Writers of Pro Football Prospectus 2008

21 Feb 2017

Failed Completions 2016

by Scott Kacsmar

Given their inability to kill clock and preserve a lead in the Super Bowl, a completed pass or two might sound really good to the Atlanta Falcons right now, even if they came up well short of the sticks. In general, though, the vast majority of the thousands of catches we see like that each year don't do much to help an offense out. Fans of PPR scoring and Tavon Austin's agent love these plays, but DVOA and general offensive success do not.

Unfamiliar with failed completions? They are one of the quickest statistics we can measure from the season's play-by-play data. A failed completion is defined as any completed pass that fails to gain 45 percent of needed yards on first down; 60 percent of needed yards on second down; or 100 percent of needed yards on third or fourth down. You can see last year's study here.

Not every failed completion is created equally -- some will move an offense into field goal range, for example -- but for this article we are making things binary, simply summing up successes and failures. For the purposes of DVOA, there are fractional points involved where a 9-yard completion on third-and-11 would generate some partial success.

Our data on failed completions currently goes back to 1989, but we can add 1986-88 next year. However, the NFL in that era featured a different brand of passing. In 1989, 18.3 percent of all completions were failures. That rate last dipped below 23.1 percent in 2007, and peaked at 25.1 percent in 2015 before falling to 24.5 percent in 2016, the third-highest season on record.

Failed completions have never been more common thanks to the overall increase in pass attempts and a metric like ALEX, which looks at where the ball is being thrown in relation to the first-down marker. In particular, ALEX's focus on third and fourth downs shows us where the most damaging failed completions are occurring. Some quarterbacks are simply more likely to complete a 2-yard pass on third-and-10 instead of trying to actually gain a first down. We tend to point out games with a high number of failed completions during the season in our weekly analysis articles. The data below is only for the regular season. The total number of completions is slightly higher than the official NFL total since Football Outsiders includes backward (lateral) passes.

Quarterbacks

In the following table, the 34 qualified quarterbacks are ranked by ascending failed completion rate (FC%). We also included failed completions as a percentage of attempts (there is very little change in these rankings), as well as the average ALEX (all downs) for the season.

Rk Player Team Passes Comp. Failed FC% FC%ATT Rk ALEX Rk
1 Matt Barkley CHI 216 129 14 10.9% 6.5% 1 1.1 1
2 Cam Newton CAR 506 270 49 18.1% 9.7% 2 1.1 2
3 Jameis Winston TB 567 345 64 18.6% 11.3% 3 0.9 3
4 Carson Palmer ARI 596 365 71 19.5% 11.9% 4 0.3 5
5 Andrew Luck IND 542 346 68 19.7% 12.5% 6 -0.5 15
6 Philip Rivers SD 576 348 70 20.1% 12.2% 5 -0.2 12
7 Matt Ryan ATL 534 373 77 20.6% 14.4% 11 -0.6 17
8 Dak Prescott DAL 458 312 66 21.2% 14.4% 10 -0.8 21
9 Aaron Rodgers GB 609 401 88 21.9% 14.4% 13 0.2 8
10 Drew Brees NO 667 472 105 22.2% 15.7% 21 -1.3 28
11 Andy Dalton CIN 561 364 81 22.3% 14.4% 12 -1.0 22
12 Marcus Mariota TEN 447 275 62 22.5% 13.9% 9 0.8 4
13 Trevor Siemian DEN 485 288 65 22.6% 13.4% 7 -0.3 13
14 Kirk Cousins WAS 603 406 93 22.9% 15.4% 18 0.0 11
15 Tom Brady NE 429 290 67 23.1% 15.6% 20 -0.8 20
16 Matthew Stafford DET 591 388 90 23.2% 15.2% 16 -1.3 26
17 Ben Roethlisberger PIT 506 328 79 24.1% 15.6% 19 0.2 9
Rk Player Team Passes Comp. Failed FC% FC%ATT Rk ALEX Rk
18 Ryan Fitzpatrick NYJ 403 228 55 24.1% 13.6% 8 0.3 6
19 Colin Kaepernick SF 330 196 48 24.5% 14.5% 14 -1.2 24
20 Russell Wilson SEA 545 354 87 24.6% 16.0% 22 -0.6 16
21 Brian Hoyer CHI 198 134 33 24.6% 16.7% 24 -1.2 25
22 Blake Bortles JAC 626 369 93 25.2% 14.9% 15 -0.4 14
23 Brock Osweiler HOU 508 301 78 25.9% 15.4% 17 0.1 10
24 Tyrod Taylor BUF 434 268 71 26.5% 16.4% 23 0.3 7
25 Alex Smith KC 489 329 88 26.7% 18.0% 26 -1.9 32
26 Carson Wentz PHI 605 379 108 28.5% 17.9% 25 -1.6 31
27 Eli Manning NYG 599 378 110 29.1% 18.4% 27 -0.8 18
28 Sam Bradford MIN 550 395 116 29.4% 21.1% 32 -2.5 33
29 Derek Carr OAK 558 357 105 29.4% 18.8% 28 -1.3 27
30 Cody Kessler CLE 194 128 38 29.7% 19.6% 30 -1.4 29
31 Ryan Tannehill MIA 387 261 78 29.9% 20.2% 31 -1.0 23
32 Case Keenum LARM 322 196 63 32.1% 19.6% 29 -0.8 19
33 Joe Flacco BAL 664 436 144 33.0% 21.7% 34 -1.6 30
34 Jared Goff LARM 204 112 44 39.3% 21.6% 33 -2.8 34

