clock menu more-arrow no yes

Filed under:

Why second-year starting quarterbacks improve when they do improve

New, comments

A good number of second-year starters do not improve, but when they do here is what they did.

NFL: Denver Broncos at Chicago Bears Dennis Wierzbicki-USA TODAY Sports

If you read my last offering, you know that on average first year starting QBs do not normally improve in their second year as starting QBs in the NFL. However, there are a few who have recently improved greatly. In this piece I am going to focus on four of those players to see if there are any parallels that we draw to Trevor Siemian. Obviously this piece is based on the supposition that he is our starter again this year.

The four recent (this century) QBs that I am going to focus on are: Matthew Stafford, Carson Palmer, Blake Bortles and David Derek Carr. Each QB improved by 15 points or more in their QB rating from their first year to their second. So that you can keep in mind what Trevor Siemian did last season I will show his stat line below so that you may get an idea about where he was statistically and where he could go, if he improves like these four QBs did.

Trevor Siemian 2016 Stats

Age GS QBrec Cmp Att Cmp% Yds TD TD% Int
25 14 8/6 289 486 59.5 3401 18 3.7 10
Int% Rate QBR Sk NY/A ANY/A Sk% 4QC GWD AV
2.1 84.6 55.8 31 6.22 6.04 6 3 2 9

Matthew Stafford

Stafford Stats for his first year and his second year

Year Age GS QBrec Cmp Att Cmp% Yds TD TD% Int
2009 21 10 2/8 201 377 53.3 2267 13 3.4 20
2011 23 16 10/6 421 663 63.5 5038 41 6.2 16
Year Int% Rate QBR Sk NY/A ANY/A Sk% 4QC GWD AV
2009 5.3 61 40.6 24 5.23 3.64 6 1 1 3
2011 2.4 97.2 64.4 36 6.84 6.98 5.2 3 4 16

Matthew Stafford was atrocious as a rookie. Trevor Siemian played much better in 2016 than Stafford in 2009. Stafford hurt himself in preseason and missed almost the entire 2010 season. Missing a year due to injury allowed him to make a huge leap in QB performance. It’s interesting to note that the sad-sack Lions, who seem to go through coaches quickly, had the same head coach and offensive coordinator in 2009, 2010 and 2011.

Matthew Stafford not only had a year “off” to learn the NFL game, but he was also in his third year in the same offense. Jim Schwartz was the HC and Scott Linehan was the OC. Stafford had roughly the same offensive talent around him in 2009 as in 2011 (Calvin Johnson and some other mostly JAG WRs, a host of mediocre RBs), but one thing that he had in 2011 that he didn’t have in 2009 was an offensive line that was healthy and intact for the entire season.

The 2009 Lions only had two offensive lineman start all 16 games (Jeff Backus and Dominic Raiola) with Gosder Cherilus (15 GS), Manny Ramirez (12 GS) and Stephen Peterman (9 GS) rounding out the line. The 2011 Lions had all five starters start 16 games and Rob Sims was playing LG instead of Manny (Sims > Manny). The other 4 OL starters were the same, but they were healthy and they had been playing together for a number years at that point - something which shouldn’t be overlooked (familiarity and communication) when it comes to effective OL play. The healthy OL also led to an improved running game for the Lions in 2011 (4.3 ypc) relative to 2009 (4.0 ypc).

So the four reasons why Stafford improved so dramatically were maturation (the game slowed down for him), familiarity with the offense, an improved running game and a healthy/improved OL. I should also note that Stafford only played 10 games as a rookie before he was injured. He missed games five and six with a dislocated patella, then he missed the final four games of the season with a shoulder injury. He was fully healthy in 2011 - his first fully healthy season in the NFL. So it’s possible that his year over year improvement was also due partly on his health.

Carson Palmer

Palmer’s Stats from his first and second year

Year Age GS QBrec Cmp Att Cmp% Yds TD TD% Int
2004 25 13 6/7 263 432 60.9 2897 18 4.2 18
2005* 26 16 11/5 345 509 67.8 3836 32 6.3 12
Year Int% Rate QBR Sk NY/A ANY/A Sk% 4QC GWD AV
2004 4.2 77.3 25 5.95 4.96 5.5 2 3 8
2005* 2.4 101.1 19 7.07 7.26 3.6 1 3 17

Carson Palmer, like Matthew Stafford, was the 1st overall pick in the draft, which pretty much guaranteed that they were both going to start as rookies, but it was a different era and so Palmer sat behind Jon Kitna and didn’t appear in a single game in 2003. Palmer did not fare as poorly as Stafford in his “de facto” rookie campaign, but he was not very good either. He had good offensive talent around him both years with Rudi Johnson at RB and Chad Johnson as his primary receiving target. He had the same head coach and offensive coordinator both years so he was running the same offensive system. The Bengals running game was the same both years (4.2 ypc) He also had a relatively healthy OL (one starter missed 6 games during his rookie year) with an elite LT in Willie Anderson (named 1st team AllPro in 2004 and 2005).

So what changed? He did. Like Stafford, the NFL game slowed down for him in his second year as a starter allowing him to improve significantly upon his 1:1 TD:INT ratio as rookie (18:18). He improved to 32:12 in his second year. He also went from a pedestrian 60.9% completion to a league best 67.8% in his second year (he wasn’t that accurate in college). Like Stafford, Palmer only played 13 games a rookie, but Palmer did not sit the final three games of the season because of injury. Marvin Lewis decided to let the veteran, Jon Kitna, start the final three regular season games in the hopes of the Bengals getting a wildcard (they didn’t).

