Beyond the stats: EJ Manuel vs. the rookie spread

Throughout the season, we’ve been tracking EJ Manuel’s similarity to recent rookie quarterbacks. Now that the season is over and Manuel, who only played ten games in his rookie campaign, has been named the starter for 2014, it’s time to dissect his rookie season some more.

Since the merger (the 1970 season), 75 rookie quarterbacks have started at least eight games. Those 75 rookies combined for a 54.1% completion percentage and 6.47 yards per attempt. Manuel’s average yards gained per pass attempt (6.44) was on par with that average. When accounting for sacks, Manuel’s net yards per pass play were exactly the same as the rookie average (5.43 yards per pass play).

Twelve rookie quarterbacks have started at least ten games for their team since 2011. Those twelve rookie quarterback starters and their net yards per attempt are depicted below. Manuel was a little behind the recent curve.



Manuel’s rookie season net yards per attempt were slightly below the middle of this impressive group. His touchdown-interception differential moves his status in the group up a bit. While +2 isn’t incredibly impressive (25 quarterbacks who started four or more games in 2013 were +3 or better), it’s solid for a rookie. Here’s how Manuel matches up against the twelve most recent rookie starters:


Throughout the season, much was said of Manuel’s accuracy issues. Articles were written about his ranking as “least accurate passer” in early December, but Buffalo’s first round pick had the highest completion rate among these twelve rookie starters. Sure, Manuel threw a lot of short passes, but this could be something the Bills’ coaching staff could build upon.


During the season, I compared each player’s stat lines to Manuel’s based on the number of games he played up to that point (the five most similar through ten games were Russell Wilson, David Carr, Joe Flacco, Andrew Luck, and Ben Roethlisberger, in order). Now that the season is over, I took all 75 starting rookie quarterbacks and compared their rookie seasons to Manuel’s.

The varying number of games has a pretty large effect (since it’s hard to match Andrew Luck’s 627 pass attempts in six fewer games), but is another factor in his assessment. Why should we insulate our analysis of the player from durability or ability to play throughout a given season?

The results were a bit different than the game-by-game comparisons. The most similar rookie performances to EJ Manuel’s 2013 season were Bruce Gradkowski (Tampa Bay 2006), Charlie Batch (Detroit 1998), and David Woodley (Miami 1980). All three players started eleven or twelve games their rookie seasons and none were first round draft picks.

Batch and Woodley went on to start more than ten games their second seasons (both “won” more than half of their starts) while Gradkowski has started just nine games in the seven seasons since his rookie year.

An interesting note is that Manuel’s first season has been the second-most similar to Ben Roethlisberger’s first year in Pittsburgh (Flacco’s rookie year was the most similar). Big Ben was the beneficiary of the league’s best defense in terms of points and yards allowed. The Steelers also threw the fewest passes in the league that year, electing to run the ball with Jerome Bettis and Duce Staley a combined 442 times. (Fred Jackson and CJ Spiller combined for 408 attempts for the Bills this season.)

While rookie year similarity scores give us an idea of how a player might develop (it’s not too inspiring in this case anyways), it’s also important to look at the population as a whole to determine whether an overall development trend exists.

65 quarterbacks started at least eight games their rookie season and then went on to start at least one game their second year. Of those 65 quarterbacks, 37 improved their overall efficiency (net passing yards and rushing yards per rush and pass play) in their second year in the league. Sixteen (almost 25%) improved their total yards per play by more than 0.85 yards, which was the difference between EJ Manuel and the 2013 NFL average. That improvement would mark a 16.4% improvement from his rookie season, a feat eighteen of the 65 players were able to accomplish in their second season.

Manuel was a work in progress when he was drafted last spring and that hasn’t changed. He was a below average in the scope of the league, but was average in the scope of rookie quarterbacks. He could become a Charlie Batch or a Ben Roethlisberger, but the outcome can’t yet be known because of our limited data. For now, I’ll leave you with a few stats on Manuel’s performance this year.

First, Manuel’s average pass depth on all passes and complete passes in his ten games this season.

Six of the ten games had an average pass depth of less than six yards on complete passes. His season average on all passes was 2.5 yards greater than his completions. That’s thanks to a 40.6% completion rate on all passes thrown ten or more yards past the line of scrimmage. Manuel had even more trouble on the deep out, completing just 26.3% of his passes ten or more yards down field and outside the numbers.

Manuel threw 210 of his 306 passes ten or fewer yards past the line of scrimmage. Over 67% of those passes were completions, which isn’t too far off the pace of league leaders (the top quarterbacks by quarterback rating (except Peyton Manning because he skews the sample a TON) complete 70-75% of their short passes). Manuel’s passes, by region, are depicted below.

Another way to analyze the rookie’s passes is to plot them on the field. In the graphic below, the complete passes are blue, the incomplete passes are red, and the interceptions are yellow.

A striking observation is that five interceptions came on passes thrown shorter than ten yards past the line of scrimmage. Another is how much Manuel struggled with the deep passes (especially early in the season).

Maybe huge proportion of short passes (almost 69% of his passes were shorter than ten yards!) were thrown because Manuel felt more comfortable making those throws. Maybe the Bills coaching staff limited his reads because he was a rookie. Maybe guys just weren’t open down field (although that’s been widely disproven with various All-22 analyses and breakdowns).

Manuel needs to develop more, there’s no denying that fact. His knee injuries muddied his progression and the prospects of his future development. He wasn’t a great rookie, a fact that disappointed some, but he wasn’t bad either. He and the rest of the offensive system will be given a chance to learn and develop together this offseason, and that is hopefully the key—the pieces need to come together to create an imposing and efficient product on Sunday afternoons in 2014.


About Joe

The Lord of Buffalo Wins