2016 QB Projections – AFC North & NFC North

2016 QB Projections – AFC North & NFC North

Introduction, Method, & Advance Mea Culpa

Before we plunge into the 2016 fantasy QB projections, allow me to pull the curtain back (somewhat) on my method. …

I’m open about my methods because I welcome challenges from different perspectives. In my line of work (new product engineering), it is critical the engineers on my team aren’t beholden to concepts, mechanisms, and “what we’ve always done.” I translate that into this space one-to-one. We can’t remain committed to methods for gaining an edge on competition… it’s the quickest way to fall behind. One must continually refine his or her method, gain perspective from disparate sources, and sometimes completely flip the process on its head to find a breakthrough.

I appreciate those who challenge my ideas — you are the whetstone I sharpen against. I can’t improve in a vacuum as efficiently as I will running this gauntlet. While I hope you gain something from these projections and the reasoning behind them, this exercise is equally for my own benefit. I’m here to learn, just like you. Winning leagues is a side effect.

The Method, Rev. 2016.4

1. Raw Material – Vegas Win Total

Using a set win total from Las Vegas is new this year. In prior years, I would “gut feel” a team’s win tendency based on my perception of their offseason moves, existing personnel, and existing philosophy. I recognize the folks inside the sportsbook are much better at this than I am, so I should leverage their knowledge whenever possible. As a result, I’ve built a database of team performances from 2006-2015, which you see below. In this case, I’ve filtered Wins (Ws) to include only teams that have won 6, 7, 8, or 9 games.


The responses from the data set vary wildly, but the 25th, 50th, and 75th percentiles (blue highlights) provide very actionable insight on a team’s probable direction based on win total. At this point, I will also review their schedule and projected opponent strength (via DAVE from Football Outsiders, as well as Warren Sharp’s Football Preview) to validate (or challenge) my thoughts.

2. Rough Out – Coach/Team Tendencies, Schedule Assessment

My next step is to observe the tendencies of the coaching staff in prior years and question what I see.

  • Are they pass- or run-heavy?
  • Do they tend to play from behind despite winning records?
  • Do they score more points than teams with their record should?

Once I’ve processed Step 1 and 2, I’ll commit to a play total (o.snaps) for that team. The play total is the bedrock that supports any projected player stat.

3. Lathe Work – Player Career Tendencies

Now that I have an idea how many plays a team will run in the coming season, I narrow my focus to the position/player(s) in question. Here is where my Player Screen data set comes in handy. Below, you’ll see slices of this data set for Matthew Stafford. When reviewing a player’s statistical history, I ask these questions:

  • Do they miss snaps consistently? I write off significant injuries (with missed games) as random events.
  • How frequently will their offensive coordinator call pass plays with them under center?
  • How frequently are they sacked?
  • Are they efficient passers?
  • Do they run strategically or to avoid death?


For quarterbacks, those five questions break out into nine different player rates (probabilities, if you will), which I project based on usage throughout their careers, their coach’s philosophy, and how risk-averse they’ll be required to play.


Once these rates are in place, some wand-waving occurs, and out pops a projection for that player’s season.

4. Finish Machining – Projection Refinement

Once I have a full projection completed, I will refine it against itself – that is, I look for counteractive projections and question their validity. This is the “smell test.” Did I over-influence at any point? Likewise, I will also review point output from that offense and compare it with projected Pythagorean Expectation, then adjust as necessary. It’s a controlled feedback loop. As the preseason wears on, I will adjust as injuries occur and team’s strategic deviations come to light.

Now that you’ve allowed me to lay bare my process, let’s move onto the projections!

AFC North

Baltimore Ravens

BAL.QB1Joe Flacco1046.315.5606.827.34114.629.112.726.452.70.9266.4324.7
BAL.QB2Matt Schaub32.20.518.71.1117.


Joe Flacco, prior to tearing his ACL, was on pace for 702 dropbacks in 2015, largely due to the Ravens’ porous pass defense. Since 2006, here are the ten quarterback seasons with the most dropbacks:

2012DETMS-4100Matthew StaffordQB756
2015SDPR-0300Philip RiversQB702
2011DETMS-4100Matthew StaffordQB699
2012NODB-3800Drew BreesQB696
2010INDPM-0200Peyton ManningQB696
2013ATLMR-2500Matt RyanQB695
2014NODB-3800Drew BreesQB688
2013NODB-3800Drew BreesQB687
2012DALTR-0800Tony RomoQB684
2010NODB-3800Drew BreesQB683
Top 10 QBs, ranked by dropbacks in a season, 2006-2015


Flacco’s 2015 pace would tie with Philip Rivers’ 2015 output for second-most in the last decade. Insane pass volume. Largely, we can say this is due to game script. In his 10 starts, Flacco dropped back 43 times per game while Baltimore ran 74 plays in those games. In 2013, he finished QB19 despite 662 dropbacks (QB5) on the blunt end of 19 TDs/22 INTs.

