are overvalued (at least compared to right guard).

If he's right, you'd probably be better off signing a good right guard and flipping him to left guard than signing a top left guard.

15
by masocc (not verified) :: Fri, 04/04/2008 - 10:56pm

If only somebody could come up with a way to statistically evaluate line play, and thus be able to make comparisons based on lineup changes...

16
by Fred (not verified) :: Fri, 04/04/2008 - 11:01pm

Of course it also makes a difference reading this article after seeing that he posted a follow up article here. Where he ranks Left Guard as the 4th hardest position to play on the the line, ahead of only Center. which would explain why he finds 40M for Faneca a bit extravagant.

17
by Bobman (not verified) :: Fri, 04/04/2008 - 11:48pm

Stuart, absolutely it's all interrelated, but if a position coach and a position player are sitting down in front of a TV to evaluate said position player, they are hopefully taking into account the TE chipping and things like that. (i.e. an OT does not get a bad grade if he succeeds with a TE's help, but he might not get a full good grade either; if he allows a sack either way, chances are he's at least partly to blame. If it's Joey H, David Carr, Ben R, he can claim extenuating circumstances)

15, Dream on. It'll never happen. Too complicated. Football is a simple game played by simple guys. 3 yards and a cloud of rubber, sub-turf cushioning pellets. No way somebody's gonna... hunh. Oh hey, how about that weather out there?

#13, Fred, good example. Maybe it's point of view: I am guessing that the delta is not that big if you're the 2nd stringer, but it's "huge" if you are the superstar, and even bigger if you are the immobile franchise QB who, while staring straight up at the stadium lights for the fifth time in one half, decided to nick-name Mister "No Big Deal" The Human Turnstyle."

18
by Tom D (not verified) :: Sat, 04/05/2008 - 12:18am

Isn't Harrington actually really good at avoiding sacks?

19
by Raiderjoe (not verified) :: Sat, 04/05/2008 - 2:02am

Fanecia not good signing for Jets. Jets are another laughinfstock team like Broncos bengals and Falcons. When was the last time the Jets were any good?

If Jets sign a player and you are betting man you should bet on the player being crappy.

I drnak some Meister Brau tonihte. so hat if the Jets are not my favorite ream? It desn't mean I cant say stuff about them. good chance Raiders drat Daren MNcFadden or Vernon Gholsto frm Ohio state. I prefer Gholton. wouldn't ind backfield of Jordan, Fargas, Mcfafdden, Griffith (best fullback in elague no) (don't think because he was noyt in Pro Bowl that he wasn't best fullback in league in 2007. He was. Lornon Neal got in because he got in other yeras and people who do voting are stupid. Neal didnnot desrevr it. Girfiiht much better player

too drunk to witr move more

20
by Raiderjoe (not verified) :: Sat, 04/05/2008 - 2:04am

meant to say I prefer D McFadden. Been saying that since 2007

21
by masocc (not verified) :: Sat, 04/05/2008 - 2:16am

I need a Raiderjoe to English dictionary. Or perhaps just a case of cheap beer that's trying to hide behind a German name.

22
by Alex (not verified) :: Sat, 04/05/2008 - 2:31am

Isn’t Harrington actually really good at avoiding sacks?

Yes, one of the best in league history, in fact. But the Falcons had such a god-awful mess of an O-line, putting players into a blocking system they simply didn't have the skill set to run, that Harrington still got sacked a lot.

23
by bubqr (not verified) :: Sat, 04/05/2008 - 7:05am

I'm really surprised a OL said this. Everyone who played the game know that, as said #7, Faneca is going to help the C and LT, 0.2 s gained on a combo block allowing the combo partner to get to the LB faster, some inches on a down block to allow a better pull block from the player next to him, some inches to allow the QB to step up into the pocket... That's what a good OL, bring, might be details, but isn't this game won or lost because of these details ?

24
by johnt (not verified) :: Sat, 04/05/2008 - 9:15am

Faneca is going to help the C and LT? Has anybody actually watched this gentleman attempt to pass block in the past 3 years?

25
by ammek (not verified) :: Sat, 04/05/2008 - 11:04am

Both Tucker articles are informed, unorthodox and clearly explained for the layman. Good find!

Tucker dismisses the value of the center, and yet several of the most celebrated lineman have been centers: Webster, Dawson, Stephenson, Otto....

Sure, free agency didn't exist in their time, but I wonder what Tucker would make of the relative value of these HoFers, compared with, say, a decent but not all-Pro left guard.

26
by Jerry (not verified) :: Sat, 04/05/2008 - 3:07pm

Re #25:

Tucker dismisses the value of the center, and yet several of the most celebrated lineman have been centers: Webster, Dawson, Stephenson, Otto….

Sure, free agency didn’t exist in their time, but I wonder what Tucker would make of the relative value of these HoFers, compared with, say, a decent but not all-Pro left guard.