We have a new record for failed completions in a single game since 1989. Oakland's Derek Carr had 16 failed completions in Week 8 against Tampa Bay. To be fair, that overtime game went nearly five full quarters, and Carr did go 40-of-59 passing for 513 yards. The previous record belonged to Peyton Manning, who had 15 failed completions (12 after halftime) in that miserable 2014 AFC divisional round loss by Denver to the Colts. But 2016 also saw that mark of 15 failed completions matched twice; first by Sam Bradford and the Vikings against Detroit (Week 12), and then by Joe Flacco and the Ravens in New England (Week 14).

Flacco actually smashed the record for most failed completions in a season with 144, breaking Matt Ryan's mark of 120 in 2013. It did not matter that the Ravens replaced offensive coordinator Marc Trestman with Marty Mornhinweg during the season. Flacco had nine of the 38 performances this season in which an offense had at least 10 failed completions, including four from Weeks 12 to 16. No other offense had more than three such games in 2016. Flacco also had five games in a row with double-digit failed completions from Weeks 3 to 7.

Here is every season with at least 100 failed completions since 1989.

Seasons with 100+ Failed Completions Since 1989
Rk Player Year Team Comp. Failed FC%
1 Joe Flacco 2016 BAL 436 144 33.0%
2 Matt Ryan 2013 ATL 439 120 27.3%
3 Sam Bradford 2016 MIN 395 116 29.4%
4 Drew Brees 2015 NO 429 113 26.3%
5 Derek Carr 2014 OAK 348 112 32.2%
6 Joe Flacco 2013 BAL 362 111 30.7%
7 Eli Manning 2016 NYG 378 110 29.1%
8 Drew Brees 2010 NO 449 109 24.3%
9 Carson Wentz 2016 PHI 379 108 28.5%
10 Drew Brees 2013 NO 447 107 23.9%
11 Derek Carr 2016 OAK 357 105 29.4%
Ryan Tannehill 2015 MIA 364 105 28.8%
Peyton Manning 2002 IND 392 105 26.8%
Philip Rivers 2015 SD 438 105 24.0%
Drew Brees 2016 NO 472 105 22.2%
16 Sam Bradford 2010 STL 354 104 29.4%
Drew Brees 2014 NO 456 104 22.8%
18 Jay Cutler 2014 CHI 370 102 27.6%
19 Ryan Fitzpatrick 2011 BUF 354 101 28.5%
Matt Ryan 2014 ATL 415 101 24.3%
21 Matt Hasselbeck 2011 TEN 319 100 31.3%
Sam Bradford 2015 PHI 346 100 28.9%
Kirk Cousins 2015 WAS 379 100 26.4%

When we say "since 1989," we might as well say "in all of NFL history," since only one season on this list (Peyton Manning in 2002) occurred before 2010. At the time, Manning's 392 completions in a season ranked fourth in NFL history. It now ranks in a tie with Ryan Tannehill's 2014 campaign for 38th. This is a great example of how much the game has changed in recent years with the reliance on more passes in general and more short passes in particular, thus a greater frequency of failed completions.

However, Baltimore fans should feel some concern here, as Flacco ranked next to last in 2016 in FC%, and had the worst rate of failed completions per attempt. He also had the highest FC% (33.0 percent) among the 23 seasons with 100 or more failed completions. While someone like Drew Brees does appear on that list five times, he had five of the seven lowest FC% on the list, including the lowest one (22.2 percent in 2016). It is different when you efficiently break the NFL's single-season completion mark several times like Brees has. Flacco got to his first 4,000-yard passing season in cheap fashion. In fact, out of the 155 4,000-yard passing seasons in NFL history, Flacco had the lowest yards per attempt (6.42) and the second lowest touchdown rate (3.0 percent).

