Writers of Pro Football Prospectus 2008

23 Jul 2007

Adjusting Completion Percentage

by Aaron Schatz

Pro Football Prospectus 2007 officially hits stores today, and it is packed to the gills with information. So packed, in fact, that we couldn't fit in everything we wanted from the Football Outsiders game charting project. You'll find quarterback hits and hurries in the book, and stats that measure defensive backs in coverage, and YAC and pass lengths for both quarterbacks and receivers. You'll find a big long article on when play-action works best, and information on how often teams blitzed or lined up in various formations. Despite all that good stuff, we've barely scratched the surface with what we can do with the game charting information.

One thing we originally planned was a page in the quarterbacks section which would give a variety of tables showing which quarterbacks had the most or fewest incomplete passes of different types: underthrown, overthrown, dropped, thrown away, defensed, and so on. It didn't fit in, so we figured we would look at that information over the next couple of days.

(I should point out at this point that Bill Moore didn't just spend hours coordinating the whole game charting project. He also put together a lot of these tables.)

The first thing we wanted to look at was dropped passes. Dropped passes is one of those stats that we all know is out there, but it isn't easy to find. STATS Inc. keeps it, so you'll find it on some individual player pages on websites where STATS provides the numbers, but not all the individual player pages, and you don't know what kind of passes were dropped -- are we talking long bombs here, or running backs who couldn't get the handle on screen passes?

We have our own count of dropped passes, thanks to the game charting project. By our count, Miami led the league with 46 drops, followed by Seattle with 43, Philadelphia with 40, New Orleans with 39, and Green Bay with 39. Buffalo had the fewest charted drops, just 12, a bit of a surprise since they picked up drop machine Robert Royal before the 2006 season. The book goes into further detail about which players had the most drops, but today we'll look at what drops mean for quarterbacks.

Not counting spikes to stop the clock, last year NFL quarterbacks completed a little over 60 percent of passes. If you were to consider drops as completed passes, completion percentage goes up to about 66 percent. Here's a look at the quarterbacks with the highest completion percentages, with and without including drops as completions. The standard completion percentage will be different from the NFL numbers because it includes only charted passes, without spikes or plays missing from our charting database. Minimum to be ranked is 100 pass attempts.

Highest Completion Percentage
(charted passes only)
  Highest Completion Percentage
(includes drops as complete)
Rk Player Cmp%   Rk Player Cmp%
1 8-D.Carr 68.6% 1 18-P.Manning 71.3%
2 18-P.Manning 65.5% 2 9-D.Brees 70.4%
3 10-C.Pennington 65.3% 3 8-D.Carr 69.9%
4 9-T.Romo 65.3% 4 9-T.Romo 69.5%
5 9-D.Brees 64.8% 5 10-M.Bulger 69.2%
6 9-C.Frye 64.5% 6 12-T.Brady 68.9%
7 13-K.Warner 64.3% 7 10-C.Pennington 68.7%
8 9-S.McNair 63.9% 8 9-C.Frye 68.6%
9 10-M.Bulger 63.8% 9 9-D.Garrard 68.3%
10 9-C.Palmer 63.0% 10 8-J.Kitna 68.1%

Hey, what a surprise, Peyton Manning is the best in the league at something. Counting drops moves up the completion percentage for Tom Brady and, surprisingly, David Garrard. We charted 33 drops by Jacksonville receivers, and 21 of those had Garrard at quarterback, even though the Week 17 game with Kansas City is one of the handful of games that was not charted.

Here are the guys at the bottom of the league:

Lowest Completion Percentage
(charted passes only)
  Lowest Completion Percentage
(includes drops as complete)
Rk Player Cmp%     Rk Player Cmp%
45 7-M.Vick 52.5% 45 17-J.Campbell 54.5%
44 10-V.Young 52.7% 44 2-C.Simms 55.3%
43 11-D.Bledsoe 52.9% 43 8-R.Grossman 57.6%
42 17-J.Campbell 53.4% 42 11-D.Bledsoe 58.7%
41 16-A.Walter 53.6% 41 7-M.Vick 59.4%
40 7-B.Gradkowski 53.9% 40 10-V.Young 60.3%
39 2-C.Simms 54.3% 39 7-B.Gradkowski 60.5%
38 8-R.Grossman 54.9% 38 17-J.Delhomme 61.7%
37 16-J.Plummer 55.2% 37 16-A.Walter 61.9%
36 4-B.Favre 56.2% 36 11-A.Smith 62.4%

Well, I guess Michael Vick has a little bit of an excuse; adding drops as complete passes means he no longer has the lowest completion percentage in the league. Vince Young also moves up a bit.

Of course, the top quarterbacks in completion percentage are not necessarily the most accurate quarterbacks in the league. We know that David Carr and Charlie Frye had high completion percentages because they were throwing a ton of screens and dumpoffs. What happens if we adjust completion percentage based on the types of throws each quarterback was making?

In the book, we separate passes into four categories. I took each category and determined the league-average completion percentage, including drops as complete passes:

  • Short (5 yards or less): 80.5 percent
  • Mid (6-15 yards): 65.0 percent
  • Deep (16-25 yards): 52.0 percent
  • Bomb (26+ yards): 33.6 percent

I then adjusted each quarterback's completion percentage based on how his completion percentage on passes of each distance compared to the league average, and added back in all the passes that don't fall into these categories: balls thrown away, tipped at the line, and so forth. Once again, I included dropped passes as completions.

