2021 Failed Completions: The Bell Tolls for Roethlisberger

Pittsburgh Steelers QB Ben Roethlisberger
Pittsburgh Steelers QB Ben Roethlisberger
Photo: USA Today Sports Images

NFL Offseason - Ben Roethlisberger has finally joined Drew Brees and Philip Rivers on the quarterback farm upstate. And not a moment too soon: Big Ben just couldn't hunt anymore. Once one of the NFL's most dangerous deep passers, Roethlisberger led the NFL in failed completions in 2021. It was his second straight year as the failed completion king.

Unlike two years ago, when he barely clawed his way to the title, Big Ben ran victory laps against his fellow dink-and-dunk stars in 2021. With 130 failed receptions to his name, Roethlisberger takes the crown by 20, the biggest gap since Joe Flacco's legendary 2016 season. It was clearly past time for Roethlisberger to hang 'em up, and we have been getting here for some time—Roethlisberger had over 100 failed completions in each of his last three healthy seasons.

A failed completion is any completed pass that fails to gain 45% of needed yards on first down, 60% on second down, or 100% on third or fourth down. You can see last year's study on the subject here.

As somewhat expected, 2020's relatively low number of failed completions was fluky, sparked by the lack of an offseason, empty stadiums around the league, and a strange drop in holding and interference calls. The league's 23.9% failed completion rate in 2020 was the lowest we had seen since 2012, and entirely out of character with the regular rise of failed completions we have seen over the past 40 years.

2021 was a significant return to form. The 3,131 failed completions of 2021 were an all-time record, although that's partially due to the addition of a 17th game last season. The 25.8% failed completion rate still sticks in the top five all-time, joining 2015, 2017, 2018, and 2019. Normal service has resumed; even COVID couldn't keep the dumpoff down for long.

Of course, not every failed completion is created equal, but for this article we make things binary, simply summing up successes and failures. A 7-yard completion on third-and-10 is better than a 2-yard completion, especially when you get into field goal range. That's reflected in DVOA, but not here—for the purposes of this article, a failure is a failure.


In the following table, the 34 qualified quarterbacks of 2021 are ranked by ascending failed completion rate (FC%). We also included failed completions as a percentage of attempts (very little change in the rankings), as well as the average ALEX (all downs) for the season. "Passes" here refers to all passing plays, including sacks.

Quarterbacks, Failed Completions, 2021
Rk Player Team Pass Cmp Failed FC% FC%
1 Patrick Mahomes KC 691 436 79 18.1% 11.4% 2 -1.1 22
2 Baker Mayfield CLE 469 253 52 20.6% 11.1% 1 -0.2 7
3 Justin Herbert LAC 713 443 93 21.0% 13.0% 4 -0.8 17
4 Matthew Stafford LAR 637 404 86 21.3% 13.5% 8 0.2 4
5 Tom Brady TB 750 485 105 21.6% 14.0% 9 -0.1 6
6 Josh Allen BUF 680 409 89 21.8% 13.1% 5 -0.3 10
7 Jimmy Garoppolo SF 472 301 68 22.6% 14.4% 10 -1.6 29
8 Lamar Jackson BAL 419 246 56 22.8% 13.4% 6 0.4 3
9 Trevor Lawrence JAX 638 359 82 22.8% 12.9% 3 -0.8 16
10 Dak Prescott DAL 639 410 96 23.4% 15.0% 13 -0.6 13
11 Tua Tagovailoa MIA 411 263 65 24.7% 15.8% 16 -1.2 25
12 Aaron Rodgers GB 570 366 91 24.9% 16.0% 17 -0.3 11
13 Kyler Murray ARI 519 333 85 25.5% 16.4% 23 -0.7 14
14 Jacoby Brissett MIA 248 141 36 25.5% 14.5% 11 -0.2 9
15 Mac Jones NE 561 352 90 25.6% 16.0% 19 -0.7 15
16 Derek Carr LV 676 428 110 25.7% 16.3% 22 -0.2 8
17 Jalen Hurts PHI 461 265 70 26.4% 15.2% 14 0.0 5
18 Jared Goff DET 531 332 89 26.8% 16.8% 26 -2.0 33
19 Teddy Bridgewater DEN 463 285 77 27.0% 16.6% 24 -1.0 19
20 Joe Burrow CIN 576 366 99 27.0% 17.2% 27 -0.5 12
21 Zach Wilson NYJ 433 213 58 27.2% 13.4% 7 -1.0 21
22 Carson Wentz IND 561 322 88 27.3% 15.7% 15 -0.9 18
23 Russell Wilson SEA 442 259 71 27.4% 16.1% 21 1.4 1
24 Ryan Tannehill TEN 587 357 98 27.5% 16.7% 25 -1.1 23
25 Andy Dalton CHI 256 149 41 27.5% 16.0% 18 -1.7 31
26 Justin Fields CHI 309 159 45 28.3% 14.6% 12 0.6 2
27 Matt Ryan ATL 610 375 107 28.5% 17.5% 29 -1.2 26
28 Taylor Heinicke WAS 537 321 93 29.0% 17.3% 28 -1.2 27
29 Sam Darnold CAR 442 243 71 29.2% 16.1% 20 -1.8 32
30 Kirk Cousins MIN 590 372 110 29.6% 18.6% 31 -1.2 24
31 Davis Mills HOU 430 263 80 30.4% 18.6% 30 -1.4 28
32 Daniel Jones NYG 393 232 75 32.3% 19.1% 32 -2.1 34
33 Tyler Huntley BAL 206 122 40 32.8% 19.4% 33 -1.0 20
34 Ben Roethlisberger PIT 652 390 130 33.3% 19.9% 34 -1.7 30

