Advertisement
  1. Bucs

Jameis Winston's completion percentage: Don't worry about it

Of the 35 quarterbacks who have thrown at least 400 passes since 2015, Jameis Winston ranks 30th in completion percentage. He is third, however, in yards per completion. [WILL VRAGOVIC  |  Times]
Of the 35 quarterbacks who have thrown at least 400 passes since 2015, Jameis Winston ranks 30th in completion percentage. He is third, however, in yards per completion. [WILL VRAGOVIC | Times]
Published Jun. 21, 2017

With 24 seconds left against the Rams last season, Jameis Winston spotted Vincent Jackson open down the right sideline.

If only he hadn't missed his target. We wouldn't be talking this summer about whether the 2017 Buccaneers can end the franchise's playoff drought … because the 2016 team would have already done it.

Since entering the NFL in 2015, Winston has completed less than 60 percent of his passes. Imagine how good he could be if he completed, say, 70 percent of his passes.

One quarterback not only crossed the 70 percent threshold last season but also broke the record. He's a Heisman Trophy winner, too. He's also a former first overall draft pick. He has earned nearly $100 million. Teams have been so desperate to acquire his services that they've surrendered premium draft picks.

That man is Sam Bradford, quarterback of the Minnesota Vikings.

Sam. Bradford.

His 71.6 completion percentage masked a critical flaw: His passes resulted in relatively modest gains. He gained 9.8 yards per completion, a league low.

Before we deride Bradford as "Sammy Checkdown," it's worth noting that completion percentage can be less an indication of accuracy and effectiveness and more a representation of the style of an offense. The Vikings weren't built to consistently beat teams down the field. Because of injuries and blocking deficiencies, they instead relied upon quick, short passes.

The disparity between Bradford's completion percentage and his yards per completion, however, was historic. It was so extreme that Chase Stuart at Football Perspective named a statistic after the former Oklahoma Sooner. According to the "Sam Bradford Index," the difference was the third-largest of the Super Bowl era, after Sonny Jurgensen in 1966 and David Carr in 2006. (How'd Stuart figure that out? If I told you the answer involves standard deviations and z-scores, would you still be interested? If yes, the "Geek box" at the end of this story is for you.)

Tampa Bay fans have witnessed a handful of Bradford-esque seasons, and many of them occurred during the Jon Gruden years (2002-2008). While Brad Johnson, Brian Griese and Jeff Garcia come to mind as comparable quarterbacks, their seasons were not nearly as extreme as Bradford's.

Winston is at the other end of the spectrum — passers who post a low completion percentage but a high yards per completion average. Stuart has a statistic for that type of quarterback, too: the "Jay Schroeder Index," after the quarterback who led the NFL in yards per completion three times during stints in Washington and Los Angeles.

What about other Bucs quarterbacks? Who are the checkdown artists? Who are the gunslingers? Thanks to Stuart's formula, we're able to put numbers to those labels and determine the most extreme seasons in team history.

Here are the results, starting with the Bradford Index. (Remember, this is a measure of a quarterback's style of play, not an argument that one quarterback was better than another.)

Sam Bradford Index (Bucs quarterbacks only)

(high completion percentage/low yards per completion)

PlayerSeasonCmp%Cmp% z-scoreY/CY/C z-scoreDiff
Josh Freeman201162.80.6510.4-1.642.28
Brian Griese200469.31.8311.3-0.412.25
Brad Johnson200160.80.3810.0-1.692.07
Steve DeBerg198460.50.7011.5-1.231.93
Trent Dilfer199959.80.8111.1-0.711.52
Brad Johnson200362.10.7410.8-0.691.43
Jeff Garcia200864.90.8911.1-0.351.24
Jack Thompson198358.90.2911.7-0.911.20
Brad Johnson200262.30.5610.9-0.561.12
Steve DeBerg198757.80.3811.9-0.751.12
Chris Simms200561.00.2210.7-0.760.98
Note: To qualify, a quarterback had to average at least 14 attempts per scheduled game. Source: Pro Football Reference

Josh Freeman? What's he doing atop the Bradford Index? There's no definitive answer; Freeman's a bit of an enigma. He's responsible for arguably the greatest single season by a quarterback Tampa Bay has ever seen (25 touchdowns, six interceptions, 95.9 rating in 2010). He is also the only quarterback beside Trent Dilfer to appear on both of these lists.