Here are the best and worst quarterbacks from 2006:

Top 10 Adjusted Completion Percentage Bottom 10 Adjusted Completion Percentage
Rk Player AdjComp    Rk Player AdjComp
1 9-D.Brees 80.3% 45 17-J.Campbell 47.6%
2 9-T.Romo 78.8% 44 8-R.Grossman 52.6%
3 18-P.Manning 78.8% 43 10-V.Young 55.8%
4 8-D.Carr 75.8% 42 2-C.Simms 56.2%
5 9-D.Garrard 73.5% 41 7-B.Gradkowski 56.2%
6 13-K.Warner 71.9% 40 7-M.Vick 56.4%
7 9-C.Frye 71.3% 39 4-B.Favre 57.2%
8 10-M.Bulger 70.6% 38 11-D.Bledsoe 58.7%
9 8-J.Kitna 69.7% 37 16-A.Walter 58.8%
10 9-C.Palmer 69.3% 36 16-J.Plummer 59.2%

Surprise! David Carr and Charlie Frye remain in the top ten. Carr and Frye didn't just throw a high number of short passes; they also threw very accurate short passes. Frye completed 89.8 percent of his pass attempts of five yards or less, the highest percentage in the league. Carr was second at 87.2 percent. Then again, maybe Carr and Frye weren't just throwing short passes -- they were throwing really short passes, stuff behind the line, or dumpoffs to uncovered receivers on third-and-10. What if we take out the short stuff?

Top 10 Adjusted Completion Percentage
(not including short passes)

Bottom 10 Adjusted Completion Percentage
(not including short passes)

Rk Player AdjComp    Rk Player AdjComp
1 13-T.Rattay 80.6% 45 4-B.Favre 37.2%
2 18-P.Manning 76.5% 44 7-B.Gradkowski 38.1%
3 9-D.Brees 73.7% 43 8-D.Culpepper 38.7%
4 9-T.Romo 71.9% 42 17-J.Campbell 40.0%
5 11-D.Huard 65.0% 41 7-B.Leftwich 41.0%
6 13-K.Warner 63.8% 40 2-C.Simms 41.3%
7 8-M.Hasselbeck 62.6% 39 8-R.Grossman 41.9%
8 15-S.Wallace 61.5% 38 11-A.Smith 43.4%
9 10-C.Pennington 60.5% 37 16-J.Plummer 45.2%
10 17-P.Rivers 60.4% 36 7-M.Vick 45.9%

OK, that looks a bit more like a table of the league's most accurate quarterbacks. Tim Rattay is number one, but that's partially due to small sample size (he barely gets over our 100 attempt minimum) and partially due to the nature of Tim Rattay. Rattay's problem has never been that he can't play, it's that he can't play without getting hurt. The two Seattle quarterbacks are a bit of a surprise, but Seattle had a big problem with dropped passes last year (what else is new) and both Hasselbeck and Wallace were quite accurate when it comes to midrange throws of 6-15 yards. That makes a lot of sense in the Mike Holmgren offense. (This is nothing compared to the other huge surprise about Seneca Wallace, but you'll have to buy the book for that one.)

Way down at the bottom of the list, however -- is that really Brett Favre? Yes, it is. Favre was terrible last year when it came to completing long-distance throws. His completion percentage was just 34.4 percent on passes of 16-25 yards, and just 20.4 percent on passes over 25 yards. It makes sense that an older Favre may have lost arm strength and the ability to be accurate when throwing deep. The Packers may want to dial down on those long passes and put more midrange throws into the playbook to take advantage of Favre's remaining strengths and hide his weaknesses.

Perhaps the most interesting player when it comes to this "adjusted completion percentage" statistic is Tom Brady. When we include all passes, Brady has an adjusted completion percentage of 69.2 percent, just outside the top ten. When we take out the short passes, Brady drops to 22nd, with an adjusted completion percentage of 54.1 percent.

Why the drop? Of the league's top quarterbacks, the only two threw more than 50 percent of their passes below six yards: Tom Brady and Drew Brees. When we take out short passes, the quarterbacks whose completion percentages are most dependent on short passes will suffer the most. Brady's completion percentage on passes longer than five yards isn't much different than the league average. You can see that taking out short passes drops Brees in "adjusted completion percentage" as well, but not by anywhere near as much because Drew Brees actually completed more than 65 percent of all passes over 15 yards last year. That includes more than 65 percent of all Bomb passes, passes over 25 yards. Tony Romo was the only other quarterback to complete more than 50 percent of Bomb passes in 2006. Not bad for a guy who was criticized for his arm strength coming out of Purdue.

In Brady's defense -- and perhaps in defense of Carr and Frye as well -- there is a skill to completing short passes, just as there is a skill to completing long passes. Brady had a DVOA of 30.0% on short passes, which ranked fourth in the league behind Trent Green, Carson Palmer, and Marc Bulger. Perhaps nobody in the league has better timing on when to throw a screen pass than Tom Brady. However, if you are expecting lots of long bombs to Randy Moss and Donte' Stallworth this year, you may be disappointed.

(By the way, David Carr was fifth on short passes with a DVOA of 29.7%, but Charlie Frye was near the bottom of the league with a DVOA of 3.0% on short passes. Nearly every quarterback has a positive DVOA on short passes because only measuring passes where we can record a pass length automatically leaves out sacks, throw aways, and passes tipped at the line.)

Wednesday, I'll look at which quarterbacks had the biggest accuracy problems (underthrown and overthrown passes) and which quarterbacks suffered the most due to the pass rush (thrown away, tipped at line, and hit in motion passes).

Posted by: Aaron Schatz on 23 Jul 2007

84 comments, Last at 20 Aug 2007, 12:26pm by Soxphan

Comments

1
by Ilanin (not verified) :: Mon, 07/23/2007 - 5:07pm

...and this didn't make PFP 2007.

Wow. Can't wait to see what you did have space for.

2
by Sergio (not verified) :: Mon, 07/23/2007 - 5:15pm

Fascinating stuff.

Does the Brady bit means that the supposed gain that Moss and Stallworth bring to the team (that being their vertical game) is somewhat negated? Or have these guys shown to be more than deep threats? I'm really asking, I'm not knowledgable on them.