Let's stick with the lowlights. The Rams brought their depressing passing game from St. Louis to Los Angeles, and the result was fitting: another lousy sequel to 2015 when Nick Foles put up the highest FC% since 1989 (41.1 percent) before he was benched for Case Keenum. This year the Rams benched Keenum and finally sacked head coach Jeff Fisher, but rookie Jared Goff did his best to try matching Foles' futility. He only ended up with the fourth-highest FC% in a season since 1989 (minimum 100 completions).

Highest Failed Completion Rate Since 1989 (Min. 100 Completions)
Rk Player Year Team Comp. Failed FC%
1 Nick Foles 2015 STL 190 78 41.1%
2 Anthony Wright 2005 BAL 164 65 39.6%
3 Tim Couch 2000 CLE 137 54 39.4%
4 Jared Goff 2016 LARM 112 44 39.3%
5 Brady Quinn 2012 KC 113 44 38.9%
6 Drew Brees 2003 SD 205 75 36.6%
7 Bruce Gradkowski 2006 TB 177 64 36.2%
8 Bernie Kosar 1993 2TM 115 41 35.7%
9 Trent Edwards 2009 BUF 110 39 35.5%
10 Blaine Gabbert 2015 SF 178 63 35.4%

Brees rebounded nicely with a breakout year in 2004, but based on this list, Goff has his work cut out for him. Maybe there is something about trying to throw to Tavon Austin (anywhere on the field) and Todd Gurley (without any blocking) that just fundamentally does not work. Maybe a new offensive-minded coach in Sean McVay will figure this out in 2017, but there has clearly been a lot wrong for a long time with the Rams' passing game.

When we look at the top of the 2016 results, we do not exactly see stellar overall performance from the passers with the best failed completion numbers. Sure, in 2015, Cam Newton and Carson Palmer were in the top three in FC%, FC%ATT, and ALEX, but things did not go so well in 2016. Of course, the same can be said of the Panthers and Cardinals in general a year after meeting in the NFC Championship Game. Jameis Winston also finished third in FC% for the second season in a row, but we'll do a more extensive statistical look at him this offseason. Let's just say the stylistic similarities to Newton are obvious.

You have to go down to the seven to nine range to start finding some of the major standouts in 2016 quarterback play with MVP Matt Ryan, Offensive Rookie of the Year Dak Prescott, and Aaron Rodgers. Ryan had a historic season with 9.3 yards per pass attempt, but his ranking of seventh in FC% made me wonder how the season's top quarterback has historically fared in this metric. So I looked at the seasonal rank in FC% (minimum 100 completions) for the AP's first-team All-Pro quarterback for each season since 1989.

Failed Completions: 1st-Team All-Pro QB
Player Year Team Comp. Failed FC% Rk
Joe Montana 1989 SF 271 52 19.2% 19
Joe Montana 1990 SF 321 65 20.2% 18
Jim Kelly 1991 BUF 304 38 12.5% 2
Steve Young 1992 SF 268 53 19.8% 15
Steve Young 1993 SF 314 51 16.2% 4
Steve Young 1994 SF 324 56 17.3% 4
Brett Favre 1995 GB 359 59 16.4% 3
Brett Favre 1996 GB 325 65 20.0% 13
Brett Favre 1997 GB 304 52 17.1% 4
Randall Cunningham 1998 MIN 259 48 18.5% 7
Kurt Warner 1999 STL 325 63 19.4% 6
Rich Gannon 2000 OAK 284 56 19.7% 10
Kurt Warner 2001 STL 375 67 17.9% 1
Player Year Team Comp. Failed FC% Rk
Rich Gannon 2002 OAK 418 81 19.4% 5
Peyton Manning 2003 IND 379 89 23.5% 14
Peyton Manning 2004 IND 337 48 14.2% 1
Peyton Manning 2005 IND 305 49 16.1% 2
Drew Brees 2006 NO 358 81 22.6% 18
Tom Brady 2007 NE 398 56 14.1% 1
Peyton Manning 2008 IND 371 79 21.3% 13
Peyton Manning 2009 IND 393 79 20.1% 5
Tom Brady 2010 NE 324 73 22.5% 11
Aaron Rodgers 2011 GB 344 63 18.3% 5
Peyton Manning 2012 DEN 401 81 20.2% 6
Peyton Manning 2013 DEN 450 84 18.7% 2
Aaron Rodgers 2014 GB 341 58 17.0% 1
Cam Newton 2015 CAR 296 44 14.9% 1
Matt Ryan 2016 ATL 373 77 20.6% 7