In 2018 and 2020, Ben Roethlisberger's failed completion rate hovered around 25%. That's not ideal, of course, and it was trending in the wrong direction, but you can work with that. The bottom just fell out for Roethlisberger last season; a full third of his completions were failures as a combination of no faith in his offensive line and no strength in his arm meant that Roethlisberger's average depth of target fell to a career low of 6.5 yards and his time to throw was a league-fastest 2.38 seconds. Roethlisberger threw escape routes and dumpoffs like his life depended on it. Najee Harris had 94 targets with an aDOT of -0.14 yards, 30 more targets than the next-most targeted player with a negative aDOT. A younger Big Ben would have stood in the pocket despite the poor pass protection, shrugged off a few hits, and found targets deep downfield. That Big Ben retired about five years ago, and last year's model just didn't have anything left in the tank.

Add a quarterback with no arm strength to a porous offensive line, a highly drafted running back, and a 17th regular-season game, and you have the recipe for a run at the single-season failed completions record. Big Ben put his all into it, but he couldn't quite seal the deal.

Most Failed Completions Since 1983
Rk Player Year Team Cmp Failed FC%
1 Joe Flacco 2016 BAL 436 144 33.0%
2 Ben Roethlisberger 2021 PIT 390 130 33.3%
3 Joe Flacco 2017 BAL 352 127 36.1%
4 Kirk Cousins 2018 MIN 425 120 28.2%
5 Matt Ryan 2013 ATL 439 120 27.3%
6 Sam Bradford 2016 MIN 395 116 29.4%
7 Eli Manning 2018 NYG 380 115 30.3%
8 Drew Brees 2015 NO 429 113 26.3%
9 Derek Carr 2014 OAK 348 112 32.2%
10 Joe Flacco 2013 BAL 362 111 30.7%
11 Kirk Cousins 2021 MIN 372 110 29.6%
12 Eli Manning 2016 NYG 378 110 29.1%
13 Drew Brees 2017 NO 387 110 28.4%
14 Derek Carr 2021 LV 428 110 25.7%
15 Ben Roethlisberger 2018 PIT 452 110 24.3%
16 Ben Roethlisberger 2020 PIT 399 109 27.3%
17 Drew Brees 2010 NO 449 109 24.3%
18 Carson Wentz 2016 PHI 379 108 28.5%
19 Matt Ryan 2019 ATL 408 108 26.5%
20 Matt Ryan 2021 ATL 375 107 28.5%
21 Drew Brees 2013 NO 447 107 23.9%
22 Teddy Bridgewater 2020 CAR 340 106 31.2%
23 Justin Herbert 2020 LAC 396 106 26.8%
24 Derek Carr 2016 OAK 357 105 29.4%
25 Ryan Tannehill 2015 MIA 364 105 28.8%
26 Matthew Stafford 2018 DET 367 105 28.6%
27 Peyton Manning 2002 IND 392 105 26.8%
28 Philip Rivers 2015 SD 438 105 24.0%
29 Drew Brees 2016 NO 472 105 22.2%
30 Tom Brady 2021 TB 485 105 21.6%

Joe Flacco remains the failed completion king. He'd still have the top two slots if it wasn't for the 17th game; while Roethlisberger put up his 130 failed completions in just 16 games, he wouldn't have had the opportunity to play that final game if it weren't for Week 18. Flacco, of course, turned the failed completion into an artform all its own. He and Drew Brees have long been the two tentpoles of this stat with Flacco showing how hollow a quarterback's completion percentage can be if every pass is a checkdown, and Brees showing that a bunch of ineffective checkdowns doesn't hurt too much if you mix in plenty of successful passes alongside them. Pre-arm injury, Roethlisberger was more of a Brees-type failed completionist, firmly in the middle of the pack in failed completion rate but with a high enough volume to appear near the top of the table anyway. But over the last couple of seasons, Roethlisberger went full Flacco. And when you go full Flacco, it's time to hang up your boots.