Elsewhere, Miami is dead last in drops. I wondered for a second if that had to do more with me than with them, but after thinking about it I don't feel I've unjustly assessed drops. I'll go back to the tape and check, just in case...

3
by Rich Conley (not verified) :: Mon, 07/23/2007 - 5:22pm

"Perhaps the most interesting player when it comes to this “adjusted completion percentage� statistic is Tom Brady. When we include all passes, Brady has an adjusted completion percentage of 69.2 percent, just outside the top ten. When we take out the short passes, Brady drops to 22nd, with an adjusted completion percentage of 54.1 percent."

What the hell have I been saying for 3 years?

That being said, take into account that he was throwing deep balls to Reche Caldwell, Jabar Gaffney, and Troy Brown.

4
by dryheat (not verified) :: Mon, 07/23/2007 - 5:24pm

Carr and Frye didn’t just throw a high number of short passes; they also threw very accurate short passes. Frye completed 89.8 percent of his pass attempts of five yards or less, the highest percentage in the league. Carr was second at 87.2 percent.

That and 41 cents will get you a stamp and a job as a backup.

5
by Rich Conley (not verified) :: Mon, 07/23/2007 - 5:25pm

"Does the Brady bit means that the supposed gain that Moss and Stallworth bring to the team (that being their vertical game) is somewhat negated?"

It could mean that, and it could mean the opposite.

We really can't know until we see what Brady does with Moss and Stallworth. If the numbers get better, the WRs were the problem, if they get worse, Brady is the problem. So its either a case of them fixing something that was broken, or them trying to improve an area that is dependant on something that is broken.

I dont think theres really any way to tell though.

Stallworth, I dont know much about, but calling Moss just a deep threat isnt really accurate.

6
by pawnking (not verified) :: Mon, 07/23/2007 - 5:29pm

Quick theories:
1) I bet the longer the yardage, the less predictable past completion % will prove to be
2) We'll see those players with very VORP for short completion % age better than those who don't
3) If and when adjusted completion % is incorporated to the college stat projections, it will improve the predictive power of the current system.

7
by James C (not verified) :: Mon, 07/23/2007 - 5:30pm

Great article. Really looking forward to the book.

8
by dryheat (not verified) :: Mon, 07/23/2007 - 5:31pm

It's tough to get a read on Brady's medium-to-long-arm from looking at 2006. His shoulder definitely was bothering him, but more damning is that Caldwell, Gabriel, and friends couldn't get any separation on medium-to-long patterns, so any throw against man coverage was fairly low percentage.

I can't say I'm surprised at any rate, because Belichick seems to prefer the strategy of getting the ball into the receiver's hands as quickly as possible and let them make a play, as opposed to letting the receiver make a play to get open and then get the ball. Some of that may be explained away by personnel, but, with the notable exception of Bethel Johnson, the WRs Belichick drafts seems to fit that profile.

9
by Sergio (not verified) :: Mon, 07/23/2007 - 5:35pm

re: 6

Theory 3 would involve charting every single college game...

10
by MJK (not verified) :: Mon, 07/23/2007 - 5:37pm

Of course, the tendency is to consider "dropped passes" as the reciever's fault, and any other type of incompletion as the QB's fault, which is not exactly the case. For example, a dropped pass may be because the QB threw it a lot harder than he needed to and the WR couldn't handle the ball without "touch" on it (hello, Mike Vick and Brett Farve). This makes sense, as both Brady and Brees are known for having good "touch".

The other side of the coin is that not all passes thrown 8 feet over the WR's head or ten yards short are necessarily the QB's fault. We all know how important good route running, and how important the WR's and the QB's making the same read, are. For example, a percentage of Brady's lack of production on medium range passes may have been due to Caldwell and Gaffney being slow to pick up the offense and running the right route. It would be interesting to run these numbers for the '04 and '05 Patriots, to see if Brady is a career "average" mid-to-long range passer, of if it was due to having inexperienced WR's last year...

I agree with the sentiment that Brady's long ball has never been anything worth getting excited over. I don't think most Patriots fans would cite a cannon arm as one of Brady's strengths--they lie more along the lines of pocket presence and short-ball accuracy. But I'm not sure that you can necessarily conclude that long balls to Moss and Stallworth won't be a factor. You could only say that if Brady had had a credible deep threat at one point and had been unable to exploit it, but I'm racking my brains and having trouble coming up with a viable deep threat WR he's had... I don't think Bethel Johnson, Tim Dwight, or Jabar Gaffney count...

11
by Rich Conley (not verified) :: Mon, 07/23/2007 - 5:46pm

"don’t think most Patriots fans would cite a cannon arm as one of Brady’s strengths–they lie more along the lines of pocket presence and short-ball accuracy."

You obviously dont have the priveledge of living in Boston, do you MJK.

I consider myself slightly homerish, but I'm often amazed by what I hear people saying around here.

12
by TuckerF (not verified) :: Mon, 07/23/2007 - 6:00pm

great read, fascinating statistics, too bad you can't get facts correct...
"Tony Romo was the only other quarterback to complete more than 50 percent of Bomb passes in 2006. Not bad for a guy who was criticized for his arm strength coming out of Purdue."

Tony "Romo criticised for arm strength coming out of Purdue?" Wasn't that Northern Illinois? OR Southern, but then I 'm not editing a major footbal website and publishing house it sort of does throw a questionable light on your conclusions.

13
by Rich Conley (not verified) :: Mon, 07/23/2007 - 6:04pm

12../

Eastern Illinois.

If you're going to nitpick, atleast do some research.

Whats interesting about the Romo thing, is he was criticized about not having the arm, and has two of the best deep threats in the league. Maybe WR makes more of a difference than the QB in bombs?

14
by Tom (not verified) :: Mon, 07/23/2007 - 6:04pm

So Grossman needs to work on his short passes. Watching games, I always felt he needs to take his check down more often, I think this supports that.

