Writers of Pro Football Prospectus 2008

Most Recent FO Features

SharpeTaj16.jpg

» Tajae Sharpe: Perfect Situation for Early Success?

What do you call a fifth-round rookie WR with real expectations? Tajae Sharpe, and there may not be another player like him in NFL history. Tennessee's poor history of developing wideouts has led to a rare opportunity that Sharpe can seize this season.

23 Feb 2016

Matthew Stafford's Obscure Record

by Scott Kacsmar

Passing records are constantly falling in today's NFL. In terms of volume, this comes as no surprise. The last five seasons are the top five seasons in history for total pass attempts and passing yards gained, and the last six seasons are the top six seasons for completions and touchdown passes. Records for passing efficiency are a little harder to come by. However, the 2015 NFL season ranks first all time in pass attempts (18,298), completions (11,527), passing yards (124,836), passing touchdowns (842), completion percentage (63.0 percent) and lowest interception rate (2.4 percent). The two highest seasons in touchdown pass rate since the 1970 merger are the last two seasons.

This passing fest did lead to a record milestone, but it is one so obscure that even us stat nerds failed to notice until mid-February. Of course, that can happen when the record was set by Matthew Stafford during a non-playoff season in Detroit.

The record: Stafford is the first quarterback in NFL history to complete at least 60 percent of his passes in all 16 games in a season.

Shocked? Amused? Perplexed? Stafford threw at least 25 passes in every game, so it was not like he had to rely on a 12-of-20 afternoon to keep the streak alive. But you could set the minimum number of attempts to one per game and Stafford would still be the only person to do it 16 times. Obviously, the 16-game requirement goes back to only 1978, but it's not like anyone was hitting 60 percent on a consistent basis in previous passing eras. The record for 60-percent games in a season from 1960-1977 was 10 by Fran Tarkenton in 1975.

I wrote about these streaks in 2012 when Matt Ryan was threatening the record, and highlighted the seasons a quarterback did it 15 times (minimum 15 attempts in each game), but Ryan suffered one off day. Since then, Philip Rivers (2013), Tony Romo (2014) and Russell Wilson (2015) have all finished one game short of what Stafford finally did. This table looks at the close calls, with each player's off day included.

Most Games Completing 60.0 Percent of Passes, Season
Quarterback Year Games "Off Game"
Matthew Stafford 2015 16 None (lowest: 60.4% in Week 2)
Kurt Warner 2001 15 Game 10: Tampa Bay, 19/39 (48.7%)
Peyton Manning 2003 15 Game 15: Denver, 12/23 (52.2%)
Daunte Culpepper 2004 15 Game 7: NY Giants, 24/41 (58.5%)
Peyton Manning 2009 15 Game 13: Denver, 20/42 (47.6%)
Drew Brees 2011 15 Game 15: Atlanta, 23/39 (59.0%)
Matt Ryan 2012 15 Game 12: New Orleans, 18/33 (54.6%)
Philip Rivers 2013 15 Game 1: Houston, 14/29 (48.3%)
Tony Romo 2014 15 None (only played in 15 games)
Russell Wilson 2015 15 Game 9: Arizona, 14/32 (43.8%)
12 times - 14 -
24 times - 13 -
Minimum 15 pass attempts in each game

Before Stafford, the 16-game benchmark was set by Drew Brees, who completed at least 59.0 percent of his passes in every game in 2011. Stafford has increased that to 60.4 percent, though there is an argument that another quarterback should at least share this record.

Romo actually completed at least 60 percent of his passes in all 15 games he played in 2014, but he missed a start against Arizona with injury. He went on to hit the mark in both of his playoff starts that year for a total of 17 games, so if you want to recognize that as the true single-season record, I won't blame you. Romo's streak actually goes back to 2013, and his lowest game was the beginning of the streak: 60.4 percent vs. Green Bay. Romo hit 60 percent in his last two games of 2013 and his first three appearances of 2015. Add in the playoffs and that is a streak of 22 games for Romo, the all-time record for consecutive games with at least 60 percent completions. So when Romo struggled to an 11-of-21 (52.4 percent) day on Thanksgiving against Carolina, we saw the end of an all-time streak that no one knew we were watching at the time.

That is really the problem with record streaks for efficiency stats. We rarely know they are happening due to games missed or a disagreement on what the minimum attempts should be.

