Running the numbers: What the projections say (and don't say) about the Eagles' ground game

Do numbers lie?

It’s a simple enough question, and depending on who you are, it’s simple enough to answer. For me, numbers can be used as a tool for a lie, but are never themselves the lie. The power of numbers is their ability to illuminate whatever we want, even if (or especially if) we don’t use them correctly.

This is as true in sports as in anything else. For the purposes of this sport and this team, the Philadelphia Eagles used numbers to help guide the future of their run game: Does LeSean McCoy's salary match his production? How many negative runs are too many? What kind of replacement can the Eagles get for a similar salary?

Where numbers can mislead us is when we use them as a predictor for events that are fueled by the human element. And let’s be real, the human element is the reason we watch football. As writers and fans, we use numbers as a predictor for things that other humans haven’t done yet. Hell, fantasy football fortunes are made and lost by these calculations.

I started this article with a simple goal, and believe it or not, the goal wasn’t to bury the lede. It was to predict what the Eagles’ run game would look like in 2015. But as I very quickly learned, predicting that human element is nearly impossible – and as a result, extremely exciting – when it's controlled by Chip Kelly.

History, DeMarco Murray and You

After the Eagles signed DeMarco Murray away from the Cowboys in one of the greatest combinations of team need, fit and division rival trolling in NFL history, the conversation quickly turned to the darker side of Murray’s stellar 2015 season.

As one of only a handful of players in the modern era to have run for more than 1,800 yards on at least 4.7 yards per attempt, Murray has proven himself to be one of the elite backs in the NFL. And while his name is admittedly near the bottom of that list for yards, he tops it in a much different category – he made the list on 392 carries, tied for the most of any of the 17 players on the list.

But it's more than just one season that can wear down a running back. Seasons of the bellcow workload can shorten a player's career. In Murray's case, last year’s eye-popping number of carries come as the high point in a career that has been prolific. Since 2000, Murray is one of only eight players to run at least 900 times for at least 4,500 yards and 28 touchdowns in his first four seasons.

So, in trying to determine what history says about Murray’s chances for a successful 2015, I looked at how the other seven players fared in year number five:

Name	Year	Games Played	Games Started	Attempts	Yards	Yards Per Attempt	Touchdowns	Yards Per Game
Arian Foster	2013	8	8	121	542	4.5	1	67.8
Chris Johnson	2012	16	15	276	1243	4.5	6	77.7
Adrian Peterson	2011	12	12	208	970	4.7	12	80.8
Willis McGahee	2008	13	8	170	671	3.9	7	51.6
Clinton Portis	2006	8	7	127	523	4.1	7	65.4
Jamal Lewis	2005	15	15	269	906	3.4	3	60.4
LaDanian Tomlinson	2005	16	16	339	1462	4.3	18	91.4

As you can see, the story the numbers told wasn’t so much an epic poem as it was a big ol' ¯\_(ツ)_/¯ emoticon. And as I mentioned earlier, the numbers don’t give the whole story. You need the human element for that.

Of the two running backs who indisputably had great fifth years, only the immortal LaDanian Tomlinson broke the admittedly arbitrary 1,000 yard mark and also ran for double-digit touchdowns. Chris Johnson ran for 1,200 yards but only recorded six scores. And of the running backs who didn't have triple-digit rushing totals, only Adrian Peterson had more than 10 touchdowns.

So that doesn’t bode well for Murray, right? Well, it doesn't look great. Take Arian Foster for example. For the first time since becoming a full-time starter, Foster’s 2013 season was the only one of his career in which he did not run for 1,000 yards. And it’s because he got hurt midway through the year. Clinton Portis suffered a similar fate in 2006.

Okay, so how about the guys who played at least 13 games? That leaves our two 1,000 yard rushers, Peterson, Willis McGahee and Jamal Lewis. McGahee played in 13 games but only started eight after a series of injuries found him in a platoon with a rookie Ray Rice and All-Pro fullback LeRon McClain. And while the bruising Lewis wasn’t as effective in 2005 (it was one of only two years in his career in which he didn’t break 1,000 yards), rushing for 906 yards is nothing at which to sneeze.

On one hand, sports science could have a very positive impact on Murray this season, and could be the ultimate test for how well it prevents the breakdown of the body after significant wear and tear. On the other hand, the injuries that slowed down Portis and McGahee occurred in bones, which unfortunately are immune to sports science. It's a stark reminder that injuries can be unpredictable, and even the most forward-thinking prevention programs aren't foolproof.

But the one wild card in all of this is that Murray will have the benefit of a multi-talented backfield, one that could very well limit his chance of injury.

The Platoon

Unlike the seven players on that previous list, Murray will have two Pro Bowl players with whom he will share the backfield. Thanks to an offseason that felt reminiscent of one your younger siblings might have in Madden, Kelly signed two running backs with three Pro Bowl appearances between them. Now, Murray will share carries with former Charger Ryan Mathews and last year's addition, Darren Sproles.