15
by dbt (not verified) :: Mon, 07/23/2007 - 6:06pm

don't look now, but i think the other three people still with me on the rex grossman bandwagon just jumped off too.

16
by CA (not verified) :: Mon, 07/23/2007 - 6:08pm

Re: 12, 13

Aaron is referring to Brees.

17
by Harris (not verified) :: Mon, 07/23/2007 - 6:12pm

#12: I think it was just poor sentence structure because Brees was criticized for having a week arm coming out of Purdue.

18
by Rich Conley (not verified) :: Mon, 07/23/2007 - 6:14pm

14.

My oppinion as to why grossman checks down so little, is that hes really bad at it. I didnt see a lot of bears football last year, but I remember the NE/Chi game, where he basically tried to check down to the TE, only Asante Samuel was standing right in front of the TE, so he chucked the ball right into the DB. I've seen him bounce balls off of recievers backs too...

19
by Fergasun (not verified) :: Mon, 07/23/2007 - 6:18pm

So this means Wednesday you're going to be shining the spotlight on Jason Campbell? Just when I've gotten used to hearing all the positive news in the Washington press about him too.

Sigh... at least we've got reigning World Bowl MVP Casey Bramlett on the team (at least until late August).

Incidentally I just watched WAS-NYG during Week 17, and did notice that although Campbell has poise and throws a very nice pass, it often was off the mark. Collingsworth pointed out that he had been 3rd on the depth chart until half-way through the year... so at least there's something to temper my negativity.

20
by Sergio (not verified) :: Mon, 07/23/2007 - 6:18pm

re: 12, 17

Hilarious.

21
by Kaveman (not verified) :: Mon, 07/23/2007 - 6:19pm

#12: That's a badly ordered paragraph; I think he's talking about Drew Brees. In fact, more kudos to Brees... he did have a rotator cuff injury even.

Thanks for this fascinating data, FO. One number that interests me is Jake Plummer's completion percentage. Here, at last, is some hard statistical evidence that Shanahan wasn't loopy when he benched the guy for the rookie. This looks like Jake just sucked, last year.

22
by Joseph (not verified) :: Mon, 07/23/2007 - 6:20pm

Re: 12 & 13--He's referring to Drew Brees. With a mom as a grammar teacher, the paragraph sentence order needs changed.
RE: the article--if you want to understand the BIGGEST reason for the Saint's 2006 turnaround and the reason for optimism in the future, this article spells it out. 80.3% adj. comp.!!!! And over 65% on the long stuff--highest in the league by a long shot. (Compare this with Aaron Brooks) Yes Reggie Bush, a healthy Deuce, and Colston all help. But those numbers are out of this world--even better the Manning and Brady. Look out for the Saints O in '07!

23
by Rex Grossman (not verified) :: Mon, 07/23/2007 - 6:24pm

F**k it, I'm still going deep. I don't care what your stats say, only a p***y throws short passes.

24
by admin :: Mon, 07/23/2007 - 6:40pm

Sigh. This is what I get for slapping something together in between radio interviews.

25
by Pat (not verified) :: Mon, 07/23/2007 - 7:10pm

#22: With a mom as a grammar teacher, the paragraph sentence order needs changed.

I'm beginning to think no one should ever criticize grammar, because there's always a mistake made in the criticism. "Needs changed" should be "needs to be changed," unless you live in Appalachia or Scotland. Or "needs changing," although I've always thought that sounded horrible.

26
by bravehoptoad (not verified) :: Mon, 07/23/2007 - 7:16pm

No, Aaron--amazing article. This is the kind of stuff that changes the way we look at football. I love the PFP, but still miss the days when articles like this popped up on FO all the time. . It's too bad books make more money than websites.

27
by Pat (not verified) :: Mon, 07/23/2007 - 7:30pm

Incidentally, regarding Brady: how much of a difference is it really going to make on long passes between Caldwell, Gaffney, etc. and Stallworth, Moss?

As bizarre as this sounds, think about it: Brady's not an idiot. He's not throwing those passes unless they've got some separation, and drops are already excluded. If you exclude drops, there's only really overthrown, underthrown, or pass defensed - so unless virtually all of Brady's missed deep passes are "overthrown," it's hard to imagine that getting Stallworth/Moss there will make things any easier.

There was a great post by a scout regarding arm strength and deep passes previously: a weak arm doesn't mean you can't make deep passes. It just means two things: first, they're going to be more variable, since you're throwing near the limit of your arm. The second point involves physics - a QB with arm strength to spare doesn't have to pass the ball at a near-45 degree angle to get the distance. He can throw it at a sub-45 degree angle, and it will get there, faster, which means the DBs deep will have less time to react.

If anything, between Moss and Stallworth, I'd imagine Moss would be the bigger impact. Stallworth's pretty normal height - he's not towering over any DBs, that's for sure. His biggest strength is his speed. If the ball can't keep up with him, that's wasted. Moss, however, is tall, and so even if the DB gets there, Moss still has a decent chance to come down with it fighting for the ball.

28
by Fergasun (not verified) :: Mon, 07/23/2007 - 7:34pm

Re: 26
Don't you mean, it's good that books can help them pay for all the bills so the Outsiders can have a website?

I'd probably pay a decent penny to have all of FO archived content in book form. In fact, I think that is possible in more-so than any other blog I read since all the content here is original.

Not only that, but there is a ton of statistical/historical value in the PFP books.

29
by Oswlek (not verified) :: Mon, 07/23/2007 - 7:39pm

#18 Rich,

I recall that play as a WR slant, not a checkdown to a TE. And Grossman is the posterchild for just how bad a QB can look if he *isn't* good at short passes.

Regarding Brady, Joyner wrote an article last year or the year before praising Brady's deep ball. It seems pretty clear why NE was so skewed to short passes, particularly when you consider that the guy who ended up being the #1 had several games with an a YPC under 5.

