DeAndre Jordan shot 71.4 percent from the field last season.
Steph Curry shot 46.8 percent from the field last season.
Not a single reasonable person would suggest that Jordan is a better shooter than Curry. Nobody would call Jordan more accurate because of the numbers. Wider NBA analysis has evolved to the point that the ability to create your own shot and knock down said shot from further afield is more valuable than simply being a dunker. That 25 percent gap between the players’ numbers is offset by the value of Curry hitting three pointers at an exceptionally high rate, three pointers that are often shot off the dribble when he has created his own opportunity, something Jordan rarely ever does.
In the NFL we are still lagging behind in the evolution department. The wider analysis of the NFL focuses on completion percentage and little else when discussing how accurate each player is. For Cam Newton, that’s a problem.
Newton has always had the reputation of an inaccurate passer. His arm strength is such that his passes travel through the air faster than anyone else’s, meaning when he misses even slightly it looks more like a wild miss. Newton has also spent most of his career playing with limited receivers who create tightened windows while playing behind leaky offensive lines. Those are all issues but the primary reason for Newton’s reputation is our bias towards completion percentage.
Completion percentage rewards quarterbacks who throw the ball short all the time.
Take Sam Bradford for example. Bradford is actually an extremely accurate passer regardless of his completion percentage, but a big reason he set the record for completion percentage in a season last year was because of where he threw the ball. 62.13 percent of Bradford’s passes travelled fewer than five yards downfield. No other quarterback threw the ball that short that often. The second-ranked quarterback, Joe Flacco, wasn’t even close as he threw 56.87 percent of his passes to that level.
If you compare Bradford’s 71.6 completion percentage to Newton’s 52.9 it would be like comparing Jordan’s 71.4 to Curry’s 46.8. Nobody threw the ball further than five yards downfield more often than Newton did last year. An incredibly 67.57 percent of his passes travelled further than five yards. Only Jameis Winston was anywhere near him.
That means that Newton threw 32.43 percent of his passes fewer than five yards downfield. Half the rate of Bradford.
Bradford is an exceptional deep passer also so it’s unfair to paint him as someone who only relies on short throws but in general it is a problem that we call players who can throw short but not deep accurate and players who can throw deep but not short inaccurate. If you look at Newton’s numbers to different levels of the field, there is no question that he is a very accurate passer.
Newton was the fourth most accurate passer in the league on passes that travelled further than five yards downfield. His 68.46 percent mark was only beaten out by Andrew Luck, Aaron Rodgers and Drew Brees. Newton was second only to Tom Brady in the 11-20 yard range, he was accurate on 70.54 percent of those throws. On 21+ yard throws, Newton ranked eighth with a 49.18 percent accuracy number.
Last year Newton played in an offense that was designed to push the ball down the field with linear receivers who couldn’t run shorter routes to get open. Throwing downfield was difficult, it is difficult in the best of conditions, but especially so because his receivers struggled to get open and he was constantly delivering the ball from condensed pockets. Throwing short was an even greater challenge because receivers such as Kelvin Benjamin, Devin Funchess and Ted Ginn aren’t good at releasing from the line of scrimmage. Even Greg Olsen is largely just a vertical threat.
It’s why the Panthers drafted Christian McCaffrey and Curtis Samuel.
Newton’s poor accuracy numbers on shorter throws, 89.83 percent on throws behind the line of scrimmage (24th in the league) and 67.86 percent in the 1-5 yard range (32nd), should get better with receivers who can run more routes and create quicker separation. That is assuming Mike Shula incorporates play designs that don’t exclusively feature vertical releases and hard play fakes.
Even if Newton remains a subpar passer in the short yardage ranges, it would still be unfair to label him as inaccurate. We don’t label quarterbacks who can’t throw deep but stockpile enough short throws to bloat their completion percentage as inaccurate. (We probably should). Kirk Cousins is the most obvious example of this.
Cousins ranked ninth in percentage of passes thrown fewer than five yards downfield, he was more than 20 percent higher than Newton in this measurement. He ranked second in the 1-10 yard range. Cousins finished the 2016 season with a 67.0 completion percentage. Only eight quarterbacks had a higher completion percentage. Yet Cousins ranked 20th in overall accuracy, 14th on throws to the line of scrimmage, 15th in the 1-5 yard range, 27th in the 6-10 yard range, 23rd in the 11-15 yard range, 25th in the 16-20 yard range and 16th in the 21+ yard range.
By throwing enough passes within five yards of the line of scrimmage, Cousins was able to elevate his completion percentage with an 86.1 accuracy percentage on throws that didn’t travel further than five yards downfield. 86.1 sounds good, until you realize there were 17 quarterbacks more accurate than him to that range. There were 24 more accurate than him on plays where the ball travelled further than five yards downfield.
You see you don’t even need to be a great short passer to bloat your own completion percentage. The average accuracy on passes thrown behind the line of scrimmage was 90.78 percent, it was 82.83 percent on throws in the 1-5 yard range. Even if you are far below average, you will still have an accuracy percentage greater than 71.6 (the number Bradford set as the record completion percentage this year).
The mention of Cousins brings us to the other necessary qualifier that must be used when it comes to accuracy: Receiver quality.
Fortunately for Steph Curry and DeAndre Jordan basketball hoops don’t randomly close when they shoot. You can be unlucky when a ball rims out but you are completely in control of whether your shot goes in or not. NFL quarterbacks are at most 50 percent responsible for whether their passes are complete or not. Cousins’ receivers rarely dropped passes, in fact his receivers created more receptions with difficult adjustments on inaccurate passes than they lost by ruining accurate passes.
Newton’s receivers are big but they aren’t good. They don’t adjust well to the ball in the air and they can’t create separation. 59 times last year Newton threw an accurate pass only to have his receiver ruin the play. Only Aaron Rodgers lost more receptions to receiver error.
Kelvin Benjamin cost him 13 completions, Ted Ginn cost him 11, Devin Funchess also had 11, Corey Brown six and Greg Olsen six.
When a quarterback throws 600 attempts in a season it only takes 50 plays to move his completion percentage eight percent in either direction. If you have two quarterbacks, one with a great set of receivers and one with an awful set, one who rarely sees any pressure and one who is constantly performing in spite of it, one who rarely pushes the ball downfield and one who lives on short throws, you can very quickly discover that completion percentage is painting a completely misleading picture.
Not all quarterbacks throw 600 passes in a season. Someone who throws 400 only needs 40 plays to jump 10 percent. The NFL plays 16 games, it’s not the NBA or MLB where the numbers have a better chance of stabilising themselves. It’s also a more interdependent sport where players are more reliant on schemes and teammates to get to the numbers they are ultimately judged on.
Calling Cam Newton inaccurate is calling Steph Curry inaccurate. It’s living in an antiquated world where evaluation relies on surface level stats and as little critical thinking as possible.
*It should be noted that all of Newton’s numbers in this article include the final quarter of the season when he played hurt. Newton had three games with an accuracy percentage below 60 percent last season, all of which came after he was hurt. If these numbers only included games he was fully healthy for they would be significantly better across the board.*
All of the numbers used in this article can be found in the Pre-Snap Reads Quarterback Catalogue.