As of press time, we don’t know what uniform Tom Brady will be wearing in 2020 (assuming there is a 2020 season, but, hey, optimism). It’s not actually all that unusual for a legendary quarterback to finish his career in an unfamiliar location, of course. If Brady does in fact leave New England, the top five quarterbacks in Pro Football Reference’s Hall of Fame Monitor will all have finished their careers wearing strange togs. It’s weird. It happens.
One of the most rumored destinations for Brady is Tampa Bay, where Jameis Winston may have worn out his welcome after the first 30-touchdown, 30-interception season in NFL history. Bruce Arians has had success with older quarterbacks in the past — see Carson Palmer in Arizona — so from his point of view you can see the appeal of adding a legend like Brady over an inconsistent player like Winston, even if Winston’s arm strength is, on paper, a better match for Arians’ downfield attack than Brady’s recent noodle-arm tendencies.
We don’t have the time or, frankly, the bytes to describe the myriad ways in which Brady and Winston are different players, but one of the biggest differences between them ties quite nicely into our yearly analysis of failed completions.
Every year, we here at Football Outsiders study failed completions around the league. 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.
2019 was sort of an odd year for failed completions. For the first time since 2014, no qualified quarterback hit a 35% failed completion rate. For the first time since we started doing this yearly analysis in 2013, only one quarterback topped a 30% failed completion rate. There certainly were some low markers, but nothing as bad as we’ve seen in past seasons. And yet, the overall failed completion rate stayed fairly constant, with qualified quarterbacks averaging 24.9%, slightly up from 2018. The extremes on both ends weren’t as strong, but failed completions certainly aren’t going anywhere anytime soon.
Not every failed completion is created equally, 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 in field goal range, and that is reflected in DVOA — but for the purposes of this article, a failure is a failure.
In the following table, the 34 qualified quarterbacks of 2019 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.
|Quarterbacks, Failed Completions, 2019|
Atlanta’s Matt Ryan led the league with 108 failed completions, but that’s mostly because he had 671 dropbacks, second only to Winston. His failed completion rate of 26.5% was just a little higher than average
Let’s talk a little bit more about Winston and Brady. Winston’s a full two standard deviations below the league average in failed completion rate, just a handful of failures behind Dak Prescott for the top spot on the table. Winston was second behind Matthew Stafford in both ALEX (1.7) and average depth of target (10.7), and it turns out that it’s very difficult to have failed completions when your average target is way past the first-down marker! It’s no surprise that eight of the ten ALEX leaders show up in the top 10 in failed completion percentage; failed completions are as much an indicator of a player’s style as they are his quality.
As such, as the league’s premier gunslinger, Winston always finishes near the top of these lists. Since entering the league in 2015, he has now finished third, third, first, second, and second in failed completion percentage. We mentioned back in 2015 that Winston’s combination of vertical nature and erratic accuracy made him a likely candidate for low FC% if his rookie season was an indicator of future performance and, well, yeah. It was. You know what you’re getting with Winston, and 3-yard checkdowns ain’t it.
In his prime, Brady could put up better numbers than Winston, despite not being a gunslinger. As recently as 2012, Brady had a 14.0% failed completion rate. But that Brady doesn’t really exist anymore. From 2013 to 2018, Brady’s FC% hovered between 21% and 24%; there was a slight year-to-year downward decline, but nothing particularly concerning for a quarterback in his 40s. The bottom fell out last year, however. Brady topped 100 failed completions for the first time in his career, despite completing his fewest number of passes in a 16-game season since 2010. His 27.9% FC% wasn’t the highest in the league, nor were his 104 failed completions the most. However, Brady did lose more DYAR on failed completions than any other quarterback; no one in the league hurt their team more with failed completions than Tom Brady in 2019. Brady also dropped 6.6 percentage points between 2018 and 2019, the biggest drop for any quarterback in the league. This is concerning!
Some of that is almost certainly age-related; even casual observers can tell that Brady’s arm strength isn’t what it once was. However, you do also have to point to the collapse of the Patriots’ receiver corps from 2019. Far too often, Brady was stuck in the pocket, waiting and waiting for someone to get separation. Putting him in Tampa Bay with Mike Evans and Chris Godwin won’t make him replicate Winston’s failed completion rate, but there’s every reason to believe he won’t have quite the same level of checkdownitis.
