Writers of Pro Football Prospectus 2008

02 Feb 2010

Stat of the Day: Quarterback Hurries

As part of our ongoing Stat of the Day series, we're digging deep into our spreadsheets to run a new stat every weekday until Super Bowl XLIV. Today, we're dipping into the charting stats to bring you the first look at quarterback Hurries in 2009. If you are a fan of caveats, well, this is your lucky day! Let's list them before we get to the numbers:

  • The whole season is not yet charted. In fact, we're running behind compared to the last couple years, and we only have full seasons for two teams right now, Buffalo and Indianapolis. So all the numbers below have been pro-rated to 16 charted games depending on how many actual games have been charted so far.
  • These have not yet been cleaned up to adjust for certain charters who may have been overzealous (or underzealous) in marking pass pressures.
  • These do not include hurries where the defender causes a holding penalty; we'll add those in when we get to Football Outsiders Almanac 2010.
  • Just so people know, we slightly changed the definition of a Hurry this season to measure defenders more than quarterbacks. In the past, we asked charters not to mark a Hurry if the quarterback still stood tall in the pocket and wasn't rattled. But we realized that we were using these stats to measure defenders more than quarterbacks, and it wasn't fair to effectively "penalize" defenders whose schedules featured more quarterbacks with good pocket presence. So this year, we asked charters to measure the action of the defender more than how the quarterback reacted to that action.

Caveats concluded? Excellent. How about some numbers?

Rank Player Team Hurries*
1 69-J.Allen MIN 34.7
2 94-J.Smith SF 34.5
3 93-A.Ogunleye CHI 32.6
4 93-D.Freeney IND 32.5
5 90-J.Peppers CAR 31.4
6 91-R.Edwards MIN 30.9
7 93-A.Spencer DAL 29.1
8 98-R.Mathis IND 28.5
9 91-R.Geathers CIN 28.1
10 90-M.Williams HOU 27.9
Rank Player Team Hurries*
11 56-L.Woodley PIT 27.7
12 94-D.Ware DAL 25.7
13 55-J.Abraham ATL 25.4
14 92-J.Harrison PIT 25.1
15 75-J.Parker PHI 24.8
16 93-K.Williams MIN 23.5
17 91-W.Smith NO 23.4
18 94-M.Kiwanuka NYG 23.4
19 58-T.Cole PHI 22.7
20 91-T.Hali KC 22.3
*Note: pro-rated to 16 charted games

Posted by: Aaron Schatz on 02 Feb 2010

45 comments, Last at 04 Feb 2010, 2:26pm by Pat (filler)

Comments

1
by Anonymous123451 (not verified) :: Tue, 02/02/2010 - 2:58pm

How many did Dumervil have?

2
by Dean :: Tue, 02/02/2010 - 2:58pm

Again, Juqua Parker is suprisingly high on the list. 8 sacks to go with the hurries, and he was ranked very highly in the run stuffing numbers presented last week.

I'm beginning to wonder if maybe Philly doesn't need a big name DE to go alongside Cole after all?

37
by Sunil (not verified) :: Wed, 02/03/2010 - 11:22am

Juqua Parker is from the TN school of defensive linemen. He was pretty good at TN until he was let go in the cap purge. Notable TN d-linemen who are doing well at other clubs include Antwan Odom (Bengals) Randy Starks (Dolphins - who TMQ has singled out as well), Travis LaBoy (Cardinals). Bottomline - I'm not surprized at the success some of these ex-TN d-linemen are having at other teams. I wonder if FO can run a stat on which schools produce good products (e.g. TN for d-linemen, Ravens for linebackers etc).

44
by Pat (filler) (not verified) :: Thu, 02/04/2010 - 2:20pm

and he was ranked very highly in the run stuffing numbers presented last week.

Yeah, but the yards/rush to his side were far worse than Cole's. When Parker could get off of a block, he didn't allow much gain, but he was blocked away from the action too easily.

Not saying Parker was terrible, mind you, and I actually agree that DE isn't a huge 'need' position. I think Philly's defense mainly needs to fix the whole safety debacle and figure out what the hell they're doing with their linebackers. Good DEs are always hard to find though, so if they want to bring more in, hey, feel free.

