2014 Depth-adjusted Completion Percentage

Mike Clay adjusts each quarterback's completion percentage based on the average depth of target.

| 2 years ago
nfl_g_brees11_576

2014 Depth-adjusted Completion Percentage


Brees - Advanced Drop BacksI’m starting to think that I should rename this annual study the ‘Alex Smith Index’.

Smith, of course, is the poster boy for rarely throwing deep balls. As conservative as NFL quarterbacks come, Smith’s average depth of target (aDOT) was 6.0 this past season. That’s one full yard per throw behind Robert Griffin III for lowest in the league. Smith’s marks of 6.9 in 2013 and 7.7 in 2014 were lowest and second lowest (Christian Ponder), respectively.

It doesn’t take a rocket scientist to figure out that shorter passes are easier to complete, which explains why Smith often ranks so highly in completion percentage each year. The question is – how much of an impact does depth actually have? To answer that question, I went play by play and adjusted each quarterback’s completion percentage based on the distance of his throws. Quarterbacks who usually throw a lot of screens and short passes will be punished. Those who aren’t afraid to chuck it deep will be rewarded.

To accomplish this, we’re using a pair of statistics. One is aDOT and the other is an adjusted version of completion percentage (aC%)*.

*aC% = (Completions + Drops – NFL average drop rate) / (Pass Attempts – Spikes – Throwaways – Batted Balls – Balls disrupted by a QB hit)

Note that quarterbacks who attempted a minimum of 100 passes (including playoffs) are included and throws with depths shorter than negative-9 yards and ones longer than 55 yards are removed. Why? Although I’m using a massive play-by-play sample that covers eight years of NFL action, there remain distances without an ideal sample size. For what it’s worth, only 117 (or 0.087 percent) of 135,193 aimed throws have been eliminated as a result of this tweak.

I’ll be referring to “aC%” throughout this piece. Don’t be confused by it. Remember, it’s just an adjusted version of completion percentage that removes wasted throws and normalizes drops.

The Good

Our first chart shows the top-10 quarterbacks in terms of their actual aC% exceeding their expected.

Rk Quarterback Aimed aDOT Actual aC% Expected aC% +/-
1 Drew Brees 607 8.1 74% 67% 7.1%
2 Tony Romo 467 9.5 69% 65% 4.1%
3 Philip Rivers 532 9.0 69% 65% 4.0%
4 Carson Palmer 209 9.2 69% 65% 3.6%
5 Teddy Bridgewater 362 7.9 71% 67% 3.6%
6 Matt Ryan 588 8.2 70% 66% 3.5%
7 Ben Roethlisberger 623 9.1 69% 65% 3.4%
8 Alex D. Smith 421 6.0 73% 70% 3.3%
9 Russell Wilson 469 9.0 68% 65% 3.3%
10 Tom Brady 671 8.3 69% 66% 2.6%

Drew Brees (+7.1 percent) was easily the league’s top quarterback in this department during the 2014 regular season. Brees (69.2 percent) trailed only Tony Romo (69.7 percent) in straight-up completion percentage, but Romo benefited from a low drop rate and had significantly fewer “unaimed” throws. Brees’ +7.7 percent is the highest we’ve seen from him in eight years of this study, but he’s been no worse than +2.3 percent in a given year. It’s far from a secret, but Brees is arguably the league’s most-efficient passer.

Teddy Bridgewater (+3.6 percent) was clearly the top rookie quarterback of 2014 and this study just adds to his intrigue. Bridgewater sported the league’s sixth-lowest aDOT (7.9), but made up for the conservative play with the fourth-best aC% (71 percent). The 22 year old will be a major breakout candidate in 2015.

Aforementioned Smith (+3.3 percent) sits seventh in the category this season after failing to crack the top-25 in 2013. As a result of his absurdly-low aDOT, Smith’s expected aC% of 70 percent was easily highest in the league. He took advantage of the “easier” throws, finishing second to only Brees with a 73 percent actual aC%. Smith isn’t nearly as good as many of the names on this list, but he’s thrived in his conservative/game-manager role.

The Bad

Rk Quarterback Aimed aDOT Actual aC% Expected aC% +/-
1 Ryan Lindley 116 12.3 50% 60% -10.0%
2 E.J. Manuel 122 7.6 62% 67% -5.7%
3 Brian Hoyer 409 10.8 58% 63% -4.9%
4 Blake Bortles 444 7.4 64% 68% -4.1%
5 Derek Carr 546 8.5 63% 66% -3.4%
6 Cam Newton 492 9.8 61% 64% -2.7%
7 Michael Vick 110 9.9 63% 65% -2.6%
8 Zach Mettenberger 168 9.1 62% 64% -2.4%
9 Jake Locker 141 10.1 61% 63% -2.3%
10 Drew Stanton 222 11.7 60% 61% -1.6%

I ran this study back in 2013 and Ryan Lindley (-8.5 percent) finished dead last in the category. Fast forward two years and Lindley (-10 percent) is back in the basement. Certainly one of the worst quarterbacks to make multiple starts over the past decade, Lindley wasn’t afraid to chuck it deep (12.3 aDOT), which allowed him some leeway in terms of expected aC% (60 percent). Of course, he completed only half of his passes, which was worst in the league by eight percentage points. Lindley is unlikely to take another snap in the NFL.

