The 10 Least Reliable Running Backs From 2014

Andre Williams failed to become the Giants' reliable back last season when given the opportunity.

Earlier today, I looked at the 10 most reliable running backs from 2014. Now, let's take a look at the opposite side of that list.

Before we get going, I'll reiterate what I said in the other article:

As most of you know, we use a metric at numberFire called Net Expected Points (NEP), which shows how many points a player performs above or below expectation. There's a difference between a 10-yard gain on 3rd-and-10 versus a 10-yard pickup on 3rd-and-15, after all. One picks up a first down and extends a drive, while the other more than likely results in a punt. Net Expected Points factors all of that in.

Each play on the football field can also be thought about in more of a binary fashion. Success Rate measures this -- if a player has a positive play in terms of NEP, it's deemed a success. If not, it's a failure. The percentage of "successful" runs, then, gives us a Success Rate.

Because rushing is far less efficient than passing, Success Rates are generally low. Logically, this makes sense -- it's much easier for a receiver to pick up big chunks of yards or make a hugely impactful play through the air than it is a running back doing the same thing on the ground.

Success Rate sort of shows us reliability and dependability. If a player is consistently picking up positive gains, you would naturally call that player reliable. The difference between Success Rate and regular Net Expected Points, or Net Expected Points per play, however, is that it doesn't tell us the extremes. For instance, a player may pick up a first down on 3rd-and-1 -- a positive play -- but if he runs for only one yard, that's much less impactful than if he were to run for 87 of them.

Make sense? Good. Now let's take a look at the 10 running backs with 150 or more carries last year who ranked lowest in the Success Rate department.

PlayerRushing NEP per RushSuccess Rate
Terrance West-0.0538.01%
Trent Richardson-0.0936.88%
Bishop Sankey-0.1036.60%
Andre Williams-0.0836.57%
Giovani Bernard-0.0436.31%
Tre Mason-0.0835.20%
Darren McFadden-0.1535.06%
Branden Oliver-0.1235.00%
Alfred Blue-0.1234.71%
Andre Ellington-0.1433.83%

What's interesting here is that we have a pair of running backs who saw their teammates make the other, better list from earlier in the day. Andre Williams not only had one of the lowest Rushing NEP per rush averages last season, but his Success Rate -- especially for a larger back -- was depressingly low. Meanwhile, Rashad Jennings, as noted earlier, was the 10th most reliable running back a season ago.

And then there's Giovani Bernard, who had a 36.31% Success Rate last year despite his rookie teammate, Jeremy Hill, leading the NFL within the metric. You'll notice that Bernard has the best Rushing NEP per rush of this group, though, which tells us that he is indeed a boom or bust runner -- when he makes a successful run, it's contributing more towards his NEP total than the other runners listed.

I won't bring up Andre Ellington (I guess I just did) because of his foot injury from last season, but do note that Alfred Blue was really ineffective behind Arian Foster in 2014. With Chris Polk now in the mix, Blue is no lock to be Foster's handcuff. Stephanie Stradley of the Houston Chronicle has noted the same.