This is normally where we turn to talk about the quarterback with the worst failed completion rate last season … but that's Ben Roethlisberger too. That has never happened in what has now been a decade of writing these stats up, and I couldn't find a case of it before that either. Typically, the quarterback with the worst failed completion rate is a player so bad he gets benched before the season is over, or who is only playing as an out-of-their-depths injury replacement. Your last three leaders were Dwayne Haskins, Mason Rudolph, and Nick Foles, none of whom had more than 305 pass attempts—borderline qualifiers. But Roethlisberger managed to hold off Tyler Huntley, Daniel Jones, and Davis Mills, all of whom fit the profile of the failed completion rate leader far more than Roethlisberger and his 652 dropbacks. Roethlisberger's 33.3% failed completion rate is the second-highest ever for a player with at least 100 failed completions. Seasons like these don't grow on trees and deserve to be immortalized.

Four other passers crossed the 100-failed completion threshold this year. Tom Brady only got there because of the 17th game, and no one cares about 105 failed completions when you're also leading the league in successful completions. Derek Carr scoring high is nothing new as he has never been super-comfortable going deep, but he has seen his failed completion rate drop from his early years. Matt Ryan is rapidly approaching the end; he's not in Big Ben territory yet, but you can here it tolling from here.

And then there's Kirk Cousins. In 2020, Cousins had the eighth-best failed completion rate, sitting at 19.2%. But last year, the bottom fell out, down to the fifth-worst rate of 29.6%. His raw totals rose from 67 to 110. It was, by far, the biggest drop-off between 2020 and 2021. Cousins had the worst game of the year with 15 failed completions in the season opener against Cincinnati, and things never really turned around. His ALEX dropped from -0.7 to -1.2, but it's not like he pulled a Roethlisberger and started throwing running back dumpoffs every other play. A large chunk of this comes from conservative first-down play calling. Cousins jumped from 15 failed completions on first downs in 2020 to 42 last year, and his first-down DVOA dropped from 23.7% to 1.5%. In 2020, about 40% of Cousins' first-down passes were below 5 air yards—in other words, so short that some receiver YAC was needed to turn them into successes. Last season, that jumped to 52%. That doesn't explain it all away, but Klint Kubiak's tendency to call dumpoff pass plays on first down, as well as Cousins' eagerness to find those dumpoffs, is a huge part of this increase. Vikings fans hope new head coach Kevin O'Connell will be a pleasant breath of fresh air strategically, though it should make Vikings fans a little nervous about Cousins' latest big-money extension.

Cousins and Roethlisberger were two of the five quarterbacks to see their failed completion percentage jump five points between 2020 and 2021. Josh Allen is another; his 2020 season went about as well as it possibly could have, so some slight regression was to be expected. Ryan Tannehill stumbled some, in part because everyone in Tennessee was hurt at one point or another. And Russell Wilson rounds out your group of five. Even Healthy Russell Wilson (Weeks 1-5 and 16-18) had a failed completion rate of 26.2%, so this isn't all due to the finger injury, either.

The only quarterback to see his failed completion rate improve by five percentage points was Justin Herbert, who went from 26.8% as a rookie to 21.0% last season—from 28th to third! Herbert had finished last among the three 2020 rookies, and we figured having Joe Lombardi calling the shots would keep him high on the failed completion leaderboards with lots of dumpoffs to Austin Ekeler. Herbert had plenty of those, sure, but his arm talent has allowed him to throw the ball deeper more effectively than Lombardi's last charge, the Aging Remains of Drew Brees. It doesn't matter how many failed completions you have if you're mixing them in with plenty of successful completions.