So how will this revamped, $11 million backfield share the load? For a better idea, let's take a look at how Chip Kelly, a coach famous for his belief in building around the run, actually uses his backs. Below is a list of how the top three running backs performed in each of Kelly's years as head coach, both in college and the pros:

Year	Name	Attempts	Yards	Yards Per Attempt	Percentage of Carries
2009	LaMichael James	230	1546	6.7	67.45
	Kenjon Barner	61	366	6	17.89
	Andre Crenshaw	50	152	3	14.66
2010	LaMichael James	294	1731	5.9	65.63
	Kenjon Barner	91	551	6.1	20.31
	Remene Alston	63	356	5.7	14.06
2011	LaMichael James	247	1805	7.3	54.41
	Kenjon Barner	152	939	6.2	33.48
	De'Anthony Thomas	55	595	10.8	12.11
2012	Kenjon Barner	278	1767	6.4	60.83
	De'Anthony Thomas	92	701	7.6	20.13
	Byron Marshall	87	447	5.1	19.04
2013	LeSean McCoy	314	1607	5.1	78.50
	Bryce Brown	75	314	4.2	18.75
	Chris Polk	11	98	8.9	2.75
2014	LeSean McCoy	312	1319	4.2	75.18
	Darren Sproles	57	329	5.8	13.78
	Chris Polk	46	172	3.7	11.08

Yikes. Let's try to make sense of this, because this is when using numbers to predict the future gets tricky.

In his six years of head coaching experience, the top three running backs under Kelly have shared the ball a fairly consistent amount: the number one guy gets around 67 percent of the carries, the number two runner gets around 20 percent, and the number three gets around 12 percent.

Interestingly, Chip has fed the number one runner more in Philadelphia than he did at Oregon. There are a number of possible reasons for this. One is that LeSean McCoy is and was better in 2013 and 2014 than any of those backs were in college. Yet another is that McCoy’s backups were Chris Polk and Bryce Brown in the NFL, not Kenjon Barner and De’Anthony Thomas against New Mexico. And finally, Oregon had dual-threat quarterbacks who were a threat to take off at any moment, and picked up large chunks of yardage on their own. In fact, Kelly's quarterbacks usually ranked second in rushing on the team.

Predicting The Future

With these figures in mind, the best way to predict what the 2015 run game might look like is to take the average number of carries each runner on the depth chart received during Chip’s time as head coach, and multiply it by the current players’ 2014 yards per carry number. Since 2009, the top three running backs in a Chip Kelly offense split an average of 419 carries per season.

Name	2014 YPA	Projected Attempts	Projected Yards
DeMarco Murray	4.7	280	1,316
Ryan Mathews	2.6	86	223
Darren Sproles	5.8	52	301
Total		418	1,840

In a word, the results are suspicious. In a few more words, the results seem very unlikely, given what we know about the talent of these runners, Chip’s philosophical belief in the run game and a hard-to-shake feeling to the effect of there is no way Murray, Mathews and Sproles run for fewer yards combined than Murray had by himself last year.

This is especially true of Mathews, the wild card of the bunch. He finished his injury-riddled 2014 season with 2.6 yards per attempt, but even if you used his career average (4.4), the projection doesn't improve dramatically. These numbers immediately made me think back to an interview Bolts From The Blue's John Gennaro gave to our own fearless leader, Brandon Lee Gowton. When asked what he saw as Mathews' role in Philadelphia, Gennaro replied:

With the Chargers, Mathews was put into a role of "1st and 2nd-down RB until we get to the red zone." That job likely won't be open with the Eagles, and it's not like he's going to beat out Sproles as a 3rd down option.

My guess is that he becomes the "2nd half workhorse" that just pounds the ball relentlessly if the Eagles have a lead and are looking to burn the clock, and I think he'll be good at that, but I wonder how many games he'll finish with 4 touches for less than 10 yards.

If Gennaro is correct (and it seems prudent to defer to his expertise until Mathews actually, y'know, plays in Philadelphia), then the projections make a little more sense. But there's also the idea that Kelly, perhaps more than other coaches, likes to play with mismatches. And who's to say that the speed of Mathews and Sproles won't present unique opportunities for him?

In year two, the league saw a clearer version of Kelly's vision on the field. Year three figures to be an even more clear vision, and it's going to include these three backs. No one could have predicted this backfield in the first place (including the Eagles' coaching staff), so there will be a built-in level of mystery that could last even in to the winter.

And this, my friends, is the beauty of football. The numbers we can run may give us a glimpse, but not the whole picture. In many ways, the numbers leave us with more questions than answers. Will Mathews really end up with 86 carries, or will he end up in the mid-100s range, as Zach Berman of the Philadelphia Inquirer recently predicted? After a historic season, will Murray end up eating less of the meat on the bone? Will the Eagles use Sproles more as a runner this year? Is Kelly going to hitch his wagon to an offense built on a three-headed running attack?

It’s what makes Chip Kelly and this Eagles offense so fascinating. You can crunch the numbers and chart the projections. But at the end of the day, the numbers can't tell us what we want to know; not entirely.

At least, not when it comes to this team.