Over the next several posts I’ll be running through some of the more convoluted NFL backfields. I’ll be using my running back model to project who has the best chance to produce fantasy points. I informally kicked off the series with a post advising caution in assuming Zac Stacy has a high likelihood of being the Rams’ most productive back, and I wrote a follow up piece looking at the dark horse in that race – lesser known UDFA Benny Cunningham. Here I’ll take a crack at one of the most vexing backfield situations for both dynasty and redraft players heading into 2013, the Green Bay Packers.
Preempting the analysis of the players involved here is the fact that, as fantasy owners know, the Green Bay backfield has been a subpar place to find running back production over the past several years. In the 7 NFL seasons since Mike McCarthy took over as head coach, Green Bay has only produced 3 top 20 RB seasons: Ahman Green (2006) and Ryan Grant (2007, 2009). The league average over that time span is 4.375 top 20 RB seasons per team, and the only teams to produce fewer than 3 top 20 RB seasons during that span were bad teams for most of that stretch (DEN, DET, OAK, SEA, TB). OK, maybe DEN was just mediocre, but the rest of those teams were categorically bad.
While we could debate whether it’s strictly the offensive philosophy, or a lack of talented players to execute a more balanced attack, we need look no further than the run/pass (im)balance of the Pack under McCarthy to answer the question of why his backs don’t produce stats. Using Pro Football Reference, I looked at Green Bay’s offensive tendencies both inside and outside of their opponent’s 10 yard line. On non-4th down plays outside of the opponent’s 10 yard line during McCarthy’s tenure as head coach, GB’s 41/59% run/pass ratio ranks 25th (i.e. only 7 teams ran less often than the Packers). And on non- 4th down plays with goal-to-go, the Packers rank DEAD LAST in those seven seasons, attempting to rush the ball only 44% of the time when their fantasy players are in prime scoring position. The average NFL team rushed 53% of the time in that same situation. To add further insult to injury, of the 221 goal-to-go rushes they attempted, OVER THIRTY PERCENT of the carries went to Aaron Rodgers and John Freakin’ Kuhn. Talk about a kick in the fantasy junk.
It appears Green Bay wants to change that trend in 2013. They drafted Eddie Lacy at the end of the 2nd and then, in an un-Thompson-like move, traded up to draft Johnathan Franklin in the 4th (two unthinkable slides based upon the analyst community’s pre-draft rankings). Even before that they’d made overtures, signing Cedric Benson, drafting Alex Green in the 3rd round, and later finding some success with their street free agent, turned practice squad player, turned starter: DuJuan Harris. But with so many runners who all bring different skills, who among these guys projects as having the best shot to succeed? Here’s where my model can help.
My projection model is based on three factors: draft slot, rushing Dominator Rating (DR) – a market share concept borrowed from Shawn Siegele’s work on wide receivers – and Speed Score (basically how fast a player is when controlling for his weight). I’ve found that successful running backs tend to be drafted in the third round or higher, account for 65% or more of their team’s rushing yards and touchdowns in their final collegiate season, and have an above-average speed score (over 100). The higher a player ranks on all three of these metrics, the more likely he is to succeed in the NFL and produce stats for fantasy teams.
Draft slot is a pretty big factor in my model because it usually explains a good deal of the success a running back might have. To me, draft slot represents what a team thinks about the player’s skill set and how they intend to use him. But since drafting is an imperfect science, it’s also worthwhile to remove the draft slot factor and re-normalize the Rushing DR and Speed Score factors to see how a running back ranks in a draft-agnostic (DA) model. That way, if a late-round or undrafted guy ends up with a starting gig for one reason or another, you can set aside talent/usage part of the equation and looking at how successful he might be based on his college production and athleticism (as approximated by Speed Score). For example, Alfred Morris scores poorly on my normal (draft included) model mostly because he fell all the way into the sixth round. But he scores a ridiculously high 1.80 on my DA model because he dominated the rushing attack at Florida Atlantic. His Speed Score was actually a little below average, but his DR more than makes up for it. Give him an opportunity in Washington, and voila.
Here’s how the Green Bay backs rank in my model, followed by my take on each guy (DA Score stands for Draft Agnostic Model Score):
|Player||Model Score||DA Score||College Year||School||Draft Slot||Height||Weight||Forty||Speed Score||Rush DR||Vert||Broad||Shuttle||3 Cone||Agility|
Alex Green – You can continue reading what I had written about Green, or you can just save yourself some time and read what Shawn Siegele just wrote, which is a much more compelling take. But for what it’s worth, here’s what I had to say: Green was a dominant force in the 2010 Hawaii rushing attack, accounting for more than three quarters of its total production. I know Hawaii runs a spread offense and perhaps anyone could have dominated rushing yards if they had Mack truck sized holes to run through. You could also quibble with the fact that Hawaii only averaged 22 rushing attempts per game in 2010, which may be another reason it was easy for one guy to dominate the production – 10 carries per game gets you to an average market share. But for better or worse, Green was the guy. In the speed department, Green boasts a forty time only one tick off of DuJuan Harris’ despite weighing 28lbs more. From a talent/usage standpoint, the Packers thought enough of him to draft him in the third round, ahead of guys like Roy Helu and Kendall Hunter. They started working him into the rotation slowly during his rookie season as Starks and Grant battled for carries. Unfortunately, Green suffered a torn ACL in week 7, landing him on IR. He still wouldn’t be quite right by the time 2012 training camp started, but he felt an urgency to get on the field. He admits he was never fully healthy, which might explain why he was so underwhelming when he had his chance to claim the starting job after Ced Benson suffered his Lisfranc injury in week 5. Supposedly, he’s back to full strength now and ready to battle for his roster spot. A lot of whether he succeeds or fails will likely come down to health – both his and that of the guys in front of him. If he’s 100% and he can show off his elite athleticism, perhaps he can return to his Hawaii form and earn some carries. Or if Lacy’s history of nagging injuries continues, he may get an opportunity by default. At the very least, I wouldn’t count him out of the running just because the team drafted two rookies this year.
