Writers of Pro Football Prospectus 2008

30 Mar 2010

ESPN Insider: Blown Blocks 2009

This week's ESPN Insider feature takes a look at the best and worst offensive linemen in 2009 according to blown blocks tracked by our game charters.

Posted by: Bill Barnwell on 30 Mar 2010

20 comments, Last at 05 Apr 2010, 7:05am by DeltaWhiskey

Comments

1
by The Other Ben Johnson (not verified) :: Tue, 03/30/2010 - 3:58pm

Those Tennessee stats seem dependent on something... maybe defenses were scared to death of Chris Johnson and kept honest by Young's rushing ability?

2
by Theo :: Tue, 03/30/2010 - 5:16pm

This is not insider but outsider... when?

3
by Jimmy :: Tue, 03/30/2010 - 6:03pm

It is a result of the quantum nature of the outsiders. They can exist both inside and outside football. It is only in atttempting to measure whether they are inside or outside that their state is resolved. It turns out they are a dead cat.

5
by Grendel420 (not verified) :: Wed, 03/31/2010 - 2:32am

That was a brilliant comment.

4
by Rick B. (not verified) :: Tue, 03/30/2010 - 7:33pm

Was Jason Peters really that bad last year?

I thought he did fairly well in pass protection in 2009, other than the few times he was injured, and the playoff loss to Dallas.

According to ProFootballFocus, he was the sixth best LT in pass protection last year.

I don't think he was elite by any means, but he wasn't a scrub.

6
by DeltaWhiskey :: Wed, 03/31/2010 - 2:56am

"the idea that you can grade players off of TV tape is just absurd."

-by Bill Barnwell :: Tue, 02/16/2010 - 6:22pm

"it is downright impossible to tell what the responsibilities are of every player on the field on a given play, and whether those players succeeded or failed at their tasks, let alone the unholy amount of time and work it would take to track that on every play."

"How can you tell whether a receiver made the proper sight adjustment and cut his route accordingly when it's off the screen? Whether an offensive lineman was supposed to go to the second level and block instead of double-teaming a nose tackle?"

- Bill Barnwell :: Fri, 02/19/2010 - 10:24am

http://www.footballoutsiders.com/four-downs/2010/four-downs-afc-west

"The 80% idea is, again, a step in the right direction, but again, you can think that you're 100% sure of what a player's responsibility was on a play and still be totally wrong. This is where the offensive lineman example in my "diatribe" comes into play."

-by Bill Barnwell :: Wed, 03/03/2010 - 8:27pm

http://www.footballoutsiders.com/extra-points/2010/dunta-robinson-and-ch...

9
by Fan in Exile :: Wed, 03/31/2010 - 12:38pm

I totally didn't catch the hypocrisy of this article until you poster these links. Thanks I had a great laugh.

10
by Aaron Schatz :: Wed, 03/31/2010 - 1:02pm

The game charting project is not *grading* players. It is simply counting when we see a blown block and presenting totals. "Blown block" has nothing to do with run-blocking or whether a guy was supposed to go to the second level. It's an attempt to count blown blocks in one-on-one blocking on pass plays, that's all. We make no attempt to judge events that take place off the screen, nor do we claim to. It's limited, but we're up front about the limits. There's nothing hypocritical when the game charting project doesn't actually attempt to do the things Bill is calling impossible to do in those quotes.

11
by Fan in Exile :: Wed, 03/31/2010 - 2:00pm

I'm sorry it's my fault. I didn't understand that you can't grade a player but you can flunk him. I also didn't seem to get that you can't grade because you don't know the assignment, but you can flunk him when he fails it. Whatever helps you sleep at night.

12
by jfsh :: Wed, 03/31/2010 - 3:11pm

Do you really not understand the difference? Or are you just being dense? You can pretty clearly watch someone get beaten in pass protection, because the goal is pretty straight-forward: get to the quarterback. To the extent you want to argue that linemen intentionally let players through for strategic reasons, I admit that it's possible, but I really can't think of too many situations where that would be the case. I would imagine that releasing your player to the running back / tight end / whatever to pick up someone else would not count as a "blown block."