3
by widderslainte :: Tue, 02/02/2010 - 3:09pm

Hmm... so what does it mean if a player has a lot of hurries, but not a lot of sacks? Bad luck, quality of the quarterback, something else?

6
by alexbond :: Tue, 02/02/2010 - 3:35pm

Bad secondary since a guy gets open fast and the QB gets the ball out?

9
by Karl Cuba :: Tue, 02/02/2010 - 3:48pm

The other alternative would be a lack of closing speed.

21
by widderslainte :: Tue, 02/02/2010 - 5:27pm

That would certainly explain John Abraham

13
by TXNiner :: Tue, 02/02/2010 - 4:12pm

In the case of a player like Justin Smith, playing DE in a 3-4 alignment, it could mean that they are getting good pressure on the QB, but because of extra blocking on them, that's all they can do is get pressure.

4
by UTchamps (not verified) :: Tue, 02/02/2010 - 3:09pm

Would it be worth it to add a number of games charted column to this chart? Otherwise, the numbers are not very meaningful.

5
by dsouten :: Tue, 02/02/2010 - 3:13pm

They are pro-rated to 16 games for that very reason.

7
by mrh :: Tue, 02/02/2010 - 3:45pm

"Pro-rated" means what he doesn't think it means.

10
by dk240t :: Tue, 02/02/2010 - 4:05pm

The number of games included in the original analysis would give an indication of the error associated with the measurement, and be useful, regardless of pro-rating.

24
by dsouten :: Tue, 02/02/2010 - 5:42pm

True, but they admit that this is a very rough data set. Given that they admit there is probably alot of error, or at least inconsistency, in the chart, an adjustment like that would probably just give the illusion of increased accuracy.

I think we may be overthinking things at this stage.

8
by Karl Cuba :: Tue, 02/02/2010 - 3:47pm

I've felt for a while that perhaps there needs to be a fourth category added to sacks, hits and hurries. I'd call this pressures. This would be where the pass rusher has beaten his block and so made the qb begin to make his throw but has not managed to get close enough to affect the quality of his throw.

The need for something like this has become more apparent this year watching the 49ers. In previous seasons the niner pass rush has had games where the quarterback has been able to stand in a state of utter serenity as no lineman has broken their block for up to 6 or 7 seconds. This year I've noticed that generally (and often it has been Justin Smith) the pass rush has been able to put a clock on the qb even when they have gotten close enough to achieve a hurry.

14
by TXNiner :: Tue, 02/02/2010 - 4:13pm

I think what you are calling a pressure is counted as a hurry.

19
by Karl Cuba :: Tue, 02/02/2010 - 5:17pm

No, it isn't. I'll try to explain what I mean again. For me a hurry should be logged when the rusher gets close enough to the qb to result in a poor throw or bad decision due to the qb not being able to step up, throw with a normal motion or his being flushed away from his desire release point. The pressure would simply be the DL breaking his block and having a clear path to the passer, preventing the qb from holding the ball indefininitely. The pass rusher might not be particularly close to the quarterback and will not impede his delivery of the ball but does prevent him from waiting even longer for a receiver to completely shake his coverage or get open deeper.

It may be that you don't agree that this stat is worthwhile but I hope that I've made it a little clearer that I see the difference between the two stats. If it helps I'm pretty sure that the 49ers at least complile this stat, as for the past two seasons they have said that Justin Smith has collected 70 or so pressures when he clearly hasn't had that many hurries.

22
by dsouten :: Tue, 02/02/2010 - 5:40pm

Last year's definition of a Hurry (according to the instructions given to charters) matches your definition of Hurry above - it impeded the throw.

This year they revised the definition of Hurry to roughly match the way you describe a Pressure.

Personally, I think the change just made the charting much more subjective.

38
by Aaron Schatz :: Wed, 02/03/2010 - 12:51pm

We ended up with some charters doing it one way, and some charters the other. So we had to re-write the rules to try to create more consistency.

A lot of these stats are going to be imperfect because of the fact that we're using volunteers and TV camera angles. And I want to keep away from adding more categories to things when we can avoid it. Just makes things more complicated, which leads to charters dropping out of the project.

30
by Dr. Mooch :: Wed, 02/03/2010 - 12:55am

Seems like every screen play would count as a pressure for at least one defender.

