Writers of Pro Football Prospectus 2008

09 Dec 2005

Elias Eliminates Roye's Sacks

I think sacks are overrated, not because I don't think tackling the quarterback behind the line of scrimmage is important, but because sacks get so much more attention than other tackles behind the line of scrimmage. A good example is what happened to Orpheus Roye this week. On Sunday he tackled Jacksonville quarterback David Garrard behind the line of scrimmage twice. He was initially credited with two sacks. Then the Elias Sports Bureau decided that Garrard was trying to run on both plays, and those sacks were taken away from him. It doesn't change what he did to help the Browns on Sunday, but it will change public perceptions of what he did to help the Browns Sunday.

Posted by: Michael David Smith on 09 Dec 2005

26 comments, Last at 19 Dec 2005, 1:22pm by Michael David Smith

Comments

1
by Alan Milnes (not verified) :: Fri, 12/09/2005 - 8:13am

So when do they get around to taking away Strahan's "sack" against Brett Favre?

2
by andrew (not verified) :: Fri, 12/09/2005 - 8:40am

I always hate it when they cite players from the 80s onwards (Bruce Smith, Reggie White, Chirs Doleman, etc) as having the most sacks in history, but fail to note that sacks were not tallied before then. Players like Deacon Jones, Alan Page, etc are not mentioned amongst the top sackers. You'll see some sack statistics for them sometimes but that' is a team source that tried to go back and catalog all the plays so they could tally sacks, but I don't think those are as accurate (and in theory not as neutral either).

3
by big_adventure (not verified) :: Fri, 12/09/2005 - 10:07am

"but it will change public perceptions of what he did to help the Browns"

Not to mention slugging him hard in the pocket come contract time. I have no idea what his stats are now (and don't have a chance to check) but stepping from, say, 8 to 10 sacks would be HUGE, I would guess, while stepping from 49 to 51 tackles is relatively meaningless.

-Sean

4
by Mike (not verified) :: Fri, 12/09/2005 - 10:31am

I've always said sacks are the most overrated stat in football while hurries are the most underrated.

It isn't getting to the quarterback and sacking him 12 times in 16 games that makes a pass rusher valuable after all. It's all the stuff they do even when they don't lay a hand on them.

5
by admin :: Fri, 12/09/2005 - 10:42am

And now, presenting the answer to the question, "Why do FO sack totals not agree with the sack totals on NFL.com?"

Elias usually changes this stuff without telling anyone, and their reasons are awfully sketchy. If it is clearly a pass play, and a guy pump fakes, but pulls his arm down, and tries to scramble a bit, and doesn't have his arm up when he's tackled, why is that not a sack?

One change I'm considering for the off-season is to count the following in adjusted sack rate: a) intentional grounding and b) any play by a quarterback that results in a loss, other than aborted snaps. That would get rid of this goofy "sack becomes a run" problem.

6
by Jason O (not verified) :: Fri, 12/09/2005 - 11:18am

The ESB also changed a play to a sack in the Colts/Titans game. The play in question featured Adam Jones running right, where he was tackled about 13 yards behind the line of scrimmage. It was originally called a TFL, but was changed to a sack, giving him the league lead back.

Click my name for the link. :)

7
by Kibbles (not verified) :: Fri, 12/09/2005 - 11:38am

Re #4: Exactly. Think back to the Denver/KC game on MNF earlier in the season. I'd imagine a lot of you guys remember Trevor Pryce wreaking havoc from all 4 defensive line positions. I bet a lot of you remember that he got named the Horse Trailer player of the game. I also wouldn't be surprised if many people didn't realize that he didn't get a single sack during that game. He was in Green's face all afternoon, and affected nearly every single play, but anyone who didn't see the game wouldn't have even considered the possibility that he had a good game, because he didn't have the sacks.

I would love it if the NFL stopped tracking sacks and started tracking hurries, instead.

8
by Ben (not verified) :: Fri, 12/09/2005 - 11:57am

Hurries are also a bit subjective. I don't know why they don't track tackles for loss like they do in college.

9
by Devin McCullen (not verified) :: Fri, 12/09/2005 - 11:57am

Hmmm...you know, I could swear that somewhere I read something about sacks being more damaging to an offense than other losses of yardage. Can't think of where it was, though. :)

10
by Matthew (not verified) :: Fri, 12/09/2005 - 12:25pm

Re: #7 - Maybe it was just because I was watching more closely than usual, but to me, Pryce had one of the most dominating performances by a defensive player I've ever seen. And you're right, he had virtual no stats to show for it.

11
by Sam O (not verified) :: Fri, 12/09/2005 - 12:28pm

#9 Devin

MDS is not saying that sacks aren't great. They are because of the loss of yardage and high rate of fumbles. But most d-linemen are judged based solely on how many sacks they have, when that only paints a small picture of a players effectiveness. So a sack is not overrated, but players are overrated because of sackss.

12
by Michael David Smith :: Fri, 12/09/2005 - 12:36pm

Right. The point is, people who judge defensive linemen by looking at how many sacks they got (and many people do, including many media types who get votes for the all-pro teams) are going to think less of Orpheus Roye not because of anything he did but because of a decision the Elias Sports Bureau made after the game.

13
by Ryan Mc (not verified) :: Fri, 12/09/2005 - 12:42pm

This is like in Jevon Kearse's rookie season for the Titans in 1999. Against the Falcons, whilst about 1 or 2 sacks short of the rookie record, he dumped the QB for a huge loss. On replay it was clear that the play was intended to be a handoff to the RB out of shotgun formation, but the snap was high causing the QB to have to jump up to catch it, miss the handoff and get dumped by Kearse. The play was changed to a tackle-for-loss rather than a sack after the game, but it didn't change the amount of yardage that Atlanta lost on the play.