It’s also worth remembering that a failed completion is not the worst result of a passing play, and Winston and Brady again give us a chance to demonstrate that. The next table breaks down every dropback for both quarterbacks. It splits completions into four types — failed completions, successful short completions (less than 5 yards in the air), successful medium completions (6 to 15 yards in the air), and successful deep completions (more than 15 yards in the air).
|Tom Brady vs. Jameis Winston, 2019|
|% of Plays||6.6%||15.3%||19.3%||15.9%||34.8%||2.1%||4.1%||1.2%||0.6%||100%|
|% of Plays||9.8%||19.0%||16.6%||10.1%||30.6%||1.7%||6.9%||4.4%||1.0%||100%|
You can see that Winston blows Brady out of the water when it comes to value from the deep ball no matter how you slice it — raw count, percentage of dropbacks, or DYAR. He also has a comfortable lead on those medium completions, and loses significantly less DYAR on failed completions. When the ball was caught, Jameis Winston was a more valuable quarterback than Tom Brady was in 2019. Adding in incomplete passes only extends Winston’s lead.
But, of course, Winston gives all that back and then some when you take into account sacks and interceptions. A completion taking you from second-and-10 to third-and-8 is a bad play, certainly. But it’s far preferable to going from second-and-10 to third-and-15, or first-and-10 for the other team! Failed completions are usually a sign of playing things safe; whether that’s a good thing or a bad thing really depends on the individual passer. If Brady does end up in Tampa Bay, I’m sure Buccaneers fans can get behind three or four extra dumpoffs per game, if that means fewer of the lowlight-reel plays Winston produced. Still, Brady’s sudden increase in failed completions is a concerning development, and should give some of his free-agency suitors significant pause. His decision-making hasn’t declined, but his physical talent has.
But we’re talking about failed completions today, not overall quarterback value. That brings us neatly to Mason Rudolph, the only quarterback to finish two standard deviations above average this year with a 33.5% FC%. Rudolph’s high failed completion rate wasn’t about him avoiding negative plays, as he still ranked near the top of the league in interception percentage. Nor was it about being overly conservative with his passes; his -0.8 ALEX and 8.4 aDOT were both roughly league average. Instead, Rudolph simply wasn’t very good. most of his deep shots weren’t caught, and all he was left with was the dinking and dunking. His Week 9 game against the Colts tied for the most failed completions in a single game this season with 13, and his shortcomings were clearly apparent week after week. For the record, however, he wasn’t the lesser of the two Pittsburgh quarterbacks; Devlin Hodges had a 34.0% FC%. Numbers like this explain how the top team in weighted defensive DVOA stays at home in the postseason.
The rest of the bottom of the list is mostly who you would expect. All-time Failed Completions king Joe Flacco stands tall with what may be his last qualifying campaign, just missing out on his fourth 30.0% season. All three first-round rookie quarterbacks show up near the bottom, as does Gardner Minshew — rookie quarterbacks often appear near the bottom of these tables as coaches give them safe options while they learn to play at the NFL level. Teddy Bridgewater is down here with his league-worst -3.3 ALEX; the Saints’ system produces a lot of failed completions, but that’s a ton even for a New Orleans quarterback. And then you have Mitchell Trubisky, the low-water mark for an established quarterback who wasn’t benched over the course of 2019. Trubisky’s rookie season had the eighth-highest FC% going back to 1989 at 36.2%, so I suppose this year was at least an improvement. Still, yikes. You would hope to see significant improvement from a quarterback in his third season, and it just isn’t showing up for Trubisky.
To talk about significant improvement, we have to go to the top of the list and Dak Prescott. Prescott’s rookie season in 2016 had a fairly solid 21.2% FC%, but he was down closer to 29% in 2017 and 2018. And now he’s on top of the world with a 17.3% failed completion percentage, just in time for contract negotiations. We can credit this, at least in part, to a new-found willingness to push the ball downfield. In 2018, his all-down ALEX was -1.5; that rose to 0.2 in 2019. His average depth of target also rose from 7.6 in 2018 to 9.2 in 2019. Yes, credit superior play calling, credit a full year of Amari Cooper, and credit the Cowboys being behind more to force Prescott to throw more downfield; all of this helped. But Prescott’s deep ball looked significantly better in 2019 than it has at any point in his career, and he continues to develop well. The fact that this is happening just as he’s negotiating a new deal with Dallas can only mean positive things for his pocketbook.