In the recent piece, Tucker said "Though some athletic centers can create additional value with their ability to pull or effectively block at the second level..." I think those are the guys you're talking about. (Merrill Hoge tells the story of when he went to the Bears, and the coaches asked him to diagram a Steelers bread-and-butter running play. He drew it up, said "The center pulls like this," and the coaches responded, "You can't pull a center like that." That was the value of Dermontti Dawson.)

27
by Joey Jo-Jo Junior Shabbadu (not verified) :: Sat, 04/05/2008 - 6:05pm

I feel like the impact of Hutchinson to the Seahawks and Vikings may be significantly overrated. Seattle fans blame his departure for the crash in their running game, but he left at the same time that Shaun Alexander decided he'd rather drop into the fetal position than run through tackles. And Minnesota fans are giddy at their current running game, but it only became deadly once the team acquired Purple Jesus. In 2006, with Hutchinson, the team gained a pedestrian 4.1 yards per carry.

I'm not saying Hutchinson isn't a good lineman; only that there are almost certainly other factors at play here.

28
by Fred (not verified) :: Sat, 04/05/2008 - 10:51pm

#27-That same Minnesota team had the DVOA 30th ranked passing offense and Chester Taylor missed a game and at 4 YPC still managed have the 9th most rushing yards in 2006.

The increase that the Vikings got with Peterson was the threat to go to the house any time which their passing game didn't really give them in 2006.

And Seattle went from 3rd in rishing DVOA with Hutch to 26th without him. People say Alexander stopped taking hits but according to TMQ (And i remember reading it somewhere else but cant find it now) Alexander scored 15 of his 27 TDs in 2005 Untouched on the way to the endzone. Can you say good blocking?

29
by Bobman (not verified) :: Sat, 04/05/2008 - 11:43pm

I second #28. Live in Seattle, though not a big fan, but I thought Jones deserved the MVP that Alexander got two years ago and that Hutch should have been re-signed. Nothing that I have seen from Minny or Seattle since then has changed my mind.

30
by masocc (not verified) :: Sun, 04/06/2008 - 2:56pm

Re #27:
Shaun Alexander has *always* dropped into the fetal position at the first sign of contact. The only difference in 2005 was that he often went untouched, as Fred in #28 pointed out.

Obviously, the loss/gain of Hutch isn't the ONLY reason for the decline/resurgence of the Seahawks/Vikings running game. He's only a "mere" 75% of the equation.

31
by Ralph Hickok (not verified) :: Sun, 04/06/2008 - 5:17pm

RE: #28
It was Tuesday Morning QB, Gregg Easterbrook, who came up with that figure. He mentioned it more than once, possibly as many as three or four times.

32
by NF (not verified) :: Sun, 04/06/2008 - 6:26pm

I think Tucker's argument is flawed, because he assumes that linemen differ the most in their contribution to the passing game but differ the least in their contribution to the rushing game. FO stats have shown that there can be wide differences in QB pressure between QBs behind the same offensive line (Drew Bledsoe and Tony Romo), but that a great rushing game is more often the product of a great offensive line than a great running back (Edgerrin James going from the Colts to the Cardinals, any Denver RB except Terrel Davis and Clinton Portis, too many backup RBs behind good lines to count).

33
by PaulH (not verified) :: Sun, 04/06/2008 - 8:08pm

As some people have pointed out earlier, Alexander has never been a physical runner. As a diehard Alabama fan, I've been watching Shaun Alexander ever since he ripped LSU to shreds in 1996 as a freshman, and I can attest to the fact that he has never been a physical runner, even going back to his college days.

Shaun, though, has been hurt a good deal by the drop in offensive line quality. More so than many other backs, he needs a good offensive line in front of him. He's not extremely athletic or a big bruiser, so he really has trouble turning busted plays into otherwise semi-productive plays. Shaun's strength has always been getting good blocking up front that allows him to get to the line unscathed, or slightly onto the second level, and from there he can use his great field vision and agility to hurt defenses.

He has never been one of those guys who could run over a linebacker in the hole, or who could juke two guys in the backfield and turn the loss into a decent gain. He has always made his money at and beyond the line of scrimmage with great vision and good hips in open space. In that sense, he needs a good OL more than a lot of backs, and that has probably what has hurt him so much the past couple of years.

34
by John Doe (not verified) :: Mon, 04/07/2008 - 2:19am

I think claiming that there is no difference in run blocking between elite linemen and average linemen is silly. Merely engaging your man is different than driving him off the line. Pass protection is actually easier on the interior linemen than run blocking. Depending on the scheme the guard may have to help turn a DT with the C or T and scrape off to blow a linebacker out of the running lane. Plenty of guards are expected to pull and lead block for RBs. Pass blocking is pretty much set your feet and try not to get blown backwards.