Is the Record Impressive?

Unless you are loading up on failed completions, completing passes is generally a good thing for a quarterback to do at a high rate. However, completion percentage gets little praise in the advanced stats community. It is often criticized for the way it is over-valued in the league's passer rating formula. David Carr may have done more harm to the stat than anyone. His 2006 season showed that you can lead the league in completion rate (68.3 percent), but still lead one of the worst passing offenses in the league. If a quarterback really wanted to complete 75 or 80 percent of his passes for a season, he could do it as long as he made sure offensive success was no longer a goal.

Completing 60 percent of your passes is no longer a big deal in the NFL, as it has been done 484 times in a season since 1970 (minimum 200 attempts). In today's game, with all the short throws quarterbacks get to exploit, you just expect the offense to finish above that benchmark. The NFL's average completion rate first cracked 60.0 percent in 2007 and has only increased since. The 63.0 percent mark this season was an all-time record, and that was even with two all-time greats struggling in Peyton Manning (59.8 percent) and Aaron Rodgers (60.7 percent). In 2015, 25 teams completed at least 60 percent of their passes.

But what about hitting that mark every week? We see that it has been really difficult for a passer to maintain that league-average performance for all 16 games in a season. Sometimes you just run into a stingy defense that forces a lot of tipped balls or throwaways, or your receivers have a bad case of drops, or you just don't have "it" that day.

Stafford is an unexpected quarterback to be the first to do it every week, because he was just a 59.6 percent passer from 2009 to 2014 before completing a career-high 67.2 percent of his throws this season. When you look at that previous table of quarterbacks who came close, those were all MVP-caliber campaigns and/or career-best seasons. Stafford's best season is still 2011 in most people's eyes, but you can definitely make an argument that 2015 was his second-best.

Still, it's not like the 2015 Lions had a special year offensively. Detroit's offense ranked 13th in DVOA, 15th in passing DVOA, 15th in yards per drive, and 13th in points per drive. Optimistic fans will point to the promotion of Jim Bob Cooter to offensive coordinator in Week 8 as proof that things got better, which they did.

The Cooter Split: Matthew Stafford's 2015
Split Att Comp. Pct. Yards YPA TD INT PR DVOA DYAR ALEX PYD
Weeks 1-7 263 171 65.0% 1866 7.10 12 9 86.8 -0.1% 205 -2.4 6.9
Weeks 8-17 329 227 69.0% 2396 7.28 20 4 105.1 14.6% 602 -3.0 6.3

The numbers certainly improved in the presence of Cooter, but we can discuss if Stafford has truly turned a corner in greater detail in something like Football Outsiders Almanac 2016. This story is about completion percentage, and the numbers for ALEX and PYD (average air yards per target) are the most telling here.

With or without Cooter, Stafford threw the shortest passes in the league. His average pass traveled 6.6 yards beyond the line of scrimmage -- even Alex Smith was at 6.8. The 2015 league average was 8.5 yards, which includes throws negated by defensive pass interference penalties. Once Cooter took over for Joe Lombardi, Stafford's passes only got shorter, and he attacked the sticks less aggressively as his ALEX went down too. The receivers helped with few drops, but this record was built on limiting the vertical threat in this offense. Calvin Johnson averaged a career-low 13.8 yards per reception. Golden Tate, who was so dynamic in his 2014 debut with the team, saw his average reception drop from 13.4 yards to a career-low 9.0 yards. Stafford was 130-of-165 (78.8 percent) on passes thrown to running backs. The rest of the NFL completed 73.9 percent of its passes to running backs, so they are generally high-percentage plays. Only Rivers (172) threw more passes to running backs than Stafford did. According to ESPN charting, Stafford completed 75-of-89 screen passes (84.3 percent) in 2015.

While easy throws were so prevalent in this offense, they alone cannot explain the lack of an off day from Stafford in 2015. Breaking down each game helps; Stafford never topped an average of 8.81 air yards in any of the 16 games. In fact, the 8.81-yard game was the last before Cooter took over. I charted Stafford's average air yards over his last 80 regular-season starts (all since 2011) and found that his average has been trending downward, and it is unusual for him not to go above 9.0 air yards per attempt.