Jay Schroeder Index (Bucs quarterbacks only)

(low completion percentage/high yards per completion)

PlayerSeasonCmp%Cmp% z-scoreY/CY/C z-scoreDiff
Doug Williams197941.8-2.8514.71.4-4.25
Josh Freeman201254.8-1.4513.31.91-3.36
Jameis Winston201558.3-1.2113.01.72-2.94
Craig Erickson199351.0-1.5613.11.35-2.91
Doug Williams198048.8-1.9213.40.76-2.68
Doug Williams198150.5-1.0415.01.63-2.67
Vinny Testaverde198847.6-1.3314.61.10-2.43
Josh McCown201456.3-1.7712.00.48-2.26
Trent Dilfer199554.0-1.0812.40.77-1.85
Vinny Testaverde199055.6-0.2813.91.39-1.67
Craig Erickson199456.4-0.2213.01.35-1.57
Josh Freeman200954.5-1.1911.70.21-1.39
Jameis Winston201660.8-0.6111.90.47-1.08
Shaun King200054.4-0.9111.90.15-1.07
Trent Dilfer199852.4-0.9212.10.05-0.97
Doug Williams198253.4-0.6912.60.19-0.87
Vinny Testaverde199257.5-0.0812.40.49-0.56
Note: To qualify, a quarterback had to average at least 14 attempts per scheduled game. Source: Pro Football Reference

Though Stuart named this index after Schroeder, he could have just as easily named it after Doug Williams. The disparity between Williams' completion percentage and yards per completion in 1979 is the 10th-largest of the Super Bowl era. Not even Schroeder himself can stake claim to a season like that. 1979 wasn't an aberration, either. Williams' 1980 and 1981 seasons rank 66th and 68th, respectively, on the index.

The table above also shows that Winston's rookie season was one of the most extreme low completion percentage/high yards per completion seasons in team history. His 2016 season was much less so.

There are a few explanations for that, with one of them being that the Bucs struggled to gain yards after the catch last season, even more than in 2015. Another explanation might be the decline in Winston's deep-ball accuracy. One more reason: Tampa Bay didn't have to play from behind as often last season. It didn't have to sling the ball down the field to catch up.

The graph below plots the completion percentage and yards per completion average for every qualifying Bucs quarterback. The numbers along the sides of the graph represent standard deviations above or below the mean. If a quarterback were exactly average in a given season, he would be a dot where the axes intersect. (Dilfer's 1997 season comes pretty close.)

The ideal quarterback would be in the upper right quadrant — high completion percentage/high yards per completion average. Only two Bucs players have ever had such seasons — Garcia in 2007 and Freeman in 2010.

Both of Winston's seasons are in the upper left quadrant — low completion percentage/high yards per completion average. His closest neighbor on the chart? You guessed it … Doug Williams. Winston's 2015 most resembles Williams' 1981, and Winston's 2016 most resembles Williams' 1982.

PlayerSeasonAgeCmp%TD%INT%Y/AY/CRating
Jameis Winston20152158.34.12.87.61384.2
Doug Williams19812650.54.037.61576.8
Jameis Winston20162260.84.93.27.211.986.1
Doug Williams19822753.42.93.66.712.669.6
Source: Pro Football Reference

The Williams comparison might not thrill some fans, but the game he played in the 1980s isn't the same as the game today. Since then, completion percentages and touchdowns have risen, while yards per completion and interceptions have declined. For his era, Williams wasn't a superstar, but he was good enough. No Bucs quarterback has started more playoff games, and it has been more than 34 years since he took a snap for Tampa Bay.

With the explosive DeSean Jackson in the fold, Winston's completion percentage could tick up this season. He probably won't ever be among the league leaders in that department, though — at least not in coach Dirk Koetter's vertical passing attack. And you don't want him to be. In this offense, that might be a sign that he's not taking enough shots down the field.

He's not Sam Bradford, and that suits the Bucs just fine.

Contact Thomas Bassinger at tbassinger@tampabay.com. Follow @tometrics.

Geek box

In developing the Bradford Index and the Schroeder Index, Stuart calculated the z-scores for completion percentage and yards per completion for every qualifying quarterback in every season since 1966, the first season that ended with a Super Bowl. Z-scores tell us how many standard deviations a value is from the mean.

Let's use Bradford as an example. His completion percentage last season was 1.95 standard deviations above the 2016 mean, and his yards per completion was 1.82 standard deviations below the 2016 mean. To get a sense of how unusual Bradford's season was, consider that in a normal distribution of data, roughly two-thirds of values will fall within one standard deviation of the mean. Ninety-five percent will fall within two standard deviations.

Stuart arrived at the number for his index by subtracting the yards per completion z-score from the completion percentage z-score. 1.95 - (-1.82) = 3.77.