30
by admin :: Mon, 07/23/2007 - 7:42pm

By the way, I wasn't complaining about the criticism, I was complaining about my poor sentence order... bad editor, bad.

31
by Gerry (not verified) :: Mon, 07/23/2007 - 8:00pm

Wow, not a single appearance on either list by Manning of the Eli variety.

Guess that spells med-i-o-cre.

32
by Theo, Netherlands (not verified) :: Mon, 07/23/2007 - 8:24pm

Madden Ratings:

Drew Brees
THP ================== 95
THA ==================== 99

Peyton Manning
THP =================== 97
THA =================== 97

Tony Romo
THP =============== 92
THA ================ 98

33
by MJK (not verified) :: Mon, 07/23/2007 - 8:36pm

Re 11:
You obviously dont have the priveledge of living in Boston, do you MJK.

I consider myself slightly homerish, but I’m often amazed by what I hear people saying around here.

Actually, I'm from New Hampshire, and lived in Boston for 10 years, from 1996 to 2006. I'm the biggest Patriots fan of any person I know, barring one friend (who named her cat "Brady"). And I watch a lot of Patriots and a lot of Brady (the QB, that is, not the cat). I was leading the "sit Bledsoe and let Brady play" cheer to my friends in 2001, most of whom were calling for Bledsoe.

But be honest. Brady's deep ball is not one of his strengths, just as his footspeed is not. That doesn't mean he can't throw a deep ball--I remember the 86 yd walkoff OT TD against Miami a few years ago well, and the deep strike to Bethel Johnson a couple of years ago that saved an important game (don't recall exactly which one at the moment...). But that's what this article is saying. Brady isn't BAD at the deep ball...but he's not great at it either. He's average. He's well above average at many other, and probably more important, QB skills, and hence he's one of the top QB's in the game. But no QB is good any everything. Now whether he merely average deep ball is because of his inherent QB skillset, or because of the WR's he's had, or because of the scheme the Pats run, I don't know. I'm pretty sure scheme has something to do with it--wouldn't you call a lot of short passes, especially if your running game was as average to bad as it has been every year since Curtis Martin left except for 2004, and if you had someone like Brady who was exceptional at the short stuff and pretty decent at the medium stuff?

To clarify, when I said "arm strength" I actually meant "deep ball". As Pat pointed out, they're not the same thing--arm strength affects speed more than accuracy of the deep ball. I was succumbing (sp?) to the sportswriter laziness and not being precise.

As to Pat's other comment about how, unless all the balls were overthrown: I assert that, by my admittedly subjective observation, most of Brady's deep ball misses last year that I remember seeing WERE overthrown. Brady is no fool. He knows if you're going to miss a deep ball and there's no safety over the top, miss by erring on the deep side so there's no chance of an INT. I saw any number of deep balls sailing over an open Reche Caldwell or Jabar Gaffney. Now I'm not saying that Moss or Stallworth would have got to those...but it's possible. I'm not expecting the Pats to go to a "fling it deep" offense, but I do expect to see a couple more deep completions than last year...

34
by Jeremiah (not verified) :: Mon, 07/23/2007 - 10:19pm

re: 32

What do THP and THA stand for?

35
by Tom (not verified) :: Mon, 07/23/2007 - 10:45pm

Throw Power and Throw Accuracy.

36
by CA (not verified) :: Mon, 07/23/2007 - 10:59pm

Great article, Aaron. Really interesting stuff. I disagree with your implication that Favre's poor downfield passing is a consequence of aging. I suspect that a similar analysis of all of Favre's seasons would show that he has a much lower ACP (not including short passes) than his overall, unadjusted completion percentage and reputation might indicate. Favre has built a record-setting career as a dink-dink-dink WCO QB. I don't think he ever has been a particularly good downfield passer, despite his gunslinger reputation and strong arm, and in recent years he has been absolutely terrible in this area. Unfortunately, the data to confirm or refute my contention is unavailable.

37
by Will Allen (not verified) :: Mon, 07/23/2007 - 11:58pm

Maybe I missed it, but what is the delineation between a deliberate throw-away, and an unintentional overthrow? I respect the effort here, and there is a lot of good stuff, but there seems to me to be a fair degree of mind-reading of quarterbacks going on, which is a bit problematic.

38
by Brett Favre (not verified) :: Tue, 07/24/2007 - 12:17am

F**k it, I’m still going deep. I don’t care what your stats say, only a p***y throws short passes.

:: Rex Grossman — 7/23/2007 @ 5:24 pm

That's the spirit, kid! Don't let some stat geeks or coaches stop you from being a gunslinger. I know I won't.

39
by Catfish (not verified) :: Tue, 07/24/2007 - 12:24am

I was actually just thinking the other day of how it would be useful to have some sort of adjusted completion percentage. Props to FO for reading my mind.

40
by Fnor (not verified) :: Tue, 07/24/2007 - 12:46am

15: If it makes you feel any better, dbt, if Grossman does actually get his stuff together, you'll be the first one in line!

41
by EnglishBob (not verified) :: Tue, 07/24/2007 - 4:03am

I have suddenly become a lot more sullen over Washington's prospects this season. I wonder how accurate Randel El and Portis are?

42
by Karl Cuba (not verified) :: Tue, 07/24/2007 - 9:24am

Fantastic article but this doesn't reduce the quarterback's dependence on the abilities of his recievers. It's a lot easier to complete a pass to a wide open receiver. This sort of data would be wonderful to look at over the length of a career so you could see how a qb's stats changed with the surrounding personnel.

43
by John (not verified) :: Tue, 07/24/2007 - 9:43am

With a mom as a grammar teacher, the paragraph sentence order needs changed.

That has to be deliberate, right? Did you really butcher the grammar that badly on accident?

You just told us that the sentence order's mom is a grammar teacher.