Five of the last 16 quarterbacks finished No. 1, but Ryan ranked the lowest since Tom Brady ranked 11th in 2010. Still, every season ranked in the top 20, even with players from a few of the more old-school West Coast offenses (Joe Montana and Steve Young) ranking 15th or lower. Drew Brees ranked 18th in 2006, though he wasn't a good choice by the voters that year over Peyton Manning, who ranked first in DVOA (doubling up Brees) and first in FC% (13.0 percent). In case you were wondering, Brees received 25 votes to 24 for Manning (with one for Marc Bulger).

Ryan was likely not expecting to get competition from Chicago's Matt Barkley, who had the third-lowest FC% (10.9 percent) since 1989. Yeah, that randomness really happened. Barkley also had the highest ALEX this season, and the correlation between FC% and ALEX was minus-0.73. We know Barkley still ranked 26th in DVOA and threw an interception on an unseemly 6.5 percent of his attempts, but his aggressive approach as a backup with nothing to lose was refreshing to see. Barkley's passes came with an average deficit of 8.6 points, the second-worst average deficit in 2016, so he had to dig the Bears out of some big holes. He nearly did so against the Packers, Lions, and Titans, but some bad drops, questionable holding penalties, and conservative coaching cost Chicago those games in the end. Combined with his low sack rate (2.7 percent), Barkley had a very interesting statistical season, and his relevance in 2016 sure is a compliment given how far he fell to the fourth round in the 2013 draft, and how horrific he looked as a rookie with the Eagles.

Looking at someone like Barkley by down is also useful. The next quarterback table compares FC% to ALEX (all passes) by down, where third and fourth down are grouped together. Rankings are from best (green) to worst (red), and the darker the color, the more standard deviations that quarterback is from the average.

Barkley had the lowest FC% for each down, and was always high in ALEX. Meanwhile, Goff was often very conservative despite dealing with the largest deficits (10.6 points on average) of any quarterback. A whopping 56.1 percent of his third-down completions were failed plays, the biggest outlier in the table. It should also be pointed out that Chicago's Brian Hoyer ranked next to last in FC% on third and fourth down, compared to Barkley ranking first. Hoyer completed 67.0 percent of his passes and did not throw an interception for the Bears, but they were still not scoring many points behind him with so many of his minimal gains, especially on money downs.

Perhaps the most fitting part of this table is that Alex Smith and Sam Bradford had the lowest ALEX on first-down passes, and the corresponding bad FC% numbers to match. Flacco was also pretty conservative on each down in racking up his record number of failed completions. Meanwhile, Ben Roethlisberger was surprisingly conservative on first down, ranked 32nd in ALEX compared to second on second down and first on the money downs. Aaron Rodgers also showed a similar split, though Rodgers and Roethlisberger are often among the most aggressive third-down passers in the NFL.

Before the Super Bowl, I pointed out that third down was not a strong point in Matt Ryan's otherwise stellar season. Here he was only 31st in ALEX and 23rd in FC%, but Ryan's failure to pull the trigger on third down may have cost him the most in Super Bowl LI. Four of his five sacks came on third down, including the game-changing fumble in the fourth quarter. The correlation between FC% and ALEX was much stronger on the money down (minus-0.74) than it was on first (minus-0.58) or second (minus-0.51) down.

Receivers

How about the players on the receiving end of these plays? I looked at the failed receptions for all wide receivers and tight ends with at least 30 receptions. I excluded running backs since they dominate these lists with all the short passes they catch. DeMarco Murray led all running backs with 29 failed receptions in his first year in Tennessee, and the Giants' Rashad Jennings had the highest FC% (61.1 percent).