11
by dk240t :: Tue, 02/02/2010 - 4:06pm

I wonder, why even bother putting together this article with all those caveats, rather than waiting till the actual data is available?

12
by Tim (not verified) :: Tue, 02/02/2010 - 4:07pm

Correct me if I'm wrong, but isn't KWilliams the only 4-3 DT on this list?

And 3 vikes in the top 20? Yeesh.

15
by TXNiner :: Tue, 02/02/2010 - 4:18pm

You are correct sir, the only DT at all actually. And Justin Smith, while we're at it, is the only 3-4 DE on the list.

16
by El Miriodor :: Tue, 02/02/2010 - 4:37pm

Not only that, but the Vikings made a significant push to sign J. Smith before the 49ers got their paws on him. The Vikes are doing a pretty good job evaluating defensive line talent it seems. (Though it is possible that having good players has a "bleed" effect where lesser players put up better stats due to having to fight through less chips and double teams.)

18
by jmaron :: Tue, 02/02/2010 - 4:45pm

Certainly seems to be their strength. They landed R. Edwards in the 4th round and if B. Robison who they also took in the 4th round got a starting job I'm pretty sure he'd put up some good numbers.

17
by Brendan Scolari :: Tue, 02/02/2010 - 4:41pm

I hope this doesn't sound too much like complaining because I enjoy interesting stats like this as much as anyone, and I have mentioned Pro Football Focus a number of times in these comment threads in recent weeks, but I figured it was worth noting. Now that PF is releasing all this data for every player and every game, doesn't it seem a bit silly to use some of these stats for articles like this?

What I mean is that this article shows a list of the top 20 leaders in hurries, with no other context. But at PFF you can look at the # of hurries for every NFL player, and sort through the leaders both by total hurries, and by most hurries for each position (4-3 DE, 3-4 DE, 3-4 OLB, etc.) They've also got sacks and QB hits for every player and another huge piece your missing, the number of snaps each player played. And not only that, but they tell you how many of those snaps that player rushed the passer, how often he played run defense, and ohow often he dropped into coverage. Basically this list is pointless, you only list the top 20 players with no QB hits or sacks referenced, no snap counts, and no division of snap counts.

It'd be like if Baseball Prospectus published an article at the end of the year showing the top 20 hitters in on-base percentage. Yes, that information is interesting, but you could instead just go to Fangraphs and not only see the on-base percentage of every player, but each player's wOBA, wRC, WAR, line drive %, UZR, how often he swings at balls out of the strike zone, and how he does against curveballs. There's so much more depth in the latter that I'm not sure why you'd even bother with the former.

I'm not saying you shouldn't run any stats articles like this, but perhaps you should focus on stats where we don't already have so much more information than what FO's game charting seems able to provide (at this point anyway). Otherwise I think you should focus your attention elsewhere (since obviously FO has done lots of insightful analysis in areas that sites like PFF do not), but of course that's just my opinion.

31
by Temo :: Wed, 02/03/2010 - 1:06am

I hear in one aspect, but I find PFF's charting to be just as iffy as FO's, if not moreso. I like to get both if I can to get a better view of what's going on.

That said, it's a shame that PFF, a free service, is putting out timely charting at a much faster pace than FO.

40
by Brendan Scolari :: Wed, 02/03/2010 - 1:39pm

Yes you're right, it is good to have more info to verify, but at the same time there's no way to verify the FO game charting numbers other than just taking them for their word. At least the PFP has the positional play-by-play available so i have something I can check when I go back to look at DVR'd games.

45
by Pat (filler) (not verified) :: Thu, 02/04/2010 - 2:26pm

Uh... try again. The detailed game charting data's available, and a few weeks ago they posted a portion of game charting as an example.

Yeah, it costs some money, but it is pretty straightforward to verify.

20
by Danish Denver-Fan :: Tue, 02/02/2010 - 5:20pm

I'm not at all surprised to discover Dumervills abscence - coverage makes or breaks a pass rushers season.