And, on that note, we can run our table of successful completion percentage—the percentage of all pass attempts that end as successful completions. This is where we count all failed completions as incomplete passes, removing some of the empty passing calories from completion percentage around the league:

Successful Completion Percentage, 2021
Player Team Att Comp. Cmp% Rk Failed Success% Rk
Patrick Mahomes KC 656 436 66.5% 18 79 54.4% 1
Jimmy Garoppolo SF 437 301 68.9% 5 68 53.3% 2
Tom Brady TB 717 485 67.6% 10 105 53.0% 3
Matthew Stafford LAR 601 404 67.2% 14 86 52.9% 4
Dak Prescott DAL 596 410 68.8% 6 96 52.7% 5
Justin Herbert LAC 671 443 66.0% 19 93 52.2% 6
Aaron Rodgers GB 528 366 69.3% 3 91 52.1% 7
Kyler Murray ARI 480 333 69.4% 2 85 51.7% 8
Joe Burrow CIN 518 366 70.7% 1 99 51.5% 9
Derek Carr LV 621 428 68.9% 4 110 51.2% 10
Tua Tagovailoa MIA 387 263 68.0% 8 65 51.2% 11
Mac Jones NE 519 352 67.8% 9 90 50.5% 12
Lamar Jackson BAL 378 246 65.1% 21 56 50.3% 13
Jared Goff DET 488 332 68.0% 7 89 49.8% 14
Josh Allen BUF 644 409 63.5% 25 89 49.7% 15
Ryan Tannehill MIA 529 357 67.5% 11 98 49.0% 16
Teddy Bridgewater DEN 425 285 67.1% 16 77 48.9% 17
Baker Mayfield CLE 415 253 61.0% 30 52 48.4% 18
Matt Ryan ATL 556 375 67.4% 12 107 48.2% 19
Kirk Cousins MIN 554 372 67.1% 15 110 47.3% 20
Jacoby Brissett MIA 223 141 63.2% 27 36 47.1% 21
Russell Wilson SEA 400 259 64.8% 22 71 47.0% 22
Davis Mills HOU 391 263 67.3% 13 80 46.8% 23
Taylor Heinicke WAS 491 321 65.4% 20 93 46.4% 24
Trevor Lawrence JAX 599 359 59.9% 32 82 46.2% 25
Andy Dalton CHI 235 149 63.4% 26 41 46.0% 26
Carson Wentz IND 515 322 62.5% 28 88 45.4% 27
Jalen Hurts PHI 430 265 61.6% 29 70 45.3% 28
Tyler Huntley BAL 183 122 66.7% 17 40 44.8% 29
Daniel Jones NYG 359 232 64.6% 24 75 43.7% 30
Ben Roethlisberger PIT 603 390 64.7% 23 130 43.1% 31
Sam Darnold CAR 404 243 60.1% 31 71 42.6% 32
Justin Fields CHI 268 159 59.3% 33 45 42.5% 33
Zach Wilson NYJ 381 213 55.9% 34 58 40.7% 34

Patrick Mahomes takes his first of what is likely to be many successful completion crowns. No one blinks twice when good players appear at the top of good stats. You may, however, be surprised to see Jimmy Garoppolo in second place, such is his reputation. When Garoppolo dropped back to pass, it was more often than not a good thing, no matter what the highlight reel may tell you. His 169.9% DVOA on successful completions was fifth highest in the league. You may think, "oh, these are all going to be little dumpoffs to Deebo Samuel that gained 97 yards of YAC and look good on the stat sheet," and well, you're not wrong. They're not really dumpoffs; Garoppolo's 7.6-yard aDOT on successful completions ranked 16th in the league, pretty much bang-on average. But then he benefitted from 7.5 YAC, the most in the league by nearly a half yard. Combine a high successful completion percentage, a solid depth of target, and insane YAC, and you see why Garoppolo ranked fifth in passing DVOA this year. A lot of that is the Shanahan system, of course, but there's skill in being able to hit George Kittle or Deebo Samuel in stride and allow them to earn that YAC. I suppose we'll get more clarity as to how much was the quarterback and how much was the coach next year, presuming literally anyone trades for Garoppolo. Anybody? Anybody? Tepper?

Ben Roethlisberger does finally get off the bottom of a table when we switch to successful completions, making way for the rookies. Zach Wilson has a solid lead over Justin Fields for the lowest successful completion rate among the first-year passers, hovering down at 40.7%. Trevor Lawrence falls into the bottom 10, beaten out by Davis Mills. Not a good year in general for the first-round passers … except for Mac Jones, ranked 12th at 50.5%. That's a very impressive debut for the rookie, and he improved as the season went along—he was at 46.5% after his first month and 47.7% halfway through the season, so despite some December rough patches, there's a positive trend going on there. I wouldn't push the panic button yet if I were the Jets or Bears, but Round 1 clearly goes to Mac Jones.

Mahomes, Herbert, Baker Mayfield, Matthew Stafford, and Josh Allen all jump at least 10 spots in the rankings when you go from raw completion percentage to successful completion percentage. Allen squeaks in to the list despite seeing his successful completion percentage drop 8.2% from 2020, which puts him in Cousins/Wilson territory in terms of year-over-year performance. Have we mentioned how good Allen's 2020 was?