Most Failed Receptions (WR/TE)
Lowest Failed Reception Rate (WR/TE)
Highest Failed Reception Rate (WR/TE)
Receiver Team Failed Receiver Team Rec. Failed Pct. Receiver Team Rec. Failed Pct.
Dennis Pitta BAL 31 J.J. Nelson ARI 34 1 2.9% Tavon Austin LARM 58 25 43.1%
Antonio Brown PIT 28 Alshon Jeffery CHI 52 2 3.8% Albert Wilson KC 31 13 41.9%
Jarvis Landry MIA 26 Dez Bryant DAL 50 2 4.0% Eddie Royal CHI 33 12 36.4%
Doug Baldwin SEA 26 John Brown ARI 39 2 5.1% Clive Walford OAK 33 12 36.4%
Golden Tate DET 26 Mike Evans TB 96 5 5.2% Dennis Pitta BAL 86 31 36.0%
Tavon Austin LARM 25 Michael Floyd ARI/NE
37 2 5.4% Cordarrelle Patterson MIN 52 18 34.6%
Larry Fitzgerald ARI 23 Dwayne Allen IND 35 2 5.7% Tyreek Hill KC 61 21 34.4%
Amari Cooper OAK 23 Kelvin Benjamin CAR 63 4 6.3% Ty Montgomery GB 44 15 34.1%
Jason Witten DAL 23 Tajae Sharpe TEN 41 3 7.3% Jason Witten DAL 69 23 33.3%
Odell Beckham NYG 21 Greg Olsen CAR 80 6 7.5% Charles Clay BUF 57 18 31.6%
Jordan Matthews PHI 21 Antonio Gates SD 53 4 7.5% Nelson Agholor PHI 36 11 30.6%
Tyreek Hill KC 21 Rishard Matthews TEN 65 5 7.7% Breshad Perriman BAL 33 10 30.3%
Minimum 30 receptions.

Jarvis Landry has ranked in the top three in failed receptions in each of his three seasons, though he did have his lowest rate (27.7 percent) yet. Some of the other names are not surprising in that they are players who catch a lot of ball, period (such as Antonio Brown and Odell Beckham Jr.), and some older players who may not be as fast as they used to be (such as Jason Witten and Larry Fitzgerald).

Then we have tight end Dennis Pitta and his league-leading 31 failed receptions. Someone had to be helping Flacco to that record, and it was his old buddy Pitta, who hadn't caught a pass since 2014 after some serious hip injuries. Pitta averaged 10.5 yards per reception through Baltimore's 2012 Super Bowl season, but those injuries have limited him to 23 games in the last four years, and slowed him down to just 8.4 yards per reception. Wide receiver Breshad Perriman was also not a real big-play threat in his recovery from a rookie season wiped out by injury. With the Ravens losing Steve Smith to retirement, a new weapon could be in order for this offense.

The players in the middle table with the lowest failed rate are used to making catches beyond the sticks. J.J. Nelson developed nicely as more than just a speed demon for the Cardinals this season. Michael Floyd became expendable in Arizona, largely due to his own stupidity, but you can see that the Cardinals had three of the top six players with the lowest failed completion rate. Meanwhile, Tajae Sharpe was not quite the fifth-round rookie phenom we thought he had a chance to be for Tennessee, but Rishard Matthews quietly had another efficient season. Matthews was 16th in DVOA after finishing second in Miami in 2015.

The table with the highest failed reception rates is chock-full of the NFL's notable gimmick players. Albert Wilson, Eddie Royal, and Cordarrelle Patterson are just a few players who seem to rarely catch anything but screens. Tyreek Hill is also at the mercy of Alex Smith in Kansas City, and Ty Montgomery converted from wide receiver to running back for the Packers. Then we have the league's most expensive gimmick. Tavon Austin ranking high with the worst rate is not a surprise due to his laughable usage with the Rams. Austin finished 93rd in DYAR and DVOA, which was dead last in both categories -- just like he nearly pulled off in 2015 as well.

Defenses

Finally, let's look at the defenses' ability to create failed completions, with a comparison to how these units fared in 2015.