23
by Dan :: Tue, 02/02/2010 - 5:41pm

For comparison, here is the PFF leaderboard (for the stat that they call "QB Pressures")

DeMarcus Ware DAL 56
Justin Smith SF 44
Dwight Freeney IND 41
John Abraham ATL 39
Robert Mathis IND 37
Tamba Hali KC 36
Jared Allen MIN 35
Ray Edwards MIN 35
James Harrison PIT 34
Julius Peppers CAR 33
Tony Brown TEN 33
Trent Cole PHI 31
Lamarr Woodley PIT 31
Elvis Dumervil DEN 31
Andre Carter WAS 30
Antonio D. Smith HST 30
Mario Williams HST 29
Calais Campbell ARZ 29
Parys Haralson SF 27
Cullen Jenkins GB 27
Mathias Kiwanuka NYG 26
William Hayes TEN 26
Kyle Vanden Bosch TEN 26
Adewale Ogunleye CHI 25
Clay Matthews GB 25
Jonathan Babineaux ATL 25

28
by tuluse :: Tue, 02/02/2010 - 9:30pm

Is this just pressures, or pressures + hits + sacks?

33
by Dan :: Wed, 02/03/2010 - 2:22am

Just pressures. They also had an article that looked at pass rushing productivity which included all three (divided by number of plays rushing the passer).

35
by tuluse :: Wed, 02/03/2010 - 2:45am

I wonder about the wildly different numbers then. From Ware's numbers, he must have been pressuring, hitting, or sacking the opposing QB on better than 1 in 10 pass attempts. Actually, with 85 hits, pressures, and sacks, if they faced about 600 pass attempts on defense it's 1 in 7.

41
by Brendan Scolari :: Wed, 02/03/2010 - 1:41pm

At least according to PFF Ware rushed the passer 614 times so your numbers seem about right.

42
by Dan :: Wed, 02/03/2010 - 1:53pm

554 plays where he rushed the passer, actually (once you subtract out the playoff numbers), which means one hit/pressure/sack every 6.5 attempts. And that only places him third, behind Freeney (67 in 393 plays, one every 5.9 att) and Woodley (60 in 391 snaps, one every 6.5 att).

29
by Chip :: Wed, 02/03/2010 - 12:04am

Simply blinking this list... it seems to make more sense. Maybe I'm hung up with Ogunleye at #3 on the FO list. Having watched every Bears game, there is no way I can believe that number.

34
by tuluse :: Wed, 02/03/2010 - 2:40am

Yeah, I'm having a hard time with Ogun too. I thought he had a very down year, more so in run defense, but still worse in pass rushing than he has been.

But I could see a case where all the Bears were handled easily and Ogun happened to finally beat his guy 5 seconds after the snap.

39
by Brian :: Wed, 02/03/2010 - 12:59pm

I think his numbers make more sense if you keep in mind that these totals are pro-rated, because he was having more success rushing the passer earlier in the year, if memory serves.

That being said, #3 in the NFL seems absurd. Not that he was awful, but he wasn't THAT effective.

25
by jmaron :: Tue, 02/02/2010 - 6:47pm

something tells me that the person charting the Dallas games for PFF has a crush on D. Ware.

26
by JMM* (not verified) :: Tue, 02/02/2010 - 8:34pm

Aaron, the precision of your table, three significant figures, overwelms your caveats.

27
by Raiderjoe :: Tue, 02/02/2010 - 9:04pm

ol at chiefs not holding onto jard allen. of coruse was jerk there but still funny KC let go of good player

32
by Anonymous6789 (not verified) :: Wed, 02/03/2010 - 2:12am

1st and 2 3rds isnt letting a guy go.

36
by FooBarFooFoo (not verified) :: Wed, 02/03/2010 - 3:23am

"But we realized that we were using these stats to measure defenders more than quarterbacks, and it wasn't fair to effectively "penalize" defenders whose schedules featured more quarterbacks with good pocket presence"

I don't get that ... If a hurry has (little to) no effect on the QB, then is it really a "hurry"?. I mean, I think I understand the reasoning that one defender who does the "same thing" compared to another defender, yet one causes the QB to panic while the other QB completes a TD pass, that makes no sense.

Or, just in case, how do you include the blocking performance by the OL in these cases? One defender performs against a bina fide OL, while the other plays opposite to a human turnstile. You don't adjust the hurries against this, so you should not adjust to the QB.

43
by Q (not verified) :: Thu, 02/04/2010 - 1:05am

I suspect facing GB twice a year helps explain Ogun and some of the Vikings on this list