Three other quarterbacks saw improvements of at least 5.0% between 2018 and 2019 — Ryan Tannehill in his Comeback Player of the Year season, Matthew Stafford, and Russell Wilson. All three of them set career highs in DVOA, Stafford and Tannehill by significant margins. Stafford and Tannehill showed a new willingness and ability at deep passing, nailing a high-risk, high-reward style to produce the best seasons of their careers. For Wilson, it’s mostly recovering from a one-year blip in 2018; a 20.8% FC% is his career best, but the improvement mostly came from 2018 being something of an outlier.
Three quarterbacks joined Tom Brady by declining by at least 5.0% from 2018 to 2019. One was Patrick Mahomes, who was injured and thus proven to be mortal. Sam Darnold can also hang at least some his decline on mononucleosis, though the lack of real improvement under so-called quarterback guru Adam Gase should be a red flag for Jets fans. The third is Andy Dalton, who struggled through 2019 with one of the worst supporting casts in football. Not having A.J. Green hurts!
Back by popular demand, we’re also going to look at successful completion percentage. 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, 2019|
Prescott and Winston both jump double-digit places in completion percentage when you ignore failed completions, Winston going from terrible to just above average, and Prescott going from just above average to nearly the very top. Rudolph falls 13 spots to end up in stone cold last, with Bridgewater also suffering quite a bit from the loss of his checkdown targets; such is the danger of a -3.3 ALEX.
Other quickfire notes from the table:
- Tom Brady actually finishes last in the AFC East in successful completion percentage, with Josh Allen just barely pipping him to the line. All four are still below average, with Ryan Fitzpatrick leading the group in what was a very underwhelming year from the division.
- Speaking of playoff quarterbacks, Jimmy Garoppolo and Kirk Cousins are in a very similar boat. Each ended the year with a top-five completion percentage; each falls significantly when you take out the failed completions, but each remains above-average at the end of the day. Ranking 11th and 13th in the league is probably closer to their true talent ranges than their raw completion percentages would have you believe. For Garoppolo, this is part of the Kyle Shanahan system which relies so heavily on YAC; he had the lowest average depth of target for any non-Saint in 2019, but led the league with 6.6 average YAC. More on that in a future article.
- Browns fans can take some solace in the fact that Baker Mayfield had a low percentage of failed completions; he jumps six slots in the rankings when you remove them from the equation. That’s the most for anyone who still ends up in the bottom half of the table; the most improvement by anyone who still had a disappointing season. Little victories.
Receivers: Failed Receptions
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, Tarik Cohen led all receivers with 41 failed receptions; Mitchell Trubisky needed someone to dump the ball off to, after all. Alvin Kamara was right behind him with 40, while Christian McCaffrey and Leonard Fournette also topped 35.
|Most Failed Receptions (WR/TE)||Lowest Failed Reception Rate (WR/TE)||Highest Failed Reception Rate (WR/TE)|
|Cooper Kupp||LAR||24||Michael Gallup||DAL||66||2||3.0%||Steven Sims||WAS||34||16||47.1%|
|Dede Westbrook||JAX||23||Mike Williams||LAC||49||2||4.1%||Vance McDonald||PIT||38||16||42.1%|
|Jamison Crowder||NYJ||22||Mike Evans||TB||67||3||4.5%||Albert Wilson||MIA||43||16||37.2%|
|Larry Fitzgerald||ARI||22||Jared Cook||NO||43||2||4.7%||Evan Engram||NYG||44||16||36.4%|
|Tyler Boyd||CIN||22||Amari Cooper||DAL||79||4||5.1%||Kaden Smith||NYG||31||11||35.5%|
|Julian Edelman||NE||21||DeVante Parker||MIA||72||4||5.6%||Geronimo Allison||GB||34||12||35.3%|
|Michael Thomas||NO||20||Preston Williams||MIA||32||2||6.3%||Dede Westbrook||JAX||66||23||34.8%|
|George Kittle||SF||20||Corey Davis||TEN||43||3||7.0%||Jonnu Smith||TEN||35||12||34.3%|
|Robert Woods||LAR||19||Breshad Perriman||TB||36||3||8.3%||Ryan Griffin||NYJ||34||11||32.4%|
|Keenan Allen||LAC||18||Chris Conley||JAX||47||4||8.5%||Jacob Hollister||SEA||41||13||31.7%|
|DeAndre Hopkins||HOU||18||D.J. Moore||CAR||87||8||9.2%||Irv Smith||MIN||36||11||30.6%|
|Odell Beckham||CLE||17||Alshon Jeffery||PHI||43||4||9.3%||Noah Fant||DEN||40||12||30.0%|
|Austin Hooper||ATL||17||DeMarcus Robinson||KC||32||3||9.4%||Gerald Everett||LAR||37||11||29.7%|
|Min. 30 receptions|
It was a very, very tight race for the top of the failed receptions leaderboard this year, with Cooper Kupp just barely managing to hold on to the lead by the end of the year. But while Dede Westbrook couldn’t quite match Kupp in raw totals, you have to remember that Kupp caught 94 passes in 2019; Westbrook had just 66. Thus, it should really be Westbrook who is acknowledged as the failed reception king for 2019. Of course, full credit has to go to catching passes both from career failed completion specialist Nick Foles as well as rookie Gardner Minshew; Jacksonville finished second in the league with 113 failed completions for a reason.