Defenses: 2016 Failed Completions Compared to 2015
Rk Team Comp. Failed FC% 2015 Rk 2015 FC% Diff Rk
1 JAC 345 109 31.6% 19 24.6% 7.0% 2
2 NYG 369 112 30.4% 24 22.5% 7.9% 1
3 ARI 344 101 29.4% 8 27.7% 1.7% 8
4 HOU 319 91 28.5% 17 25.2% 3.3% 6
5 DEN 307 87 28.3% 1 32.3% -4.0% 27
6 DET 399 113 28.3% 5 28.0% 0.3% 12
7 PHI 333 94 28.2% 28 21.8% 6.4% 3
8 PIT 381 105 27.6% 29 21.6% 6.0% 4
9 LARM 378 103 27.2% 6 27.8% -0.6% 14
10 CIN 374 101 27.0% 3 28.1% -1.1% 17
11 SEA 330 89 27.0% 10 27.3% -0.3% 13
12 MIA 368 98 26.6% 27 22.0% 4.6% 5
13 NE 367 96 26.2% 14 25.8% 0.4% 11
14 BAL 368 93 25.3% 4 28.0% -2.7% 23
15 DAL 425 105 24.7% 9 27.6% -2.9% 24
16 NYJ 347 84 24.2% 15 25.4% -1.2% 18
Rk Team Comp. Failed FC% 2015 Rk 2015 FC% Diff Rk
17 CLE 347 83 23.9% 25 22.4% 1.5% 9
18 CHI 342 81 23.7% 20 24.5% -0.8% 16
19 CAR 411 97 23.6% 2 31.3% -7.7% 32
20 MIN 332 78 23.5% 13 26.2% -2.7% 22
21 TB 348 78 22.4% 31 20.1% 2.3% 7
22 SD 364 81 22.3% 7 27.7% -5.4% 29
23 OAK 329 73 22.2% 23 22.8% -0.6% 15
24 WAS 387 85 22.0% 22 23.2% -1.2% 19
25 ATL 422 90 21.3% 11 27.0% -5.7% 30
26 SF 324 69 21.3% 18 25.0% -3.7% 25
27 KC 349 72 20.6% 16 25.2% -4.6% 28
28 TEN 388 79 20.4% 32 19.4% 1.0% 10
29 BUF 307 62 20.2% 26 22.1% -1.9% 21
30 IND 378 76 20.1% 12 26.6% -6.5% 31
31 GB 370 73 19.7% 21 23.6% -3.9% 26
32 NO 378 71 18.8% 30 20.4% -1.6% 20

Last year, Denver and Carolina finished as the top two teams, meeting in the Super Bowl with the two best defenses. Denver was still a top-five unit in 2016, but Carolina suffered the largest drop in percentage points (7.7) of any defense at forcing failed completions. Luke Kuechly's concussion, which knocked the linebacker out of the lineup for six games, certainly played a factor, but the Panthers just never looked right this year.

Tom Coughlin might be interested to know that the team he just returned to as executive vice president (the Jaguars) and the team he coached in 2015 (the Giants) made the two biggest leaps in 2016, securing the top two spots in the process. Both defenses were hindered by their struggling offenses, but both swarmed to the catch well this season. Both units also were hyped heavily in the offseason after acquiring a lot of premium talent. The Giants successfully purchased the core of a new defense with Oliver Vernon, Damon Harrison, and Janoris Jenkins. The Jaguars signed Malik Jackson in free agency, got high draft pick Dante Fowler back from injury, and pulled off a draft haul of Jalen Ramsey and Myles Jack. While neither team experienced any playoff success like Denver and Carolina did the year before, both of these units are moving in the right direction. Now we just have to see if they can sustain their 2016 success, as the Jaguars have moved on from Gus Bradley at head coach to an offensive-minded Doug Marrone.

While we ended last year's study with a quasi-positive note about the Saints finishing 30th instead of 32nd, the defense went right back to 32nd even after firing defensive coordinator Rob Ryan. Look at the other teams ranked near the bottom. The Bills finished 29th and fired head coach Rex Ryan (and his twin brother Rob). Colts fans have been calling for Chuck Pagano's job for a few years now, and similar things can be said about defensive coordinator Dom Capers' job in Green Bay. Given that the Saints have continued to waste the remaining years of Brees' prime with 7-9 seasons because of a lack of a defense, at some point the sword has to fall on Sean Payton too. For as prolific as his offense has been, opposing offense have been just as prolific in each o f the last three years.

Posted by: Scott Kacsmar on 21 Feb 2017

19 comments, Last at 24 Feb 2017, 4:05pm by Vincent Verhei

Comments

1
by Bright Blue Shorts :: Wed, 02/22/2017 - 4:02am

re: "In 1989, 18.3 percent of all completions were failures. That rate last dipped below 23.1 percent in 2007, and peaked at 25.1 percent in 2015 before falling to 24.5 percent in 2016, the third-highest season on record."

When I first read this it sounded like the NFL has become worse at passing because of the focus on efficiency but then I realised that while 18.3% is lower, the overall completion rate was also lower than today.

Scott - is it easy to calculate what the "failed incompletion rate of all pass attempts" or therefore the "successful completion rate"? Would it make sense to?