With Calvin Johnson eyeing retirement, you have to wonder if the days of vertical Stafford are over. If Stafford completes at least 60 percent of his passes in the 2016 season opener, his streak will move up to 18 games, the second-longest streak in history including playoffs. It would also be the longest uninterrupted streak in NFL history, as Stafford has not missed any games like Romo did multiple times during his 22-game streak. Fittingly enough, Stafford's streak actually started in a playoff loss to Romo's Cowboys to end the 2014 season.

Other Benchmarks to Watch

Consistency is the hallmark of great quarterback play, and doing things thought to be average on a consistent basis is harder than it sounds, so Stafford deserves some credit even though his 2015 season was nowhere close to being great. A discussion of great seasons should factor in consistency, as a quarterback can produce impressive averages with a few big-game outliers.

It is likely just a matter of time before other efficiency stats have a new 16-game benchmark. Peyton Manning could have set two of them in his 2004 season. In each of his first 15 games, Manning averaged at least 7.54 yards per attempt and had a passer rating of at least 93.5. In Week 17, with the Colts locked into the No. 3 seed, Manning played one series and only threw two passes against a Denver defense he would torch for 458 yards and five total touchdowns a week later in the playoffs. So no quarterback has ever averaged 7.0 yards per attempt or had a passer rating of 85.0 or better in all 16 games of a season. The 16-game benchmarks instead belong to Kurt Warner (at least 6.87 YPA in 2001) and Philip Rivers (at least an 84.5 passer rating in 2009). Surprisingly, Rivers (2009 and 2013) and Aaron Rodgers (2012) are the only quarterbacks to have at least an 80.0 passer rating in all 16 games.

Given that passer rating and yards per pass are more reliant on great play than completion percentage, it is a safe bet that someone far better than Stafford will eventually set marks at a high level in those categories for 16 games. Given the direction the league is going with shorter passes, we may be checking to see who breaks Stafford's 60-percent benchmark in 2016.

Posted by: Scott Kacsmar on 23 Feb 2016

16 comments, Last at 06 Mar 2016, 7:38pm by LionInAZ

Comments

1
by dmstorm22 :: Tue, 02/23/2016 - 3:48pm

Like the analysis, but needed to call out this:

" Calvin Johnson averaged a career-low 13.8 yards per reception."

if there's any reason why he needs to be in the HOF is that through a high-volume career, this is the LOWEST y/r he has had.

2
by Independent George :: Tue, 02/23/2016 - 4:59pm

I support Johnson in the HoF, but let's not get carried away here. The biggest argument against him is that he only played nine seasons, and retiring at 30 means he cut off all of his decline years.

6
by RickD :: Fri, 02/26/2016 - 12:46pm

"Decline years" shouldn't add much to any player's candidacy.

Nine is not an abnormally low number of productive seasons for any NFL player. Gale Sayers only had five productive seasons (though, as always, Sayers is by far the outlier here and I don't think he should be seen as a model candidate.)

Calvin Johnson's career totals are already higher than Michael Irvin's, who played 12 seasons. The "only nine seasons" argument should be summarily discarded.

7
by Tom Landrys Hat :: Fri, 02/26/2016 - 5:48pm

I think cutting the 'cutting off the decline years' was cited as a reason for his lowest YPC being as high as it was as YPC will often decline late in a receivers career.

8
by Thok :: Fri, 02/26/2016 - 8:40pm

I tried posting a link to a full list at pro-football-reference and hit their spam filter, but here's the short version

Number of seasons with >=60 receptions, >= 13.0 Yards per reception
Jerry Rice 14
Terrell Owens 11
Isaac Bruce 10
Randy Moss 9
Michael Irvin 8
Calvin Johnson 8
Steve Largent 8
Andre Reed 8
Marvin Harrison 7
Andre Johnson 7
Chad Johnson 7
Jimmy Smith 7
Steve Smith 7

(And yes, Jerry Rice is awesome, Terrell Owens should be in the Hall of Fame eventually, this list does favor modern wide receivers, and this list is slightly cherry-picked to favor Calvin Johnson.)

9
by tuluse :: Sat, 02/27/2016 - 1:20am

I could see minimum reception numbers favoring modern wide receivers, but yards per reception shouldn't, possible it should even favor older receivers.