A trio of rookies shows up on our list. Blake Bortles (-4.1 percent), Derek Carr (-3.4 percent) and Zach Mettenberger (-2.4 percent) didn’t exactly have a great supporting cast, but each struggled in the efficiency department. Bortles had the best aC% of the trio, but was also easily the most conservative. With what is shaping up to be one of the league’s top/deepest wide receiver units at his disposal in 2015, he’ll need to be much better as a sophomore.

Yes, Johnny Manziel was horrific when called upon as a rookie, but Brian Hoyer (-4.9 percent) is obviously not the answer in Cleveland. Hoyer completed 55 percent of his throws, which was worst among 30 passers who attempted at least 300 aimed throws.

Eight Years Later

There are 40 passers who have 1,000 aimed throws under their belt since I started this study in 2007.

Aaron Rodgers (+5.8 percent) is tops in the category during that span. Brees (+4.9 percent), Kurt Warner (+4.1 percent), Peyton Manning (+3.9 percent) and Philip Rivers (+3.8 percent) round out the top-five.

On the other hand, Derek Anderson (-4.8 percent) is worst in the category since 2007. He’s followed closely by Marc Bulger (-4.7 percent), Mark Sanchez (-4.2 percent), Josh Freeman (-2.5 percent) and Matt Hasselbeck (-2.3 percent).

The complete 2014 rundown

Rk Quarterback Aimed aDOT Actual aC% Expected aC% +/-
1 Drew Brees 607 8.1 74% 67% 7.1%
2 Tony Romo 467 9.5 69% 65% 4.1%
3 Philip Rivers 532 9.0 69% 65% 4.0%
4 Carson Palmer 209 9.2 69% 65% 3.6%
5 Teddy Bridgewater 362 7.9 71% 67% 3.6%
6 Matt Ryan 588 8.2 70% 66% 3.5%
7 Ben Roethlisberger 623 9.1 69% 65% 3.4%
8 Alex D. Smith 421 6.0 73% 70% 3.3%
9 Russell Wilson 469 9.0 68% 65% 3.3%
10 Tom Brady 671 8.3 69% 66% 2.6%
11 Ryan Tannehill 555 8.2 69% 67% 2.6%
12 Aaron Rodgers 559 8.6 69% 66% 2.5%
13 Kirk Cousins 192 9.0 68% 66% 2.3%
14 Robert Griffin III 199 7.0 72% 70% 2.2%
15 Joe Flacco 581 9.2 67% 66% 1.8%
16 Colin Kaepernick 445 9.5 66% 64% 1.6%
17 Andrew Luck 682 9.2 66% 65% 1.5%
18 Shaun Hill 211 8.3 68% 67% 1.3%
19 Ryan Fitzpatrick 294 9.1 66% 65% 1.2%
20 Eli Manning 564 9.2 66% 65% 1.1%
21 Peyton Manning 623 9.0 66% 65% 0.7%
22 Jay Cutler 529 7.7 68% 68% 0.4%
23 Kyle Orton 415 7.5 68% 68% 0.2%
24 Geno Smith 329 8.6 66% 66% 0.2%
25 Colt McCoy 124 7.3 69% 68% 0.1%
26 Austin Davis 264 8.7 66% 66% 0.0%
27 Mike Glennon 188 11.6 61% 61% -0.3%
28 Andy Dalton 482 8.1 66% 67% -0.8%
29 Charlie Whitehurst 163 10.7 62% 62% -0.8%
30 Mark Sanchez 301 8.8 65% 66% -1.0%
31 Nick Foles 285 10.3 62% 64% -1.3%
32 Matthew Stafford 593 8.2 66% 67% -1.3%
33 Josh McCown 300 10.7 61% 63% -1.5%
34 Drew Stanton 222 11.7 60% 61% -1.6%
35 Jake Locker 141 10.1 61% 63% -2.3%
36 Zach Mettenberger 168 9.1 62% 64% -2.4%
37 Michael Vick 110 9.9 63% 65% -2.6%
38 Cam Newton 492 9.8 61% 64% -2.7%
39 Derek Carr 546 8.5 63% 66% -3.4%
40 Blake Bortles 444 7.4 64% 68% -4.1%
41 Brian Hoyer 409 10.8 58% 63% -4.9%
42 E.J. Manuel 122 7.6 62% 67% -5.7%
43 Ryan Lindley 116 12.3 50% 60% -10.0%

Follow Mike Clay on Twitter: @MikeClayNFL

  • PetEng

    Can we include aDot in the signature stats? Seems like a highly valuable piece of information.

  • MikeClayNFL

    It’s available with the PFF Fantasy Gold package. Thanks!

  • flounder

    I wonder how much a feedback loop like this plays into these results:
    1. Coach knows QB is not good, so tries to protect him by running the ball.
    2. Offensive Line/RB are terrible, so they get stuffed/take holding call.
    3. “Not good” QB has to attempt hard 3rd and long pass and is not successful.

    Also, I am a Chargers fan, and one common criticism this year was Reich had gotten away from the short passing game that SD had success with in 2013. Seeing Rivers’ 9.0 aDOT confirms this.

    • Mike Clay

      Hey Flounder – You’d be right that Rivers’ aDOT spiked quite a bit in 2015. He went from 7.8 in 2012 to 8.1 in 2013 to 9.0 in 2014. For what its worth, he was at 9.0+ every year from 07 – 11