3
by nat :: Wed, 02/22/2017 - 9:57am

Sure it would make sense. Here it is for this year:

Rk__Player_____________ Tm_ Att Scs SuccessRate
1___Matt Ryan__________ ATL 534 296 55.4%
2___Drew Brees_________ NO__667 367 55.0%
3___Dak Prescott________DAL 458 246 53.7%
4___Matt Barkley________CHI 216 115 53.2%
5___Tom Brady___________NE__429 223 52.0%
6___Kirk Cousins________WAS 603 313 51.9%
7___Aaron Rodgers______ GB__609 313 51.4%
8___Andrew Luck_________IND 542 278 51.3%
9___Brian Hoyer_________CHI 198 101 51.0%
10__Sam Bradford________MIN 550 279 50.7%
11__Andy Dalton_________CIN 561 283 50.4%
12__Matthew Stafford___ DET 591 298 50.4%
13__Jameis Winston______TB__567 281 49.6%
14__Carson Palmer______ ARI 596 294 49.3%
15__Alex Smith_________ KC__489 241 49.3%
16__Ben Roethlisberger__PIT 506 249 49.2%
17__Russell Wilson______SEA 545 267 49.0%
18__Philip Rivers______ SD__576 278 48.3%
19__Marcus Mariota______TEN 447 213 47.7%
20__Ryan Tannehill______MIA 387 183 47.3%
21__Cody Kessler________CLE 194 90__46.4%
22__Trevor Siemian______DEN 485 223 46.0%
23__Tyrod Taylor________BUF 434 197 45.4%
24__Derek Carr_________ OAK 558 252 45.2%
25__Colin Kaepernick___ SF__330 148 44.8%
26__Carson Wentz________PHI 605 271 44.8%
27__Eli Manning_________NYG 599 268 44.7%
28__Blake Bortles______ JAC 626 276 44.1%
29__Joe Flacco_________ BAL 664 292 44.0%
30__Brock Osweiler______HOU 508 223 43.9%
31__Cam Newton_________ CAR 506 221 43.7%
32__Ryan Fitzpatrick___ NYJ 403 173 42.9%
33__Case Keenum_________LAR 322 133 41.3%
34__Jared Goff_________ LAR 204 68__33.3%

Even better would be to have columns for the different kinds of failures: Completed with gain, no gain/incomplete, loss/sack, sack+fumble, interception. Those extra columns would show how - despite a high success rate - Barkley's propensity for the worst kind of failure makes him a weak starting QB. Stubbornly throwing into coverage beyond the sticks is exciting and occasionally works, but shows a low football IQ.

9
by Bright Blue Shorts :: Wed, 02/22/2017 - 4:10pm

Nice work.

If you were then to substitute Success% for Completion% in the standard NFL passer rating system, would the QB list look better? That would enable you to account for Barkley's weakness.

10
by nat :: Wed, 02/22/2017 - 4:42pm

I'm not sure whether that would work. The passer rating already has factors that account for yards per attempt. Trying to include "success" by this measure might amount to double-counting some yardage.

Besides, on first and second downs at least, the "success" criterion imposes a step-function on something that is in reality a smooth spectrum of value.

But you could try it if you want. You might need to come up with an offset value or scaling factor or both to bring success rate to the same scale as completion %.

11
by Bright Blue Shorts :: Wed, 02/22/2017 - 4:58pm

I was thinking that those Success% look a lot more like Completion% of the 70s which is when the passer rating was introduced. To my thinking, the Failed Completions have just replaced the downfield incompletions that would have occurred in earlier era's.

Jared Goff at 33% is perhaps a bit too much of an outlier but as I recall passer rating does top and tail extreme values to avoid giving too much credit.

I might have a little go tomorrow if I can get my Excel skills together.

12
by nat :: Wed, 02/22/2017 - 5:45pm

Stubbornly throwing into coverage beyond the sticks is exciting and occasionally works, but shows a low football IQ.

For the record, I'm just guessing at a diagnosis here. Chicago fans might have a better idea of why Barkley "succeeded" a lot, and had a high ALEX score, while also throwing a lot of interceptions and thus getting a low DVOA ranking. I'm assuming that those interceptions came from ill-conceived throws beyond the sticks. I could easily be wrong about that.

15
by xMRNUTTYx :: Thu, 02/23/2017 - 10:39am

Barkley's first pass attempt this year traveled further in the air than I think all but a couple of passes Brian Hoyer threw. It was a revelation...

Honestly, I'm surprised to see him at the top here. I don't think you could watch him play and not think about Hoyer's ultra-conservative decision making (he's a poor man's Alex Smith). So I was convinced Barkley just looked more daring in comparison. But apparently that wasn't the case. It's weird to think that they opened up the passing game when Barkley took over, but it certainly seemed that way- or he just decided "f*** it" and threw it downfield. I think the emergence of a running game via Jordan Howard greatly helped, too. I feel they had generally short 2nd and 3rd down conversion distances, allowing underneath routes to actually be at or beyond the sticks.