Davis Mills and Tyler Huntley drop at least 10 spots flipping to successful completion percentage; there's a lot of empty calories in the "Houston has Found Their Guy!" narrative.


What about the receivers on the other end of those failed completions? It's worth taking a look at that, even though appearances here generally have more to do with usage and scheme than a receiver's individual talents. We exclude running backs from these tables; they would otherwise dominate due to their roles on checkdowns and emergency outlets. For the record, however, Najee Harris led all players with 33 failed receptions, as those Big Ben wobblers had to go somewhere. Leonard Fournette was the only other running back above 25.

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
Diontae Johnson PIT 30 Tee Higgins CIN 74 4 5.4% Ray-Ray McCloud PIT 38 16 42.1%
Cooper Kupp LAR 29 Van Jefferson LAR 50 3 6.0% Noah Fant DEN 68 26 38.2%
Marquise Brown BAL 28 Hunter Henry NE 50 3 6.0% Rondale Moore ARI 54 20 37.0%
Hunter Renfrow LV 27 Byron Pringle KC 42 3 7.1% Zach Pascal IND 38 14 36.8%
Noah Fant DEN 26 Mike Evans TB 74 6 8.1% Chester Rogers TEN 30 11 36.7%
Keenan Allen LAC 24 Michael Gallup DAL 35 3 8.6% Ricky Seals-Jones WAS 30 11 36.7%
Davante Adams GB 23 Robert Woods LAR 45 4 8.9% Devin Duvernay BAL 33 12 36.4%
Jaylen Waddle MIA 23 Rob Gronkowski TB 55 5 9.1% Jalen Reagor PHI 33 12 36.4%
Cole Beasley BUF 22 DeAndre Hopkins ARI 42 4 9.5% Sterling Shepard NYG 36 13 36.1%
Brandin Cooks HOU 21 Emmanuel Sanders BUF 42 4 9.5% Geoff Swaim TEN 31 11 35.5%
Tyler Conklin MIN 21 Jaylen Guyton LAC 31 3 9.7% Quez Watkins PHI 43 15 34.9%
Adam Thielen MIN 20 DeVante Parker MIA 40 4 10.0% Evan Engram NYG 46 16 34.8%
Amon-Ra St. Brown DET 20 Allen Lazard GB 40 4 10.0% C.J. Uzomah CIN 49 17 34.7%
Rondale Moore ARI 20 Austin Hooper CLE 38 4 10.5% Tyler Conklin MIN 61 21 34.4%
Min. 30 receptions

Being on the failed receptions table is not, in and of itself, a bad thing. Cooper Kupp, Keenan Allen, and Davante Adams all make return appearances from last year's list; they rack up a lot of failed receptions because they're the guys their quarterbacks look for when they're in trouble. If you catch the ball 145 times, not every single one of them is going to be a winner, and that's OK.

Diontae Johnson also returns from 2020's list, but I think we can blame a lot of that on Roethlisberger's dead arm; it's conceivable that one day, Pittsburgh may have a quarterback who can throw the ball more than 2 yards beyond the line of scrimmage. That also partially explains Ray-Ray McCloud's appearance atop the failed reception rate leaderboard, but I'd give McCloud a larger portion of the blame than I'd give Johnson. Johnson is a solid receiver hamstrung by a quarterback who can't throw. McCloud is a punt returner and below-average slot receiver who couldn't produce with the ball in his hands; maybe he doesn't finish dead last in receiving DVOA with an average quarterback targeting him, but a 42.1% failed reception rate basically meant he was juking nobody. When you're sitting on a 4.3-yard aDOT on your receptions, averaging just 2.9 YAC is not going to cut it.

Rondale Moore at least averaged 8.1 YAC, but seeing as how he was the only qualified wide receiver in the league with a negative depth of target on his receptions, it's no surprise to see him on the failed reception list. He was at least able to do more with the ball in his hands than McCloud, Devin Duvernay, or Jalen Reagor, all of whom also were near the bottom of the league in aDOT and show up on the failed reception table. It feels like Moore is salvageable with some better play design; not so much the other three.

We should congratulate Allan Lazard for being the only player to repeat from last year's lowest failed reception rate table. He did double his failed receptions, but going from two to four is not precisely the end of the world. I suppose we should also congratulate Evan Engram for being the only player to repeat from last year's highest failed reception rate table; he did see his percentage drop from 38.1% to 34.8%, but eef. And this man was a coveted free agent! That's betting a lot on Daniel Jones just being bad … which, I suppose, is fair enough.


