Accuracy is a key when it comes to NFL success. Drew Brees, Peyton Manning and Aaron Rodgers are the top three NFL quarterbacks in completion percentage over the past five seasons. Those three have a combined regular-season record of 157-57 and have made four Super Bowl appearances during that time.
When taking accuracy into consideration, however, not all passes are created equal; a player can have an inflated completion percentage if the majority of his passes are screens and short slants.
Look no further than Geno Smith's AQ-leading 71.2 completion percentage in his final year at West Virginia. That season, Smith threw more screens than any other player in the nation, and his average pass traveled 7 yards past the line of scrimmage, a yard and a half shorter than the FBS average.
When comparing Smith to other quarterback prospects, completion percentage alone is not a good gauge of overall accuracy. A much better indicator is a concept known as standardized completion percentage.
Quarterbacks drafted in 2012 and 2013
(from AQ conferences)
Standardized completion percentage
Standardized completion percentage is very similar to effective field goal percentage in basketball because it accounts for the distance of a quarterback's passes, just as effective field goal percentage accounts for both 2-pointers and 3-pointers.
So a player like Smith receives credit for his short passes, but the amount of credit that he receives depends on how often a standard quarterback would throw that short.
For example, Smith threw 32 percent of his passes behind the line of scrimmage in his final year of college. A standard NFL quarterback would throw 13 percent of his passes that distance. Smith's standardized completion percentage reduces the importance of his 90.9 completion percentage on passes behind the line by assuming that he was throwing only 13 percent of his passes this distance.
What this does is give credit to players making more difficult throws at a higher completion percentage, not just increasing their overall completion percentage with short throws. Drops and throwaways are removed from the calculation because drops are not the quarterback's fault and throwaways are generally good decisions.
Last season, the top five quarterbacks in standardized completion percentage in the NFL were Rodgers, Manning, Brees, Philip Rivers and Nick Foles. All made the playoffs, and all finished the season ranked in the top six in Total QBR.
Predictive power
Among the 17 quarterbacks drafted from AQ conferences in the past two seasons, Russell Wilson, Andrew Luck, Foles and Robert Griffin III all ranked in the top five in standardized completion percentage in their final college season.
Looking at the bottom five draftees in this category, only Matt Barkley has taken an NFL snap. While Brandon Weeden may stand out as an outlier in the graphic to the right, it is important to note that he was 27 years old in his final college season, far older than the other prospects on this list.
Wilson's standardized completion percentage at Wisconsin was the best of any AQ quarterback in the past three years. Griffin ranked second in his final year at Baylor, followed by Teddy Bridgewater at Louisville in 2013.
The 2014 draft class
Draft prospects, 2013 stats
If this statistic is the truest measure of overall accuracy, Bridgewater is among the most accurate quarterbacks who have come out of college in the past three seasons.
Johnny Manziel and Zach Mettenberger also possessed "elite" accuracy in their final college seasons. Both players benefited from above-average accuracy with their deep balls and intermediate throws.
Logan Thomas was the least accurate of the top QB prospects. Even after accounting for the fact Virginia Tech dropped 36 passes, the most of any AQ offense, Thomas had a below-average 63.1 standardized completion percentage (AQ average is 66 percent).
Similarly, Tom Savage and Tajh Boyd struggled with their accuracy and posted standardized completion percentages below 70 percent. Savage threw a higher percentage of his passes off target (overthrown, underthrown or wide) than any other prospect, and Boyd relied heavily on screens, with more than 35 percent of his passes thrown behind the line of scrimmage.
It is important to note that, because quarterback is such a multidisciplinary position, there is no one statistic that will be able to predict NFL success. But this new take on completion percentage, which normally is misleading because of the different offenses run in college, may be a truer measure of accuracy and how a player like Bridgewater will adjust to the NFL.