(Reminds me of the time that a local university radio station broadcast an editorial defending the fact that the students often mispronounced words, "because we're just students." The beauty of the editorial: she mispronounced "mispronunciation" three times.)

44
by Fergasun (not verified) :: Tue, 07/24/2007 - 10:57am

Well, the truth about Campbell is that he was pretty inaccurate last season. I hope we'll see more about him on Wednesday. He really did vault from 3rd string into starter during the bye week (maybe I don't remember correctly).

I'd like to know how someone that inaccurate ended up ranking 20 spots higher in DVOA/DPAR. It's not a 1:1 correlation... but you would expect it to track pretty well. I take it to mean the 50% of the time he hit his receivers, it went for a first down and was a positive gain.

45
by DGL (not verified) :: Tue, 07/24/2007 - 11:05am

#25: "'Needs changed' should be “needs to be changed,� unless you live in Appalachia or Scotland."

Actually, the dropped infinitive is characteristic of Pittsburgh. The joke was always that if Hamlet was from Pittsburgh, his soliloquy would begin "Or not."

46
by admin :: Tue, 07/24/2007 - 11:11am

Overthrown and Thrown Away were two separate designations. Thrown Away specifically means "five feet over everybody's head out of bounds to avoid pressure," not just any long pass.

I can't believe the Madden people gave Tony Romo a higher passer accuracy rating than Peyton Manning. The NFL should pull their license for that.

47
by Theo, Holland (not verified) :: Tue, 07/24/2007 - 11:22am

#46.
The Madden people didn't do that, you did.

48
by MRH (not verified) :: Tue, 07/24/2007 - 11:52am

Great article.

Re Jason Campbell: as I understand it (Chiefs fan), the Vermeil/Martz/Saunders offense relies heavily on QB and WRs working together well (not that every offense doesn't, but it's more important in this scheme, it's a relative thing).

As the #3 guy, Campbell didn't get the reps w/the WRs necessary in pre-season to be completely effective in this offense. He also may not have been totally familiar with this offense (which is reputedly one of the more complex to learn). Todd Collins was specifically brought in as the backup from KC because he was familiar with the offense and could be the #2 while Campbell learned. But the Redskins never saw Collins as the #1, so when Bruenll was ineffective they gave Campbell the game experience.

It will be great when there are enough years of game charting to look back at, say, Trent Green's 1st year in KC where he really knew the offense but the WRs were still getting the hang of it (plus they sucked - Kennison was a huge upgrade when he was kicked out of DEN for going AWOL) - and Gonzo had held out so there was no chemistry there either. If Green's numbers reflected the same problems as Campbell's, then we could expect Campbell to improve. If not, then we'd expect Campbell to be lousy.

49
by James, London (not verified) :: Tue, 07/24/2007 - 11:59am

I find it astonishing that Joey Harrington doesn't make any of these lists, for good or ill. Miami led the league in dropped passes, and even then he's firmly middle of the pack in all categories.

Joey was 29th in DPAR last year and 31st in DVOA, while Miami had the 24th ranked pass offense. I'd expected his name to show in some of the rankings, even if the "My Receivers are Horrible!" adjustment took him out again.

50
by Omar (not verified) :: Tue, 07/24/2007 - 12:37pm

Re: 47

Well played, sir.

51
by RobinFiveWords (not verified) :: Tue, 07/24/2007 - 1:29pm

I'd be interested in seeing Drew Brees's splits when LaDainian Tomlinson lined up in the backfield vs. when Tomlinson was lined up out wide or not on the field. That's a crazy completion percentage for deep passes, but many of those attempts may have come against eight-in-the-box defenses. Perhaps the underlying factor that took Brees to a higher level is that he and Cam Cameron knew when to go deep.

52
by Tracy (not verified) :: Tue, 07/24/2007 - 2:06pm

51 -
I don't think those splits would tell you much, seeing as how Tomlinson and Brees did not play on the same team in 2006.

53
by seabos (not verified) :: Tue, 07/24/2007 - 2:36pm

I'd be interested to see drops against for some of the teams, especially Seattle. Seattle played maybe 5 or 6 games in rain/snow/sub freezing temperatures and I think it caused a lot of drops for them AND their opponents.

54
by CA (not verified) :: Tue, 07/24/2007 - 3:53pm

This article received a mention in The Wall Street Journal Online's "The Daily Fix" (click on the link in my name if you're reading this on Tuesday, July 24 and you have access to wsj.com content):

Two of the Fix's favorite sites believe stats can tell sports stories. Each delivers a particularly good story today...

Football Outsiders' Aaron Schatz's focus is on the field. He factors in pass length, dropped passes and spiked balls to examine NFL quarterbacks' true completion percentage. His most interesting findings include some surprises about the deep-throwing ability of Tom Brady and Brett Favre.

55
by Will Allen (not verified) :: Tue, 07/24/2007 - 4:32pm

Yeah, I thought so Aaron, but a fair number of overthrown passes have the same intent as throwing it out of bounds.

56
by Bobman (not verified) :: Tue, 07/24/2007 - 4:45pm

52 Tracy, you are so polite. Sometimes in these threads, someone might have been challenged to a knife fight for such an error.

Nice to have some civilization here.

Oh, and Aaron, stupendous work. To paraphrase "Jerry Maguire," your offseason articles complete me. (and your in-season stuff, along with the asssociated commentary, just causes me to piss away about 4 otherwise productive hours every damn day!)

57
by Pat (not verified) :: Tue, 07/24/2007 - 5:59pm

#55: I wouldn't really say 'same intent' - similar intent, maybe. A throwaway just ends the play with very little risk, and absolutely no chance at success. Intentional overthrows are similar to throwing the ball intentionally low, rather than high (like McNabb) - less chance of success, but less chance of dramatic failure.

That's really a separate issue, though, if you think about it, because it applies equally well to short passes being underthrown - it's an intentional reduction of accuracy to avoid a higher interception rate.