Frankly, I'm sick of the endless complaints on this site whenever FO's methods are shown to be less than perfect. If you have constructive criticism to add, you can do so with a civil tone. I love this site, and I'm extraordinarily grateful that I've been able to read their analysis -- for free -- for so many years. If you don't feel that way, fine, just stop reading / commenting. Don't ruin it for the rest of us.

13
by Fan in Exile :: Wed, 03/31/2010 - 5:22pm

Wow, do you not understand what's going on. To actually add anything to this conversation you need to argue that what FO does is significantly different from what PFF does. This isn't a complaint about what they're doing it's a comment on the attacks that one of the authors has made about another site. Also some of us pay for some of what they do, so just because you've been mooching off of them doesn't mean you need to lay a guilt trip on other people.

15
by dmb :: Wed, 03/31/2010 - 10:08pm

You can count blown blocks without trying to infer the intent of the play. If an offensive lineman is engaging with a defender, and that defender is able to beat the lineman to make a play, then you've got a blown block. That says nothing about whether the lineman was trying to engage the correct defender, nor does it assign responsibility to a lineman if someone comes in unblocked. It's a reasonably substantial difference in methodology from PFF, and probably the more sound one.

However, as Rick B. points out, I would agree that going on to rate linemen based on this alone may not be any less speculative than what PFF does. Since we don't know how often each lineman blew an assignment, it could be that a lineman who came out as average using only blown blocks would turn out to be a bottom-feeder if we factor in missed assignments. There are also the issues of splitting QB/lineman responsibility, etc.

Finally, I don't understand the animosity in this thread. You may disagree with Aaron, or PFF, or another commenter, but that doesn't mean you have to write in a tone that one might use when accusing someone of peeing in his cereal.

17
by DeltaWhiskey :: Thu, 04/01/2010 - 10:17am

It is somewhat different from what PFF is doing. As noted it is pass/fail, but I dont' see anywhere on PFF's site that they try and infer what the player was supposed to do, they grade what they saw, just as you describe the process in the Game Charting Project.

RE: Aaron's comments (BTW, thanks for commenting)

- PFF seems pretty damn open about what they're doing. They too acknowledge they can't see every player on every play, and my point is that I believe that if Mr. Barnwell's criticism of PFF is accurate, then regretably, much of it is accurate for the Game Charting Project as well.
- Maybe blown pass blocking is more objective, but there are still many factors that cannot be accounted for. For example, is a chip block with the expectation that the running back will pick-up the rusher a possible missed block? How about if a blocker is supposed to "hand off" a player to another blocker, but the latter misses it? What if the blocker was expecting help to double team, but that help never came?

My point is and was that I believe that if Bill's criticism of PFF is accurate, then regretably, it is on some level, accurate for the Game Charting Project as well.

Frankly, Bill has been a complete and obtuse ass in regards to the PFF data and has been throwing stones in a glass house.

19
by Jimmy :: Fri, 04/02/2010 - 12:16pm

I do think FO do deserve some credit where it is due regarding the claims they make on the back of their charting data. FO admit that while they are charting every play they are doing it from TV footage and are therefore inherently limited by the quality of the angle the producer has decided best shows (or hides entirely) the action. For the most part they only express some fairly basic numbers from the charting, eg. number of blockers and rushers on pass plays or who was the closest defender in zone coverage. Any judgements subsequently made aren't normally simply an expression along the lines of 'limited charting says X therefore it must be true'.

What sets FO above its rivals in their chosen field is DVOA. The fact that the majority of the charting data is used to help analyse the information DVOA is producing is a good reason why FO are better than PFF. You can start to see whether a team is succesfull off play action or if they run the ball well on first down. Using the charting to examine which formations or strategies are most effective according to DVOA seems more likely to produce valid results than just looking at the footage shot from bad angles.

20
by DeltaWhiskey :: Mon, 04/05/2010 - 7:05am

" I do think FO do deserve some credit where it is due regarding the claims they make on the back of their charting data. FO admit that while they are charting every play they are doing it from TV footage and are therefore inherently limited by the quality of the angle the producer has decided best shows (or hides entirely) the action."