Eddie Lacy – Lacy is just a mediocre prospect using my model. Alabama continues to use their platoon system, continually rotating in two workhorse backs behind their perennially dominant offensive lines. And time and time again, the scouting community seems to agree that these are high quality running back prospects. My model isn’t buying it. It’s possible that a backfield run this way will produce a truly elite prospect to whom my model won’t give enough credit. Then again, Trent Richardson actually grades out pretty well (a starter-caliber 1.26) since he managed to claim about 60% of Alabama’s yards and touchdowns, even while splitting time with Lacy. To Lacy’s detriment, T.J. Yeldon managed to steal away 34% of the 2012 Tide rushing market, probably because he’s a much better talent. Lacy’s speed score using his pro day 4.55 forty actually isn’t bad at 102.5. You could even make the excuse that he was slower than he could have been because of the hamstring injury, or the toe injury…What’s hard to ignore is that the Packers drafted him at the end of the second round. They obviously like what he brings to the proverbial table and it’s likely they plan to use him a fair bit, presumably near the goal line…err, I guess. TDs are what fantasy football are all about, and in all likelihood that will be Lacy’s calling card since he’d project to take over the John Kuhn carries. But between GB’s historical RB usage, the fact that Lacy wasn’t dominant in the Alabama running game, and the fact that he’s all but guaranteed some degree of a timeshare, he just doesn’t excite me as a top 5 rookie draft pick.
Johnathan Franklin – I’ll admit it: I loved Franklin on tape. I really wanted my model to like him because my eyeballs did. He looked agile yet patient, powerful yet fast. He could catch. He could make guys miss. What more could you want? Well, a better draft slot, for one. There’s a chance he fell for fluky reasons (Sigmund Bloom envisioned it in one of the FootballGuys podcasts, saying he might have been the next back on Bengals’ board behind Bernard, and the next back on the Steeler’s board behind Bell, and so on). But more than likely, teams just didn’t view him as highly draftable. And I’ve found that tends to make it less likely for a running back to be a fantasy success. His 55% market share of UCLA’s rushing yards and TDs is actually above average, and his Speed Score is fine. Give him an equivalent draft slot to Lacy and my model would actually prefer Franklin. I am pulling for Franklin because the scout in me wants to be right, but at the end of the day he’s not the biggest, the fastest, the most agile, or even the most productive running back in this group. Maybe he’ll be the guy who has a knack for ripping off long TD runs? Perhaps he’ll be the healthiest and earn a role by default? Unfortunately, the numbers just don’t suggest a high likelihood of success.
James Starks – Starks is the geezer of the bunch at 27, which tends to be the beginning of the end for running backs. However, my model does like him (at least it does if you ignore the fact that he wasn’t drafted until the 6th round). He was 65% of the Buffalo rushing market it 2009 and his 106 speed score is better than average. He’s flashed at times for Green Bay as we all remember, with his 123 yard performance against Philly in the wildcard game. But realistically, he is probably the least likely to do anything productive since he’s been unable to claim a starring role during his first three years in the league and now he has all of this fresh young talent behind him.
DuJuan Harris – Let’s start with the positives. Harris was the most effective running back on the Packers last year with his 4.6 YPC. He also managed to tie for the most touchdowns of any running back on the team with…wait for it…two. In the “push” category, Harris is the fastest of the bunch but he should be since he’s also the lightest – no extra credit, yet no dings for the 101 speed score. But on the negative side, his Rushing DR is very low, which tends to be the case with smaller/speedier running backs. He did miss one of Troy’s games in 2010, so if you look at his DR for the games he played in it improves – but only to .31. It’s not impossible Harris enjoys fantasy success in the Green Bay Backfield. But based on the numbers, it’s very unlikely.
If it seems like I haven’t given you a clear answer, you’re right. My model doesn’t love any of these guys Green probably has the most upside, Franklin the least downside, and Lacy the most potential to affect our collective perception of Alabama running backs. But ultimately it could be health that determines who is the best fantasy option of this bunch. If you’re drafting any of them expecting a lot of touchdowns, you’d better hope for a major change in Green Bay’s goal line tendencies. Otherwise, expect a kick in the fantasy junk.