Finally, let's look at the defenses' ability to create failed completions, with a comparison to how these units fared in 2020.

Defenses: 2021 Failed Completions Compared to 2010
Rk Team Comp Failed FC% 2020 Rk 2020 FC% Diff Rk
1 CLE 367 121 33.0% 23 22.4% 10.6% 1
2 SF 372 122 32.8% 7 26.4% 6.4% 4
3 DAL 364 108 29.7% 11 25.6% 4.1% 9
4 TB 445 131 29.4% 3 27.9% 1.5% 18
5 CAR 337 99 29.4% 9 26.1% 3.3% 13
6 CIN 420 119 28.3% 26 21.5% 6.8% 3
7 CHI 314 88 28.0% 8 26.3% 1.7% 17
8 BUF 297 83 27.9% 18 23.3% 4.6% 7
9 NO 369 103 27.9% 14 24.3% 3.6% 10
10 LV 400 110 27.5% 24 21.5% 6.0% 5
11 PHI 409 112 27.4% 6 26.9% 0.5% 22
12 TEN 395 106 26.8% 31 19.4% 7.4% 2
13 NYG 402 106 26.4% 20 22.8% 3.6% 11
14 ATL 391 103 26.3% 12 25.4% 0.9% 21
15 MIA 374 98 26.2% 22 22.7% 3.5% 12
16 KC 401 105 26.2% 28 20.6% 5.6% 6
17 LAR 416 106 25.5% 2 28.8% -3.3% 30
18 GB 379 96 25.3% 17 23.4% 1.9% 16
19 JAX 377 95 25.2% 16 23.8% 1.4% 19
20 PIT 355 89 25.1% 10 25.8% -0.7% 25
21 LAC 357 88 24.6% 4 27.1% -2.5% 28
22 ARI 367 89 24.3% 19 23.3% 1.0% 20
23 MIN 401 97 24.2% 27 21.0% 3.2% 14
24 BAL 397 95 23.9% 15 23.9% 0.0% 24
25 DET 359 83 23.1% 32 18.8% 4.3% 8
26 DEN 341 78 22.9% 5 27.0% -4.1% 31
27 NYJ 401 91 22.7% 30 19.8% 2.9% 15
28 WAS 400 90 22.5% 1 29.1% -6.6% 32
29 SEA 443 97 21.9% 13 24.9% -3.0% 29
30 NE 319 69 21.6% 25 21.5% 0.1% 23
31 IND 390 83 21.3% 21 22.8% -1.5% 27
32 HOU 363 71 19.6% 29 20.4% -0.8% 26

Defensive numbers are far less sticky year-to-year than quarterback stats. A lot of this depends on the type and quality of the quarterbacks on your schedule, and so the year-to-year correlation is typically down in the 0.25 range. 2021 was even a little below that because of the oddness of 2020.

Nearly a fifth of Cleveland's forced failed completions came courtesy of one Ben Roethlisberger. Add in matchups against Kirk Cousins, Derek Carr, Tim Boyle, Tyler Huntley, and Brandon Allen and, well, you begin to understand. Those six passers account for 59 of Cleveland's 121 failed completions forced. Credit where credit is due when it comes to forcing those dumpoffs, and we should note that Cleveland forced Aaron Rodgers to 12 and Lamar Jackson to 10, so it's not all the strength of schedule here, but the Browns did not exactly face a murderer's row of deep threats this season.

There's usually a solid correlation between a pass rush and a high number of failed completions as offenses try to keep their quarterbacks upright. Cleveland, San Francisco, and Carolina were all in the top five in pass rush win rate, with Tampa Bay and Dallas finishing in the top 12—that explains the top of the table quite nicely. Washington had enough injuries and regression on their heralded defensive line to explain a large chunk of why the fell from the top slot last year down to 28th this year, and the rest of the bottom five is composed of below-average pass rush units. You have to go the Jets in 27th to find the first team ranked in the top 16 in pass rush win rate on this chart; it turns out it doesn't matter how much pressure you get if you have no cornerbacks whatsoever.

The Houston Texans round out the bottom, as they do in so many of these stats. They were able to force eight-failed completion days from Trevor Lawrence and Ryan Tannehill, but that's fairly faint praise. Baker Mayfield, Josh Allen, Jacoby Brissett, Kyler Murray, Russell Wilson, and Carson Wentz (twice!) were able to have games with three or fewer failed completions. It's a good thing the Texans went through a coaching change this offseason, and didn't do anything foolish like promote their defensive coordinator to the center seat.


