Lamar Miller, Daniel Thomas and the Running Back Projection Model
The Lamar Miller hype train already has plenty of passengers. That much is clear. He’s creeping up to the top of the 3rd round in redraft and he’s coming off the board as RB17 in dynasty leagues. Heck, he even went 28th overall in our RotoViz Dynasty League (hosted by Fleaflicker.com)! With a skyrocketing price, the question for Miller owners becomes: have they bought a ticket to Fantasy Studville or are they on a train to nowhere? Let’s find out…
First Stop: The Numbers
I think one reason people clutch their Lamar Miller ticket so nervously is because of his (lack of) production in his rookie season. That’s understandable when you consider that despite being touted as one of the more talented runners in the 2012 draft class, Miller finished behind 10 other rookie running backs in yards from scrimmage:
Obviously a big driver of the lack of stats is Miller’s sample-low 51 rookie carries. To put a positive spin on it, perhaps you’d point out that Miller’s 4.9 yards per carry average is tied for best along with two other backs who are generally viewed as talented players with a chance to break out if given the opportunity: Pierce and Brown. Before you decide to stand behind that argument, take a look at this PFF article by Mike Clay where he shows that even above-average rookie running back efficiency doesn’t correlate strongly with future success.
Now that I have you considering jumping off this train, let me give you some good news. One of the biggest reasons I’m confident that Miller can be a solid fantasy producer is his college market share numbers. In developing my running back model, I’ve found that college running backs who account for around 60% or more of their college rushing stats are more likely than not to produce a top 30 fantasy season.
For me, the logic behind the larger-market-share-is-better concept goes like this: most if not all football teams, be they college or pro, have multiple running backs to carry the rock. Some of this is because running backs have different skill sets and therefore fill different roles in the offense. But I’d argue teams use multiple backs mainly because of the brutal nature of football and the amount of physical punishment that running backs take. Overusing one back is basically guaranteeing he’ll be injured. Spreading carries between multiple backs reduces the workload and hopefully preserves the health and effectiveness of a team’s runners.
Since there are several players who share the rushing workload (and the beating that comes with it) and they’re all playing in the same offense and running behind the same offensive line, the only thing left to distinguish their productivity from one another is their relative talent levels. If we assume rational coaching, the most talented runner will receive more touches than his teammates because he’s the most likely to gain positive yards on any given carry. Those additional carries lead to more yards and more touchdowns for the talented player relative to his teammates. To boil it down to a sentence: the more talented a runner is, the more workload he will get. He will dominate his team’s yards and touchdowns.
Now let’s bring it back to a concrete example. In Miami’s 2011 offense Miller received 277 carries, a full 155 carries more than the number two back Mike James. Miller averaged 5.9 YPC on his touches while James managed only 3.8. Because of his superior talent, Miller dominated the Miami rushing attack accounting for 72% of the rushing yards and 50% of their rushing TDs for a Dominator Rating (DR) of 61%. I’d argue that the 2013 NFL draft also confirmed the disparity in talent when James was chosen in the back half of the 6th round.
Speaking of the NFL draft, another important factor in my running back model is draft slot. Generally, I like to see running backs drafted in the third round or higher. When teams burn a high draft pick on a player it tells me two things 1) they thought highly of his talent (and assumed other teams would too) and 2) they probably intend to use him early and often. My model doesn’t like that Miller fell to the fourth round, but this is a situation where the slide was most likely due to uncertainty about Miller’s shoulder injury, rather than a concern about a lack of talent.
Miller sustained a “shoulder injury” (details of the specific injury were hard to come by despite a lot of Google searching) in the third week of the 2011 season against Kansas State. Despite the lingering injury, Miller played in the remaining 9 games that year averaging 18.3 carries, 95.8 yards, and .7 TDs per game. (To me, it makes his domination of the Miami ground stats that much more impressive.) He needed surgery in December of 2011 to fix the shoulder issue and it limited his participation in some of the scouting combine drills. While we can’t know for sure why he fell to the 4th round, consensus opinion among draft analysts suggests his injury had a lot to do with it:
Lamar Miller falling likely thanks to his need for surgery after the draft, from what I heard. Trying to find where surgery is required.
— Eric Galko (@OptimumScouting) April 28, 2012
Miller is a perfect example of a scenario where I’m willing to look past a lower draft slot when there is a pretty clear non-football-talent-related reason a player fell from an otherwise high spot. As much as it pains the scout in me to say so, Johnathan Franklin is the counterexample – a player who seemingly fell without any overt injury concerns.
Second Stop: Athletic Measurables
Let’s start with the official combine numbers, since those are…well…official, and are better suited for comparison against other players since they happen in a somewhat controlled environment (concurrent timeframe, identical field surface, official timing equipment, etc.) The only athletic measurable that my model uses (thus far) is speed score, which is essentially a way of describing how fast a player is when you control for his weight (smaller guys should run faster than bigger guys). Miller weighed in at 212 lbs at the Combine and clocked an official 4.40 forty yard dash, giving him a speed score of 113, which is very good. For context, a speed score of 100 is average with anything above being fast and below being slow. Speed scores approaching 120 are exceedingly rare and represent the biggest of big-play threats. Miller’s 113 is well above average and indicative of the speed and burst that draftniks describe when talking about his skill set.