I think the Ravens will still have a poor defense. However, I do believe they will keep games closer than they did last season, when they were behind 4.1 points per offensive snap (3.9 points/snap with Flacco upright).

It’s difficult to tell whether Flacco will improve his good play-to-bad play ratio with larger volume. In 2015, he accounted for 14 TDs/12 INTs on 429 dropbacks before going on IR. In his most effective seasons (2009-2012, 2014), Flacco averaged 23 TDs/11 INTs on 555 dropbacks. It’s clear that he plays better with a lead. If Baltimore is truly the 8.5-win team Las Vegas claims, he should have that opportunity more often than not.

Cincinnati Bengals

CIN.QB1Andy Dalton1030.016.0566.525.53895.321.516.855.6139.12.5237.3280.4
CIN.QB2AJ McCarron0.


Andy Dalton, on the other hand, enjoyed cruise control up to his season-ending injury. Positive game scripts and an opportunistic defense led Cincinnati to the third-highest rate on touchdowns-per-drive (26.3%) and a 5-year low turnover-per-drive rate (10.3%), powered by a -4.9 points/snap differential.


All in all, the Bengals were able to sit on large leads, run the ball, and pass with efficient discretion. This led to Andy Dalton finishing fourth in fantasy points per dropback (0.613 fpts/drpbk) among active starters. His previous high-water mark was .478 fpts/drpbk.

I don’t expect this type of season to happen again for CIN, and I expect the 2016 Bengals to play from behind more, throw more passes, and experience similar efficiency to their 2012-2014 campaigns as a result.

Cleveland Browns

CLE.QB1Robert Griffin III675.010.0371.329.72629.914.910.245.6296.23.2192.9222.6
CLE.QB2Cody Kessler270.04.0153.910.81016.24.66.412.851.10.352.962.1
CLE.QB3Josh McCown135.


The preseason has led us to believe the Browns will play in track meets every week, with Robert Griffin III launching moonshot after moonshot to Corey Coleman and Terrelle Pryor. Based on Hue Jackson’s last two years with the Bengals, I think that is a distinct possibility.

Hue Jackson’s play split tendencies were well-documented by Chris Raybon of 4for4 over the summer, and the plots below are icing on that cake.

cin_split2h_2014 cin_split2h_2015

When he’s behind, he throws. When he’s ahead, he runs. It’s that simple.

With Las Vegas projecting Cleveland to win a mere 4.5 games in 2016, we can paint a vivid picture where the Browns throw 60-70% of the time in second halves when they’re behind, after an even play split in the first half.

Pittsburgh Steelers

PIT.QB1Ben Roethlisberger1035.014.4631.431.64618.328.416.828.367.80.0271.6328.4
PIT.QB2Landry Jones115.01.667.94.7441.


The first thing you notice in this table is I’m giving Ben Roethlisberger less than 100% of the snaps. There is a reason behind that, and it is directly related to his play style. Big Ben is notorious for patting the ball 37 times, waiting on someone to come open, with linemen draped all over him. Let’s face it, he misses snaps every season with his bell rung. In his career, he has played a full compliment of games only three times. His average snap share between 2006-2015 is 90.1%, meaning he misses just over a game’s worth of snaps every season for one reason or another.


The Steeler passing game also takes a dramatic hit when field-stretcher Martavis Bryant isn’t playing. Rather than trying to say Sammie Coates, Markus Wheaton, or the corpse of Limas Sweed will fill that void, we should admit the offense must be different, and project accordingly. Roethlisberger is a borderline QB1 once we consider this.

NFC North

Chicago Bears

CHI.QB1Jay Cutler975.015.0546.035.53675.726.215.336.5200.60.7245.7298.1
CHI.QB2Brian Hoyer71.

I’m not a big fan of John Fox. He takes the air out of the ball, plays “not to lose,” and generally ignores his best chance to win. How he will handle the Bears this season is a mystery to me. He did well getting out of the way often enough in Denver to advance to a Super Bowl; however, he still managed to damper the mood enough for John Elway to show him the door.