23 comments, Last at 30 Mar 2022, 3:52pm

1 Minor editorial

Is it unusual for a player to be so low in ALEX, as Mahomes is at 22, yet still bottom of failed completion %. It might be a reflection of scheme, but even in Alex Smith's best KC season (2017) he still ranked low in ALEX and high in failed completion %. I checked back through 2012 and couldn't find a comparable type of season.

Matt Ryan is rapidly approaching the end; he's not in Big Ben territory yet, but you can here it tolling from here.

Think this should be hear :)

5 Likely due to the fact that,…

In reply to by HitchikersPie

Likely due to the fact that, against the chiefs, defenses sat deep in the cover 2 shell, giving up the underneath routes and allowing for additional YAC.

4 Considering my last…

Considering my last impression of Herbert was a QB who could not accomplish a completion of any kind unless it was 4th-long, it seems impossible to replicate the Chargers ALEX or FC% rates.

3 You don't need to wait for…

You don't need to wait for next year -- Goff and Stafford switched teams in 2021.

They were basically the same guy in 2020 (12th and 13th) and if you averaged 2018 and 2019 (2018 was aberrant for Stafford). There was a 5 point swing this year.

There seems to be some QB effect, but a lot of team effect -- a mixture of surrounding talent and offensive habit.

Arians teams are mad bombers. They've been top-5 in ALEX and FC% since 2018, excepting last year when Brady was 15 in FC%. Chicago, Miami, Minnesota, The News York, Washington, and in recent years Pittsburgh have tended to be high in FC%. There are some exceptions -- Fitz, an inveterate mad bomber, has defied team trends in Miami.

Chicago has done this to Trubisky, Foles, Dalton, and Fields. (2020 was a weird Trubisky year) Fields tried defying the odds -- Fields and Dalton have about the same FC%, but their ALEX is completely different. I wonder if that happened with Russell Wilson and Geno Smith, too.

Ryan has swung up and down over his career. I can't help but wonder if receiver talent isn't deeply entangled with this. I also observe that the count of failed completions stat is kind of useless -- there sure are a lot of HOFers in that list. Considering the moral judgment aspect FO imparts to FC% and ALEX, I can't help but observe it's not apparently justified. There are a lot of bad high-ALEX QBs and a lot of good low-ALEX QBs. I'm not confident either correlate to DVOA or DYAR all that strongly, if at all.

22 There are some exceptions --…

There are some exceptions -- Fitz, an inveterate mad bomber, has defied team trends in Miami.

Oddly enough, Tua rated comparatively well this year in both FC and SC despite a low ALEX and ultra low YAC. No idea how that happened.

6 right about ALEX and FC

They don't correlate well with good QBing. A failed completion is better than an incompletion generally, even an incompletion that is thrown downfield. A failed completion represents an accurate throw that between defense recognition, receiver agility, and luck, doesn't result in an overall improved situation. I know this is pedantic and everyone here knows it, but still, better to have failed completions than sacks, incompletions, or interceptions.

I like the success % though, it correlates, this year at least, largely with our perceptions of QB play and rank order, with the possible exception of Garropolo and Josh Allen. Although some good QBs may generate more FCs because they are accurate in their checkdowns, they should also generate more SCs.

As for ALEX, I don't know, it doesn't show much correlation year over year, weird combination of scheme, defensive opposition, tendencies of your WR1. it originated, if memory serves me correctly, to denigrate Alex Smith and his tendency to not throw the ball downfield even on third and long, but I don't give it much credence as a predictive of, well, anything...

7 A failed completion…

A failed completion represents an accurate throw that between defense recognition, receiver agility, and luck, doesn't result in an overall improved situation.

It's not entirely clear from the commentary, but I think a failed completion can still accrue positive DVOA, and thus result in an objectively improved situation. (As opposed to only the relative improvement of a small gain versus zero or negative gain)

This sets EPA and WPA aside entirely.

8 Yes, a failed completion can…

Yes, a failed completion can accrue positive DVOA, although it's obviously somewhat uncommon.  About 15% of failed completions end up with positive DVOA each season.

The best failed completion in terms of DVOA last season belongs to Jalen Hurts against the Cowboys.  He hit Miles Sanders for a 19-yard gain on 3rd-and-24, moving the ball just across midfield.  The Eagles opted to punt on 4th-and-5 rather than go for it with less than a minute left in the half.

The best failed completion in terms of DYAR last season, oddly, came from the same quarter of the same game.  Facing 1st-and-goal from the 19, Dak Prescott hit Amari Cooper for 6 yards, setting up 2nd-and-13.  They ended up scoring two plays later, though that was probably more due to Zeke Elliott gaining 10 yards on the second-down play than it was the little screen and shimmy that picked up half of the penalty yards. 