Regarding his picks, 10 of his 14 came in the final 3 games of the season (he started 6 total). I'd argue that most of those were due to bad reads or decisions on his part, but frankly, he was throwing to a lot of players who probably wouldn't be on many NFL rosters. And in those last two games, the defense had basically rolled over. He was playing from behind with huge deficits so risky throws were fine in my book.

Personally, if the Bears don't draft a start-now QB, I don't see why they don't have Barkley penciled in as the starter in 2017. He'd be cheap, competent, and be a perfectly fine stop-gap until they find a franchise player to replace him.

16
by Vincent Verhei :: Thu, 02/23/2017 - 4:45pm

Just because I love looking stuff like this up...

Barkley's first pass of the season came off the bench in Week 7. It only traveled 10 yards downfield. However, he did have passes that traveled 24 and 49 yards downfield in a little more than a half of football.

However, his first pass as a starter was in Week 12, and that traveled 33 yards downfield. Only six of Hoyer's passes all season went so far.

The Bears, as a team, had 45 passes travel more than 25 yards past the line of scrimmage. Barkley had 19, out of 216 passes overall. Cutler had 15 out of 137. Hoyer had 11 out of 200.

17
by Bright Blue Shorts :: Fri, 02/24/2017 - 5:49am

However ... Barkley's pass did travel about 20-yds through the air as he threw it from 10-yds behind the line of scrimmage ... roughly his own 38 to GB's 41 ...

Can see why that play would get your hopes up after six weeks of conservative dink&dunk. QB drops, runs deeper from pressure, eventually thows a pass and it's complete for a 1st down. It's all about context!

19
by Vincent Verhei :: Fri, 02/24/2017 - 4:05pm

However ... Barkley's pass did travel about 20-yds through the air as he threw it from 10-yds behind the line of scrimmage ... roughly his own 38 to GB's 41 ...

Ha! Hadn't thought of it that way.

2
by Bright Blue Shorts :: Wed, 02/22/2017 - 4:15am

re: the table of 100+ failed completions being almost exclusively since 2010, that also tallies with how the number of 4,000yd passers has ballooned since the mid-2000s.

Up until 2005, it was never more than 5 players in a season passing for 4,000 yds.

The past few years it's been 9 to 11 QBs - about a third of the league achieving it including players like Josh Freeman!

4
by Otis Taylor89 :: Wed, 02/22/2017 - 10:08am

"With the Ravens losing Steve Smith to retirement, a new weapon could be in order for this offense."
I know just the guy - Tavon Austin (Mind Blown).

13
by Jim C. :: Wed, 02/22/2017 - 6:13pm

You do know that Austin is a Baltimore native, don't you?

5
by Otis Taylor89 :: Wed, 02/22/2017 - 10:13am

I seriously don't know what to think of Matt Barkley as I saw him play really well for a good stretch and then he kind of came unglued near the end. Do we give him the benefit of the doubt since he hadn't played much prior to this year, except preseason and he didn't exactly have Jerry Rice circa 1989 to throw to? There just seems to be so many worse QBs that are starting or projected to start out there that Barkley may be an OK option.

6
by MilkmanDanimal :: Wed, 02/22/2017 - 10:52am

Mike Evans. 96 catches. FIVE failed completions.

My kingdom for a real #2 WR. And by "my kingdom" I mean "somehow Corey Davis plummets to #19 so we can draft him without trading up also I'd like a magic pony".

7
by Guest789 :: Wed, 02/22/2017 - 11:22am

Nah, if the Bucs trade up it'll be because they need a kicker to replace Aguayo. They can probably get the best one in the draft in the first round.

8
by MilkmanDanimal :: Wed, 02/22/2017 - 12:33pm

Well, Corey Davis can probably kick about as well as Aguayo did in the first five or six games of the season, so we'll just double-dip him in a kicker role.

14
by Dr. Bill :: Wed, 02/22/2017 - 9:34pm

I really, really wish the D got some credit in this and it was "Foiled Completions."

18
by jtr :: Fri, 02/24/2017 - 12:20pm

I'm not exactly sure how to test this but is be curious to see FO try to tease out how much of this is on the QB and how much is on the OC. This year we only had 2 teams qualify multiple passers, and they showed opposite effects. Hoyer and Barkley are quite far apart on the charts, suggesting QB responsibility, while the two QB Rams are both at the bottom, suggesting OC responsibility (consistent with the craven offensive approach of the entire Fisher era). It would be interesting to see how closely QB teammates have correlated with one another during the years FO has been keeping these stats.