The numbers bear out this narrative, thus far:

chi_split_1h chi_split_2h

Despite being behind, on average, 4.5 points every snap on offense, the Broncos’ play-calling remained horrendously run-heavy. Only in the fourth quarter would the Pass Play Percentage break 55%, despite still being behind 4.3 points per offensive snap. I don’t see these tendencies changing and, as a result, Jay Cutler’s potential will remain stifled by his high interception rate and the Fisherian offensive philosophy “supporting” him.

Detroit Lions

DET.QB1Matthew Stafford1140.016.0684.037.64718.634.212.945.6114.02.1323.4391.8
DET.QB2Dan Orlovsky

It’s difficult to temper my enthusiasm for this offense, but I cannot. I see high potential for Detroit to erupt offensively in 2016. When studying their shake-n-bake bye week evolution last season, I keep coming back to the fundamental changes in their passing offense. In weeks 1-9, Joe Lombardi appeared to put Matt Stafford in -EV situations, and Stafford threw 11 INTs. Over the back half of 2015, after James Robert Cooter assumed play-calling duties, Stafford threw only 2 INTs.

What’s more, Cooter was a very aggressive playcaller, opting pass-heavy, even with leads, and even early in games.

det_split_LOM det_split_JBC

det_split1h_JBCIf you’re keeping score at home, the last two Game Split Plots indicate anywhere from 1120-1180 offensive snaps for Detroit this season, with roughly 60% being passes. If Stafford’s newfound turnover resistance remains intact, there is a real possibility he could sneak into QB6 range in 2016.

Green Bay Packers

GB.QB1Aaron Rodgers1050.016.0609.039.64669.239.68.548.5242.61.5361.0440.2
GB.QB2Brett Hundley

The entirety of Fantasy Footballdom has beaten this drum for over a year now (since Jordy Nelson’s ACL tear), but Green Bay really missed his abilities in their offense.


From the outset, Green Bay struggled to run the ball in Nelson’s absence. This, in addition to the compounded failure of non-Jordy wide receivers on the roster, drove Rodgers to an abysmal 6.5 YPA, down from a running average of 8.4 YPA between 2009-2015. That’s enough to shake faith in most quarterbacks and offensive systems.



I believe, based on the re-commitment of Eddie Lacy to game fitness, and Jordy Nelson returning to at least 90% of his former self, the Green Bay offense will return to its 2014 form.

Minnesota Vikings


With the season-ending injury to Teddy Bridgewater, and the shocking (and awesome) move to acquire Sam Bradford early Saturday morning, I have updated the Minnesota QB projections to incorporate these factors.

MIN.QB1Sam Bradford727.512.0385.627.02510.
MIN.QB2Shaun Hill242.54.0133.410.7883.

Minnesota won 11 games in 2015, but their Pythagorean Expectation indicated they should’ve only won 9.6. They face a much more difficult schedule in 2016, and with a Las Vegas win total set at 9.5, it’s safe to assume they aren’t as prepared to take the next step as many thought. I’m in agreement, and expect they will be forced to throw more in the second half of games.


min_split_2015 As you can see, in the first half, they skew run heavy relative to the pack. This is Mike Zimmer’s philosophy front-and-center, and will severely limit Teddy Bridgewater’s ability to excel in the fantasy realm.

Many conjecture that a move into a domed facility will also encourage Norv Turner to skew more pass heavy, but there appears to be a strong commitment from the organization to mirror the early 90s Cowboys – efficient passing, stout defense, and a clock-killing, demoralizing run-heavy scheme when ahead. This works if you’re ahead. If you’re behind, you’ll have to throw at some point if you want to win.

min_split2h_2014 min_split2h_2015

Second-half splits tell the true tale. If the Vikings are ahead, you will see very judicious use of the passing game. If you’re drafting Teddy Bridgewater, you’re hopeful their defense craters and they’re forced to play from more than a touchdown back throughout the second halves of games.

Wrapping Up

Thanks for taking time to read through my philosophy. Hopefully these projections and the method behind them aid in your upcoming drafts. If you have questions about the plots or statistics used in these projections, please contact me on Twitter @FantasyADHD or leave a comment below.

Josh Hornsby

Josh Hornsby leads engineering teams in the oil & gas industry. His background in new product development, combined with nearly 20 years of data-driven fantasy experience, compels him to think outside the box and wreck the echo chamber of current fantasy analysis. Josh loves to challenge popular thinking and typically does so with numbers in hand. You can find him on Twitter @FantasyADHD

Latest posts by Josh Hornsby (see all)

1 thought on “2016 QB Projections – AFC North & NFC North”

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.