I think sacks get more attention, though, because generally speaking I would guess that sacks are usually for a larger loss of yardage than a tackle on a running back, and more likely to cause a fumble. I don't have any stats to back this up, though, just seems that way (with the exception of tackles on end-arounds/reverses which can be for huge loss of yardage sometimes.)

14
by Sophandros (not verified) :: Fri, 12/09/2005 - 12:55pm

#8

I don't see why they don't do that, because to me it's counter intuitive to count a sack as negative passing yards.

15
by benw (not verified) :: Fri, 12/09/2005 - 1:33pm

Along the same lines, Derrick Burgess had two sacks week 12 against Miami (and the team had three). That might imply that they were getting a lot of pressure on Ferrote. But the truth is that those were the only three times they got any pressure on him. He had plenty of time to throw on almost every other pass play.

Hurries or hits on the QB or intentional throw aways would help round out the statistics so that they paint a better picture of how effective the pass rush is.

16
by admin :: Fri, 12/09/2005 - 1:59pm

In the game charting experiment, we're trying to track not just hurries and intentional throwaways but which defensive lineman was most responsible, similar to a sack.

17
by Will Allen (not verified) :: Fri, 12/09/2005 - 2:11pm

Another example of how statistically evaluating individual line play is so difficult. The Vikings probably aren't going to give up a whole lot more sacks this year than in the previous three, which, for someone just looking at sack totals, might mean that the Vikings offensive line is not thought to have declined all that much.

The reality, of course, is far different. The Vikings yielded a lot of sacks in the previous years because Culpepper holds the ball for a very long time, especially when he had Moss running deep patterns or free-lancing. The Vikings were able to absorb those sacks and still be very effective on offense. This year, the Vikings offensive line has just stunk, no matter if Culpepper was sticking to his old habit, or trying to get rid of the ball quickly, or if Brad Johnson was making pretty rapid decisions. Similar sack totals, yet completely different offensive efficiency.

18
by SJM (not verified) :: Fri, 12/09/2005 - 3:41pm

Why isn't every QB tackle for loss counted as a sack, regardless of whether he was attempting to run? QB runs are so rare that Elias shouldn't bother with sorting them.

19
by Joon (not verified) :: Fri, 12/09/2005 - 3:54pm

vick?

20
by Kite (not verified) :: Fri, 12/09/2005 - 4:07pm

# 8, I think this following website gives Stops For Loss.

http://snap.stats.com/stats/nflinfo/index.asp

21
by Zac (not verified) :: Fri, 12/09/2005 - 4:26pm

Click my name for a link that shows stat corrections made after the fact. This is courtesy of Sideline Software, which is the company responsible for myfantasyleague.com . It's listed so that commissioners can decide whether to use the initial stats, or change to the "real" stats. My favorite is that in Week 7 of this year, a 21-yard reception by Dane Looker was incorrectly scored to Kevin Curtis. It hasn't been updated with this new information yet, but it should sometime today.

Also, the article states that Roye doesn't have a sack incentive in his contract.

22
by bobman (not verified) :: Fri, 12/09/2005 - 4:58pm

Aaron,
I am all for adding hurries and throwaways in the adjusted sacks dept. It will affect my Colts a lot, on both sides of the ball. Why is Manning "never" sacked? The line treats him well and James picks up blitzes well, to be sure, but also, he senses pressure, scans the field, and is okay with throwing it away (finally!) if nobody's free. He gets another down and avoids a 5 yard loss and potential fumble/injury. May even pick up a roughing call (cough cough). So adjusting for hurries and throwaways (which really can be credits to the secondary as much as the DL) will slam his line and stats. So be it.

On the other side of the coin, the Colts' speed rushers, who very often seem to "miss by this much" as Maxwell Smart used to say, will look a lot better, if you can believe it. To my untrained eye, it looks like they have about 4 or 5 legit hurries for every sack in the games I've seen this year.

And in the end, adding these items more accurately reflects their importance to the game. IMHO.

23
by Jim A (not verified) :: Sat, 12/10/2005 - 1:59am

I think it's puzzling that in order to award an "official" sack, the scorekeeper apparently has to judge the original intent of the play, i.e. guess what call was made in the huddle. This seemingly makes sacks even more subjective than tackles, which is not an official stat.

There was a recent discussion of the sack rule on the PFRA forum (linked) in response to another recent change in sack stats. Love the responses by Bob Carroll there!

24
by Andrew (A.B. (not verified) :: Sat, 12/10/2005 - 11:27am

Aaron (#5) -- Just use all tackles for loss. And you can do the same thing for non-QBs too, as part of RB success rate -- or OL success rate.

25
by Dave Gansereit (not verified) :: Mon, 12/19/2005 - 12:26pm

In the Atl-Chi game last night - how could the Vick 2nd interception not be ruled a catch? Jenkins had both feet down, turned and was pummeled - although still having the ball - when he hit the ground - the ball popped up and was caught by Vasher -

I think Vick is getting screwed by the INT and by losing the 20 or so passing yards - What is a football move - he had two feet on the ground and the ball - That was a huge call in the game and in Fantasy and I hope Elias reviews it

26
by Michael David Smith :: Mon, 12/19/2005 - 1:22pm

Elias doesn't review those things. Even if the league were to review it and conclude that it should have been a catch and down by contact, Elias would still rule it the way the officials on the field ruled it.