Miller didn’t participate in the broad jump, 3 cone drill, or the short shuttle at the Combine, but he did attend Miami’s Pro Day. I always take Pro Day numbers with a large grain of salt, since they’re usually done on a player’s college “home turf” (both literally and figuratively). The surfaces they run on might be faster than what players run on at the combine. Plus, I’m not even sure how “official” the timers are at these events, and in athletic measurement every hundredth of a second counts. But with that said, Miller did put up some impressive numbers at his Pro Day. Here are Miller’s numbers alongside his primary backups Daniel Thomas and Mike Gillislee:
|Name||Data Source||Height||Weight||40||Vert||Broad||Shuttle||3 Cone||Agility Score|
|Lamar Miller||Pro Day||5’10”||212||4.4*||35.5″||120″||4.08||6.94||11.02|
|Daniel Thomas||Pro Day||6’02”||230||4.62||34″||123″||4.29||7.06||11.35|
*Miller did not run a 40 at his pro day, official combine 40 time.
Miller’s broad jump (usually associated with a player’s “burst”) was tied with Doug Martin’s official number. His Agility Score (which Shawn Siegele has found to be correlated with running back success) is actually better than David Wilson’s. Most importantly, with the exception of Thomas’ broad jump, ALL of this athletic measurements (which you could refer to as his Explosion Index) are better than those of both Thomas and Gillislee. Which leads us to the next stop…
Third Stop: Competition
Ladies and gentlemen the train will be rolling through this stop, because Miller isn’t facing much competition at all. Daniel Thomas has been in the league for two full seasons and has been pretty terrible. Despite being drafted by the Dolphins in the 2nd round in 2011 he showed signs of being a bad pick right from the start. Let’s not forget Thomas was a quarterback before he got to Kansas State. Apparently, making the transition from QB to college RB to NFL RB, is not an easy one. During the preseason he was already on the outs with the coaching staff that drafted him because, despite his 6’0” 230 lb frame, he was running timid at the goal line. Things never got much better for Thomas from there. He was supposed to carve out a significant portion of a timeshare but got left in the dust by a 26 year old Reggie Bush who claimed the Dolphins’ lead back role for himself. On his 256 total attempts he’s averaged only 3.5 YPC and 4 rushing TDs. Thomas will be 26 in October and despite some reports that he’s locked up the backup job, he’s probably practicing alongside his eventual replacement, 22 year old Mike Gillislee.
Despite some draftnik love for Mike Gillislee, he fell to the end of the 5th round. Using my projection model, which takes Draft Slot into account, he scores an abysmal -0.30. Add in the fact that he’s smaller, slower, and less explosive than Miller and I just can’t see how Gillislee earns any playing time over a healthy Miller. Heck, at this point it sounds like he’s even behind Thomas on the depth chart. My point here is not that Thomas and Gillislee have zero talent or are completely worthless as real life backups or fantasy handcuffs, but rather that neither poses a legitimate threat to Miller’s playing time, as the Miller-doubters might have you believe.
Fourth Stop: Sim Score Projection.
I love using the Similarity Score Apps for a glimpse at what we can expect from a player in 2013. What makes these apps especially cool is that you can customize the inputs to explore the range of possible outcomes for a player. For Miller’s projected stat lines and fantasy points I looked at three variations from the most conservative, to the most aggressive.
Conservative: No low-usage games removed and the games played penalty left at 0.5
Medium: No low-usage games removed, games played penalty dropped to 0.
Aggressive: Low-usage games (<9 carries) removed, games played penalty dropped to 0.
Keep in mind that all of these projections assume a range of 10 – 12 carries per game. Reggie Bush averaged 14.3 carries per game during his stint with Miami and I see no reason Miller couldn’t approach that workload level. Ultimately, that makes all three of these projections relatively conservative and speaks to the major upside that Miller has this year and beyond.
Fifth Stop: Positive Media Buzz
While this stop isn’t very analytical, the scenery sure is pretty. You need look no further than Miller’s Rotoworld news page to find blurbs littered with positive sound bites:
5/26/13 – The Miami Herald says “there is a quiet confidence” at Dolphins camp that Lamar Miller is ready to break out.
6/4/13 – “He’s doing a great job, really has a great handle on the offense and he’s doing a great job protecting,” Tannehill said.
6/13/13 – Coach Joe Philbin says Lamar Miller has “really progressed” over the course of OTAs and minicamp.
6/23/13 – Ryan Tannehill says “what impresses (him) most” about Lamar Miller is the second-year back’s improvement in pass protection.
And from Frank Gore on 7/15/13 – “Lamar is probably the fastest,” Gore said when comparing Edgerrin James, himself, Portis, and Willis McGahee. “He reminds me of Portis.” Per the Miami Herald, the Dolphins “privately” believe Miller can develop into “not merely a good starting back, but a great one.”
Ladies and gentlemen, we’ve arrived at our Final Stop: Fantasy Studville. I hope you’ve enjoyed the ride.