Only two players make a return appearance on the failed receptions table, as longtime stalwarts Jarvis Landry, Golden Tate, and Nelson Agholor dropped off in 2019. This is Michael Thomas’ third year in a row amongst the league leaders; when you have 149 receptions, it’s not surprising that a handful of them won’t work out. Austin Hooper also makes a return appearance as one of four Falcons receivers to finish with double-digit failed receptions. Hooper is a free agent, and is expected to become the highest-paid tight end in the NFL when he eventually signs. He surely has the volume to make teams want to sign him, and he had a better successful completion rate than George Kittle did in 2019, but he does get more than his fair share of third-down dumpoffs.
Deep-ball specialists make up the table of fewest failed receptions. Mike Williams, Breshad Perriman, Mike Evans, DeVante Parker, Michael Gallup, and Amari Cooper all ranked in the top 15 in average air yards per reception, so you would expect most of their catches end up as successful completions. With Gallup and Cooper, you can really see how Dak Prescott’s improved willingness to throw the ball downfield helped. Evans makes his fifth consecutive appearance on this table, nicely matching Jameis Winston’s career to this point. It will be very interesting to see how he does if he plays with someone else under center.
All of the 13 names on the highest failed reception rate were in the bottom 30 in air yards per reception. Vance McDonald’s average reception had a -7.0 ALEX! Even getting the ball in the hands of someone as speedy as Steven Sims isn’t going to work on a regular basis when they’re catching the ball 5 yards behind the first-down marker.
Finally, let’s look at the defenses’ ability to create failed completions, with a comparison to how these units fared in 2018.
|Defenses: 2019 Failed Completions Compared to 2018|
|Rk||Team||Comp||Failed||FC%||2018 Rk||2018 FC%||Diff||Rk|
As a reminder, defensive numbers are far less sticky from year-to-year than the quarterback stats; there’s about a 0.25 correlation for defensive failed completions. A lot of this depends on the types of quarterback on the schedule before you even get to year-over-year turnover. Still, a 0.25 correlation isn’t nothing; half of 2018’s top dozen teams repeated their performance last year, as did half of 2018’s worst twelve teams.
You do have to be careful when looking at this table, of course. Vic Fangio’s Broncos saw the biggest year-to-year increase in failed completions, but their pass defense DVOA fell from -10.6% to 1.9% as they experienced some transitional growing pains. Broncos opponents did average fewer yards per completion in 2019, but the Broncos allowed a better completion percentage and had worse sack and interception rates than they did in 2018. Those extra failed completions weren’t successful receptions that became failures thanks to better coverage and tackling; they’re passes that were hitting the ground a year ago turning into small gains. Those are still net “wins” for the defense, but better results for the offense than they were getting two years ago. Less than ideal, and a reminder that allowing failed completions is not necessarily the sign of a good defense.
That being said, we can still learn something from this table; it’s still correlated with good pass defense in general. What you’re seeing here is a list of teams that force the most dumpoffs, checkdowns, and low-ALEX plays. Six of the top ten teams in FC% were also in the top ten in ALEX allowed on completions. If you can constantly take away the deep pass — either by pass rushes that don’t give routes time to develop or by lockdown coverage deep downfield — you’re going to see your FC% rise, and quickly.