Interception rates are unfortunately low that looking at splits by distance would be a bit difficult, but you might be able to do it on a long enough average.

58
by Lou (not verified) :: Tue, 07/24/2007 - 7:39pm

Great work. And these numbers aren't adjusted for opponent either correct? So that makes Grossman's numbers that much worse?

le sigh

59
by OMO (not verified) :: Tue, 07/24/2007 - 8:10pm

"What the hell have I been saying for 3 years?"

You can't write this stuff. Comedy Gold.

60
by OMO (not verified) :: Tue, 07/24/2007 - 8:12pm

"His (Brady's) shoulder definitely was bothering him, but more damning is that Caldwell, Gabriel, and friends couldn’t get any separation on medium-to-long patterns, so any throw against man coverage was fairly low percentage."

And the retroactive "ball-washing" of Tom Brady's 2006-2007 season begins.

61
by NewsToTom (not verified) :: Tue, 07/24/2007 - 8:22pm

#58
Actually, I'm mildly curious about that, if certain teams are better at getting opponents to drop passes than others.

#60
begins continues.

Will Allen:
As a game charter, I generally didn't find it very hard to distinguish between passes that were overthrown and those that were simply thrown away. Except when Kerry Collins was at QB, because he was bad enough that throwing the ball 10 feet over an open receiver's head was plausible. In some sense, though, yes, these numbers are based in part on my sense of karmic justice, and what I think would accurately reflect to someone who didn't watch the game how the QB performed. Yes, that sentence scares me, too.

62
by morganja (not verified) :: Tue, 07/24/2007 - 8:38pm

Is this a northern Appalachian thing? I don't get the reference and I live in the heart of Scottish descended Appalachia.

63
by Pat (not verified) :: Tue, 07/24/2007 - 9:26pm

It's mainly northern Appalachia. Western PA/eastern Ohio (sometimes called "North Midland," although regionally, it's still Appalachia, although the Appalachian dialect region is south of it, and parts of the Appalachian and Midwest regions use it too). It's originally from a Scottish dialect, though. See here - down in the section on Scotch-Irish origins.

64
by MJK (not verified) :: Tue, 07/24/2007 - 9:30pm

And the retroactive “ball-washing� of Tom Brady’s 2006-2007 season begins.

I don't know about that. People were saying before last season "oh my god, the Patriots have no good recievers, they're going to be terrible, Brady will finally have a bad year".

And now people are saying "yeah, the Patriots had no good recievers, and yeah, Brady didn't have a fantastic year". Although to come within 2:34 of the SB and play in probably the most exciting football game of the last ten years counts for something... :-)

But as far as Tom Brady goes...it all depends on where in the QB-vs-WR camp you come down. If you think WR's are completely fungible, than you have to admit that Brady is a very good QB, but not so good that he never has off years. If you think WR's are very important, than you don't have enough data to say anything about Brady--he's had three excellent years (with decent WR's, 2002-2004), and three pretty good years, with average or below average WR's (2001, 2005-2006).

65
by David Brude (not verified) :: Tue, 07/24/2007 - 11:36pm

I did a comparison where I took the standard deviations on VOA and comp%.

Then I did VOA(stdev)-comp%(stdev) to see who is worse or better than their completion %

Overinflated comp%

8-D.Carr -2.85 1
9-C.Frye -2.10 2
14-B.Johnson -1.17 3
7-J.Losman -1.14 4
10-C.Pennington -1.05 5
2-C.Simms -1.00 6
2-A.Brooks -0.88 7
8-J.Kitna -0.85 8
3-D.Anderson -0.83 9
10-T.Green -0.71 10

Underinflated comp%

5-D.McNabb 2.15 1
17-J.Campbell 1.97 2
11-D.Huard 1.46 3
10-V.Young 1.38 4
18-P.Manning 1.37 5
7-M.Vick 1.13 6
4-B.Favre 1.00 7
8-R.Grossman 0.84 8
11-D.Bledsoe 0.80 9
17-P.Rivers 0.72 10

66
by Josh (not verified) :: Wed, 07/25/2007 - 8:25am

Great article.

The overweighting of short passes by Brady/Brees begs a question: what would a QB's completion % be if he threw the typical mix of short, medium, and long passes and after removing drops? This seems to fit the FO spirit of applying stats to averages.

The value could be tested by seeing if this adjusted, adjusted % predicted DVOA across seasons and/or across coaching changes.

This isn't perfect; in theory players who throw many short passes should find bombs easier because the defense will play the short pass, and vice versa. But it could improve the current system.

It would also be interesting to know who has the most consistent completion % across distance. Somebody who performed at 130% of the average % at all distances would be much harder to defend than somebody who was well above average short but well below average long.

67
by dryheat (not verified) :: Wed, 07/25/2007 - 8:38am

#60, beautiful pre-emptive strike.

However, in the paragraph that you cite, I don't see anything that can be remotely construed as ball washing. Tom Brady did have surgery prior to the season. His receivers did have problems separating from defenders. These are facts.

I think every Patriots fan here knows that Brady had his worst season since probably 2002 last year. Rich and I were fairly vocal stating this as it was happening.

68
by SGT Ben (not verified) :: Wed, 07/25/2007 - 9:19am

Woohoo! Charlie Fry is accurate! Hope Edwards learns to concentrate more...we might be able to pull a few upsets and go (gasp!) 8-8 this year!

/dreamlikestate

OK...maybe my Browns won't be that good...but it's nice to see we've got some talent there. All those people clamoring for Quinn...I hope this shuts them up.

69
by dryheat (not verified) :: Wed, 07/25/2007 - 9:58am

Isn't Derek Anderson the favorite to start right now?

70
by MJK (not verified) :: Wed, 07/25/2007 - 12:32pm

dryheat,

By conventional stats, Brady had a pretty good 2002. He led the league in TD's (and possibly in yards?). The Patriots just couldn't run or stop the run, which, though we know is not the way to win games as CW would have us believe, is certainly not a formula for success.