Ok, I think my hypothesis is correct, check out Don Maynard http://www.pro-football-reference.com/players/M/MaynDo00.htm or Cliff Branch http://www.pro-football-reference.com/players/B/BranCl00.htm

Teams didn't throw as much in the past, but when they did, they threw deep. It would be interesting to drop the catch category to 3 catches per game instead of total catches. That would account for 14 game seasons and include a lot of old guys. It would also coincidentally grab CJ's rookie year.

10
by Thok :: Sat, 02/27/2016 - 2:17pm

Doing seasons with Yards/reception >13, Receptions>=3*games, games >=5 (to filter out injury years) gives

Jerry Rice 15
Issac Bruce 14
Terrell Owens 13
Andre Reed 13
Henry Ellard 11
Billy Howton 11
Steve Largent 11
James Lofton 11
Don Maynard 11
Randy Moss 11
Charley Taylor 11
Raymond Berry 9
Wes Chandler 9
Mark Duper 9
Irving Fryar 9
Joey Galloway 9
Michale Irvin 9
Calvin Johnson 9
Charlier Joiner 9
Steve Smith 9

Calvin Johnson would look better with a 10/11th season of doing that, but not so much better that it makes or breaks his HoF case.

13
by tuluse :: Tue, 03/01/2016 - 2:43pm

Thanks for doing the work Thok. Interesting stuff.

12
by SandyRiver :: Tue, 03/01/2016 - 12:40pm

Or Bob Hayes at the extreme: Career avg YPC was 20.0, and 19% of his catches went for TDs.

11
by Dan :: Mon, 02/29/2016 - 5:51am

Number of seasons with 70+ rec yards per game & 10+ games:
Jerry Rice* 14
Torry Holt 8
Calvin Johnson 8
Brandon Marshall 8
Randy Moss 8
Terrell Owens 8
Lance Alworth* 7
Tim Brown* 7
Larry Fitzgerald 7
Marvin Harrison 7
Michael Irvin* 7
Andre Johnson 7
Steve Largent* 7
James Lofton* 7
Jimmy Smith 7
Reggie Wayne 7

Number of seasons with 8+ rec TDs:
Jerry Rice* 13
Terrell Owens 11
Randy Moss 10
Marvin Harrison 9
Cris Carter* 8
Larry Fitzgerald 8
Antonio Gates 8
Steve Largent* 8
Don Hutson* 7
Don Maynard* 7
Tommy McDonald* 7
Lance Alworth* 6
Tim Brown* 6
Tony Gonzalez 6
Bob Hayes* 6
Calvin Johnson 6
Art Powell 6
Paul Warfield* 6

3
by Raiderjoe :: Tue, 02/23/2016 - 9:02pm

interesting star. Could be done again by othrt passrs but not guarsmyee becuusde always chance soem other guy has 15 games over 60% but adds a 17 for 36 clunker in yhetr. So very hard to match this thing by Stafford.

5
by rich006 :: Thu, 02/25/2016 - 5:02pm

took me a few minutes: guarsmyee = guaranteed
Raiderjoe's posts could be used as captchas.

4
by justanothersteve :: Wed, 02/24/2016 - 11:23am

Matthew Stafford proving you can set records if you dare to be mediocre.

14
by rj1 :: Wed, 03/02/2016 - 12:43pm

The problem with Stafford's outrageous passing statistics for lack of game success (and by extension, Calvin Johnson's) is it's easy to accumulate positive quarterbacking stats when the game's won and the defense is in prevent. I feel nowadays statistics as far as analysis is concerned should cease counting when there is nothing at stake. If a team is down 35-17 with 4 minutes to go, the statistics gained during the losing team's touchdown drive to make it 35-24 are nothing more than stat padding against a defense not trying too hard, and they don't mean anything for the offense because you know ahead of time they're losing. Call it the Stafford Theory.

15
by tuluse :: Wed, 03/02/2016 - 10:46pm

The Lions aren't going 0-16 these days, so I don't think this is fair. They've only been worse than 7-9 once in Stafford's career. So most games he's not facing garbage time (since not all losses even have garbage time).

16
by LionInAZ :: Sun, 03/06/2016 - 7:38pm

Under 7-9 twice, actually, in his first two years, but I agree with you that it's unfair to say that the Lions have been constantly playing garbage time. I don't know how they could have gone to the playoffs twice since 2010 if they have been so awful.