Blake Bortles

Bortles’ stats from his first and second year

year Age GS QBrec Cmp Att Cmp% Yds TD TD% Int
2014 22 13 3/10 280 475 58.9 2908 11 2.3 17
2015 23 16 5/11 355 606 58.6 4428 35 5.8 18
year Int% Rate QBR Sk NY/A ANY/A Sk% 4QC GWD AV
2014 3.6 69.5 31.7 55 4.84 3.81 10.4 1 1 5
2015 3 88.2 55.9 51 6.25 6.09 7.8 3 4 14

Unlike the last two QBs, Bortles was not the first overall pick in the draft (he was 3rd). His struggles were similar to Stafford’s as a rookie. Bortles did not win the starting job in training camp. Head coach, Gus Bradley, decided to begin the season with Chad Henne as the starting QB. Despite being the veteran and much less prone to mistakes, the offense was inept under Henne (scoring 10, 17, and 10 points in the first 3 games) and Bortles came off the bench in relief in game four to lead the team on two garbage-time TD drives in the 44-17 blowout loss to the Colts.

Bortles would go on to start the rest of the regular season games with notably poor results in most games - he only had three games where he threw mores TDs than INTs. The Jags actually had a decent running game in his rookie year (4.5 ypc - 6th in the league) despite having an OL where the RT, Austin Pasztor, missed eight games with injury. His 2015 offensive line was very different (three new starters), but they at least were healthy for most of the season (all five started 13 or more games).

In 2015, he was the starter from day 1 of training camp. He now had a high priced pass-catching TE in Julius Thomas and his favorite WR, Allen Robinson, had matured with him (also in his second year) to become a legitimate threat as a #1 WR in the NFL. While he had the same head coach in Gus Bradley, he had a new offensive coordinator and a new offensive scheme (although Jedd Fisch and Greg Olsen run similar West Coast offenses).

Greg Olson had been the OC for the Faiders in 2014 when they were the second lowest scoring offense in the league. The running game was actually worse for the Jags in 2015 - only gaining 4.2 ypc - 15th in the league. So why did Bortles improve despite having to learn a new scheme, having a mostly new OL and having a worse running game? He had better chemistry with his receivers and better receivers, but the main reason was that game supposedly slowed down for him (at least during garbage time when he would rack up most of his stats). So maybe we don’t want Siemian to be like Bortles since I hope we aren’t down big late in many (if any) games this year.

Derek Carr

Carr’s stats from his first and second year

year Age GS QBrec Cmp Att Cmp% Yds TD TD% Int
2014 23 16 3/13 348 599 58.1 3270 21 3.5 12
2015* 24 16 7/9 350 573 61.1 3987 32 5.6 13
year Int% Rate QBR Sk NY/A ANY/A Sk% 4QC GWD AV
2014 2 76.6 45.7 24 5.01 4.82 3.9 1 1 5
2015* 2.3 91.1 46.3 31 6.22 6.31 5.1 4 4 13

Unlike the other three QBs that I have discussed, Derek Carr started all 16 games for the Faiders in 2014 as a rookie. He had a mostly healthy, mostly veteran offensive line to play behind and was sacked only 24 times all season (one of the lowest in the league). His OL was about all he had going for him as a rookie. His head coach, Dennis Allen, was embattled and would be fired after an 0-4 start.

His team was one of the worst in the league at running the ball (3.7 ypc - 27th) and his WRs were all “has-been”s or “never were”s (James Jones, Andrew Holmes, Kenbrell Thompkins). His offensive coordinator was the much maligned (rightfully so) Greg Olson who, despite having been an OC in the NFL for 8 seasons, had never had an offense rank higher than 15th in scoring (2010 TAM). His offenses generally rank in the mid-to-high 20’s in scoring.

Contrast this with his season in 2015. Bill Musgrave was the new offensive coordinator under new head coach Jack of the River. He still had that healthy and now fully veteran offensive line in front of him (who still couldn’t stop Von Miller). He also had a marginally improved running game (up to 3.9 ypc - 22nd in the league - thanks to Latavius Murray) and two new WRs in Amari Cooper and Michael Crabtree. Despite having to learn a new offensive system, his improved weapons and good healthy veteran offensive line would allow him to see great improvements in his passing stats in his second year. Some of his improvement has to be attributed to his maturation (game slowing down) and I’m sure if I search for it I could find some quotes from him about the game slowing down for him in 2015 relative to 2014.

What does this mean for Denver in 2017?

Well we’ve got a new head coach, new offensive coordinator, new offensive scheme and potentially four new starters on the OL. Those things would seem to be detrimental to Trevor Siemian improving this year relative to last season. We might have an improved running game this year. Given that we were 28th in ypc (3.6 ypc) in 2016, it would be hard to get worse. Having our center healthy enough to actually practice with the rest of the OL this year should help both the running game and pass protection/communication on the OL.

It’s amazing what an extra half a second of time in the pocket can do for a QB in terms of allowing plays to develop and routes to uncover leading to more 3rd down conversions. There are some quotes from Trevor Siemian indicating that the game is “slowing down” for him. If he is to be our starting QB, we can only hope that it truly is slowing down for him.

I chose not to focus on four QBs who regressed terribly during their second year (parallelism), mainly because the four worst examples of second year regression all had amazing stats in their first years as starters (Nick Foles, Josh Freeman, Chad Pennington and Robert Griffin). Trevor did not. Trevor can’t regress very much because he was already in the bottom third in most passing stats in 2016. I’m hopeful that our OL is improved which leads to a domino effect. OL improvement means a better running game. A better running game means it’s easier for our QB to throw the ball. Easier passing situations for Trevor lead to use more of the field. Using more of the field leads to better passing stats.