13 I believe that goes to Teddy…

I believe that goes to Teddy Bridgewater against Washington.  On 3rd-and-1, he passed to Melvin Gordon five yards deep in the backfield; Gordon managed to go forward one yard, but that's still a loss of four, and Denver opted to punt from the Washington 49.

Tyrod Taylor came dang close, though, hitting Chris Moore for a four-yard loss on a 3rd-and-1 against the Jets.  And in third place it's...Teddy Bridgewater again, with a nine-yard loss on a pass to Javonte Williams on a 2nd-and-5 against the Eagles.

Jalen Hurts, Taylor Heinicke and Kirk Cousins tied for the lead with 10 failed completions worth at least -10 DYAR, with Ryan Tannehill and Carson Wentz just behind at nine a piece.

Mahomes had eight, tied with Justin Herbert, Tom Brady, Sam Darnold, Trevor Lawrence, Josh Allen, Matt Ryan and Derek Carr, which may well be the most random collection of passers you could put together.

9 A failed completion can also…

A failed completion can also be an inaccurate throw that takes a receiver out of his stride and prevents them from picking up any YAC.  That's the ultimate difference between the Jimmy Garoppolo 49ers' offense and, say, the Nick Mullens' version. 

I would agree that the raw total of failed completions doesn't correlate with overall quarterback success.  There's a reason Drew Brees pops up all over the total failed completions list; a combination of offensive philosophy and sheer volume.  It is interesting and useful to look at those top failed completion players and try to figure out if it's just the team throwing short or a lack of arm strength or checkdownitis or whathaveyou, but a high level of failed completions does not mean a quarterback isn't good.

Failed completion percentage on the other hand, does a better job of sorting good passers from bad.  Yes, sometimes a good throw ends up as a failed completion due to factors outside of a quarterback's control, but when you start getting up to 30% of your completions going nowhere, you can start pointing fingers.  We last ran the table of worst failed completion percentage back in 2018, and that was much more a list of disaster passers -- Nick Foles, Anthony Wright, Tim Couch, Brady Quinn, Mitch Trubisky, Bruce Gradkowski, Joe Flacco, Trent Edwards and Blaine Gabbert, as well as Terrible Rookie Year Jared Goff, Bad Charger Year Drew Brees, and Broken Down Bernie Kosar.  Dwayne Haskins would have joined the list in 2020.

As for ALEX, while other writers in the past have used it as more of a club, I do tend to think of it as a descriptive stat more than anything else, especially when we use all-down ALEX like this as opposed to third-down ALEX.

11 It's hard to separate ALEX's…

It's hard to separate ALEX's roots from how it's currently used -- it gets tarred by its own history.

Like if I named Mike Tanier's ironic, oft-pejorative analytics comments as Average Analytics Ridiculousness Over Normal. It's hard to ignore that it's called AARON as a slight. ;)


16 And yet, of the QBs with non…

And yet, of the QBs with non-negative ALEX (only 5 total) only one was below average in failed completions. On the other hand, the bottom feeders in ALEX were all worse in failed completions. 

If you're not aiming for the sticks, you're setting yourself up for failure. I think Stafford has finally figured that out - after years of negative ALEX, he's had consistently positive ALEX.

19 High ALEX seems to convert…

High ALEX seems to convert failed completions to failed incompletions. Fields and Hurts were high on ALEX and low on successful completion rate. (which is (completions-failed completions)/attempts)

10 Are per down FC rates…

Are per down FC rates interesting?  Obviously they would be correlated (a FC on first down makes a FC on later ones more likely). I wonder if there was anyone who had 2 FC's and then would convert the third anyway. (penalties are another thing. Much easier to fail on third and 25 vs third and 5. I doubt anyone had enough penalties to affect the results much however. )

Are the percentages arbitrary(they do make sense) or are those distances some sort of breakpoint for probability of a first down?(i.e. if you don't get 45% of the first down the it reduces the overall probability of new downs alot)

It does seem to be kinda a mix stat that has aspects of the QB play, playcalling and receivers mixed together. Might be a more intersting measure of offences and defences than players. 

21 How well does the league…

How well does the league-rate of FC% compare to their rate of going for it on 4th?

Is there a place on the field where FCs cluster?

23 I would expect them to be…

I would expect them to be more common in the red zone for obvious reasons.  I wonder if there is any variability between first/second/third/4th? downs?  Is one down more likely to have a FC? (relative to number of attempts for that down).