PFF does this as well:

"As with any type of analysis we are always at the whim of the TV companies who seem to think showing a QB or HBs face right up until the snap somehow makes good coverage."

http://www.profootballfocus.com/about.php?tab=about#grad9

I just don't think that FO should be throwing stones without some sort of empirical support for their assertions.

I agree that DVOA is certainly FO's bread and butter and the the the charting project can enhance DVOA significantly. FO is clearly interested in team performance, while PFF is focused on individual performance.

14
by Rick B. (not verified) :: Wed, 03/31/2010 - 5:23pm

Here's my problem with this piece: when you give an article a loaded title like "The best and worst O-linemen", you are insinuating that whatever conclusions you draw based on the number of blocks blown leads to a definitive analysis of offensive linemen.

When you write:

"Left Tackle
Worst: Jason Peters, Philadelphia Eagles (7 sacks/4 hurries/11 total blown blocks)"

You are condemning Peters as the worst tackle in the NFL based on something as nebulous as "blown blocks". I fully understand if you wanted to use this article to draw attention to, simply, an isolated stat, but to imply that your analysis has the power to objectively quantify the total performance of an offensive lineman based on that isolated stat is very misleading and disingenuous.

For example: if Peters completely shuts down an elite pass rusher in a single game (DeMarcus Ware in Week 16), but has a very bad "off" day on another occasion in which he has multiple blown blocks, do you still call him the worst offensive lineman in the NFL? Would you consider him worse than a LT who consistently struggles against elite pass rushers and never has the ability to shut someone out for a game? I would call him "inconsistent", but hardly the worst LT in the NFL.

16
by tuluse :: Thu, 04/01/2010 - 12:41am

I'm pretty sure an ESPN editor writes the headlines, not the author of the article.

18
by Rusty G. (not verified) :: Thu, 04/01/2010 - 10:21am

I love the work you guys do in general, but with all due respect, I'm not so sure I agree that you were "up front" about the limitations of this type of analysis at all.

Where in the article do you take into account the number of drop backs over which those "blown blocks" occurred, or note that the analysis does NOT account for this? You show the "number of games" for Tyler Polumbus, I see, but that's all I could account for.

Where do you note that the presence or absence of tight ends, and biases of certain offensive schemes with respect to additional blocker utilization, may impact whether an individual has been categorized as the "best" or "worst" at his position?

Where do you note that quarterback delivery has an impact, beyond bifurcating sack vs. hurry totals? You do not point out that this factor also has a material impact on whether a given play is a hurry or a successful block, as well!

You certainly state that this is all based on how you look at blown blocks, but it's just a bit unsatisfying to me when you surround that with, at worst, deceptive, and at best, misleading copy that draws a pretty straight line between this stat and conclusions on individual skill. Hard to criticize the headline - ESPN probably added that - but it, too, contributes to an unsatisfying whole.

I could be in the minority - but part of me just revulses at the idea of presenting interesting analyses... and drawing tenuous conclusions from them. Wasn't this what FO was about? Combatting uninformed conclusions media, players and fans drew from loosely or equivocally related stats presented without context and rigor?

7
by Noah of Arkadia :: Wed, 03/31/2010 - 10:17am

What happened to the old spam bot? I don't like the new one :(

8
by Mr Shush :: Wed, 03/31/2010 - 11:10am

I'm torn. On the one hand, it's kind of annoying having to scroll past a massive list of nonsense. On the other, I'm genuinely intrigued by this particular spam outbreak. What the Hell is it for? Is there anyone, anywhere in the world who would click on those links? What does the writer of the spambot think it will achieve? Or is this some kind of emergent spam, a byproduct of the slow, bumpy and frankly terrifying road towards an accidental gestalt internet consciousness? Will Skynet subjugate humanity by means not of H/Ks and death camps but porn and mail-order viagra? I admit, it sounds like a more pleasant way for a species to go out . . .