I think O linemen are worth the money because they improve all of your skill positions. Quarterbacks benefit from extra seconds to hold on to the ball, Running Backs benefit from bigger holes, and WRs get more time to get open.

35
by John Doe (not verified) :: Mon, 04/07/2008 - 2:29am

I'll add that I'm glad the Giants have Chris Snee. If they could get an above average LT and move Diehl back to guard they'd have a shot at being the best O line in the league.

36
by Paul (not verified) :: Mon, 04/07/2008 - 9:32pm

I'm new to this site, but has anybody ever attempted to do a study that would correlate success with salary cap allocation between the various positions?

I don't know how much variance there is from team to team when it comes to allocation, but I think it would be interesting to chart whether salary cap allocation makes a difference.

You could measure standard deviation. (Is it better to pour money into big stars or spread it evenly?)

You could measure offense vs. defense (is it better to put more money, relatively speaking, on one side of the ball?)

You could measure starters vs. bench (Is it better to pour all money into the frontline guys or pay for depth?)

You could could measure draftees vs. free agents.

I think it might be interesting. Any thoughts?

37
by Paul (not verified) :: Mon, 04/07/2008 - 9:35pm

One more - skill position vs. linemen. Where do you get more for your salary cap buck?

Such a study would not account for poor talent evaluation (i.e., Millen's skill position players do not bring skill), but it still might give you a pretty good idea how to build a winning team.

38
by masocc (not verified) :: Tue, 04/08/2008 - 12:26am

Re #33: Thanks. *Just* when I had almost gotten memories of Shaun Alexander bumping into a lineman and falling down for a loss of a yard on 3rd and 1 (usually on a draw) out of my head, you bring the nightmare crashing back. You farking icehole bastich.

39
by Cyrus (not verified) :: Tue, 04/08/2008 - 9:22am

Paul--

I expect there to be a huge variance, based on two completely opposite examples: the Patriots and the Colts.

The Patriots put a lot of money into certain positions, leaving others weak. (DL is one money position, and I hope LB, whereas CB and WR is not)

They typically don't pay a lot of money to individual players, preferring to sign people to veteran minimum. I would wager that the majority of their cap money is on defense.

The Colts, on the other hand, put a lot of money into certain positions (QB, WR) and typically use a lot more of their cap room to reward these players.

They then use the draft, instead of veteran minimums, to fill in the holes, which can sometimes make the positions that would be considered "weak" to be a strength. RB and CB are positions I consider them to shy away from paying big money to, but since they used their 1st/2nd round picks on them, they are still a strength.

Once Addai gets to free agency we will see how things change, but as of now, I see two completely different approaches.

Patriots/Colts:
Defense/Offense
Money spread/Money in a few elite players
Veterans/Draft

40
by CA (not verified) :: Tue, 04/08/2008 - 4:48pm

Re: 1 Does anyone believe Hutchinson hasn’t made a huge difference for the Vikings?

[Raises hand] I think you picked the worst example possible to make your point. Hutchinson received what was then by far the biggest contract in history for a guard, and he proceeded to play mediocre football over the next two seasons. He appears to me to have been an above average but inconsistent and by no means great run blocker, and, if anything, he has been a liability in pass protection. That didn't stop him from making the Pro Bowl, and the largely self-generated success of Adrian Peterson and the very strong performance of the underrated (and relatively underpaid) Bryant McKinnie have helped to perpetuate the myth of Hutchinson-as-elite-lineman. Clearly, conventional wisdom with regard to Hutchinson has yet to catch up with the reality on the field. That's JMHO as someone who has watched him closely during his time with the Vikings. Full disclosure: I was highly critical of the signing at the time because I thought no guard was worth that kind of money. If there is a guard worth that kind of money, I can now say with some confidence that it sure ain't Hutchinson.

Re: 10 A claim that a market is massively inefficient, with nearly every participant spending their money nonsensically, should be viewed very skeptically.

Yes, of course, be skeptical, but just such a claim is the foundation of Moneyball, a book and movement with which the Outsiders associate themselves.

41
by Aaron (not verified) :: Wed, 04/09/2008 - 3:15pm

There's actually quite a lot of irrationality in market systems. Their strength is efficiency and adaptability, not necessarily in being "right" (though these often intersect).

Further, the rationality of an open market decision is entirely dependent on the accuracy and objectivity of the information available to the one making the decision. In the stock market, this is much easier to do - a lot of the indicators are already quantified for you, and the methods of quantification are pretty obvious.

In football, this is practically impossible. Even efforts such as FO can only attempt to approximate objective, quantifiable standards of value for players.

In other words, the free agent market is a prime candidate for irrational decision-making, due largely to extremely imperfect information.

42
by crack (not verified) :: Thu, 04/10/2008 - 2:44pm

shorter 40:

I didn't like the Hutchinson deal and will ignore any evidence that I was wrong.