Josh,

That's a really cool idea. Sort of reminds me of what Aaron has told us of how DVOA got started, looking at special teams averages... The only problem is, while it seems like it would be a nice tool for comparing different QB's across different systems and gauging how good a QB might be in a generic system, a good coach will taylor his system to the strengths of his players (or look for players whose strengths fit his system), so it will NOT have a lot of predictive value. I.e. if it tells us that Drew Brees is more accurate than Donovan McNabb by this average measure, for example, that tells us something about how each might perform in a controlled passing contest, or if they were forced to play for the Giants, but does not predict how each might perform on their respective actual teams.

71
by OMO (not verified) :: Wed, 07/25/2007 - 1:00pm

"Tom Brady did have surgery prior to the season. His receivers did have problems separating from defenders. These are facts."

Actually the surgery is a fact. The WR separation is your opinion.

Additionally to call it (WR sep) a contributory factor to Brady performance you are going to have to show that the WR performance was worse than other WR unit's performance in prior years.

Another curious mention...if he had surgery in the off season...don't you think the resulting affects from the surgery would have been a slow start?

Why then was Brady's best month in October and his worst month in November?

Also...isn't this the hallmark of the greatness that is Tom Brady? He can lead his team to victory even with a WR unit that is beyond horrible? As if his passes just seem to will themselves into the arms of 2nd and 3rd rate WR? ;)

72
by SGT Ben (not verified) :: Wed, 07/25/2007 - 1:31pm

#69

I wouldn't call him the favorite. The Coaches do like the fact that he has a more powerful arm and his size...but Frye has more starting experience (game speed) which is far more important than just who performs well when nobody's wearing pads.

73
by dryheat (not verified) :: Wed, 07/25/2007 - 1:47pm

OMO, I understand that your love to bash Brady and the Patriots is as great as the need for many FO posters to praise them. I get it. Honestly, I do.

Without answering the questions/comments you pose, which would be simply argumentative, the point I was trying to make is that in order to get a truer read on Mr. Brady's accuracy, it's far, far better to get a longer view - perhaps the last three seasons - than looking at last year and inferring that he can't throw the deep ball. I expect that to be true of all the quarterbacks on the list who have been in the league to have built up a credible sample size. I'm willing to bet guys like Jason Campbell and Alex Smith will have better numbers going forward.

I have absolutely no problem stating that Brady struggled throwing the ball last season. I know it as well as anyone -- I watched it happen. However, I also think it erroneous to say that Brady is an inaccurate quarterback in the passing game.

74
by OMO (not verified) :: Wed, 07/25/2007 - 3:00pm

So the posting name of dryheat also refers to your playful sense of humor?

Lighten up Francis.

75
by Pat (not verified) :: Wed, 07/25/2007 - 3:47pm

#66: That's what that "adjusted completion percecntage" is up there.

#71: WRs getting or not getting separation is a fact, not opinion. How much of a problem it was is the opinion, but anyone can tell whether or not a WR is gaining any separation on some of the plays.

And jeez, play nice. Posting random insults (especially with no content) is below this site.

76
by NY fan in MT (not verified) :: Wed, 07/25/2007 - 3:50pm

#31 pointed out that Eli Mannning didn't show up in any of the bottom 10 lists, but equally surprising (according to conventional wisdom) is that Chad Pennington is in the top 10 after removing short passes.

77
by OMO (not verified) :: Wed, 07/25/2007 - 4:23pm

"but anyone can tell whether or not a WR is gaining any separation on some of the plays."

Anyone? On some of the plays? Under what defination? 3 yards? 5 yards? 1 yard? What's separation?

Unless you define separation as a yes/no condition, it's a relative term, thus judgment comes into play, thus it's opinion-based, not fact-based.

And what random insults? And better yet...when did you annoint yourself as the moral compass of this site?

78
by billvv (not verified) :: Wed, 07/25/2007 - 4:54pm

#76 Amen brother!

79
by Pat (not verified) :: Wed, 07/25/2007 - 8:57pm

#77: About two yards or so. One yard isn't enough separation, as a well timed leap and reach could just about hit it away.

And c'mon. #74 was a random insult, and so was the "moral compass" accusation. There's just no point. It doesn't add anything to the discussion.

80
by Brian (not verified) :: Wed, 07/25/2007 - 10:27pm

Completion percentage is double counted in the conventional passer rating formula. It is one of the 4 components and also half of the yds/attempt component. (Yds/att = Yds/completion x completion percentage.)

This article underscores just how ridiculous passer ratings are.

81
by Stuart Smalley (not verified) :: Thu, 07/26/2007 - 9:37pm

Hey Pat, don't be such an over-sensitive pussy.

82
by OMO (not verified) :: Thu, 07/26/2007 - 10:07pm

Ok, even I think that was totally over-the-top...

83
by Doug Farrar :: Sat, 07/28/2007 - 10:56pm

#33 - I believe the Bethel Johnson reception you're referring to occurred in a 30-20 win over the Seahawks in 2004. I remember that play distinctly because Seattle was uncharacteristically hanging in with the NFL's best team (the 2004 Seahawks squad was maddeningly inconsistent and could be taken out of games very easily) until Johnson got past the Seahawks' entire secondary, which had been unwisely set to "NFL Films Slo-Mo" in that particular instance.

84
by Soxphan (not verified) :: Mon, 08/20/2007 - 12:26pm

I love the digs on Brady that he can't complete long passes. Give him the receivers to do it (Moss & Stallworth this year) or (Patton/Givens/Branch) in past Superbowl years, and he comes through fine. Sure he threw short passes last year, you would too with that receiving core. Not all teams are lucky enough to have receivers like Harrison & Wayne, two of the main reasons why Peyton Manning is so good at completing long passes.