Last season, it was #FreeSpiller. If you're not familiar, #FreeSpiller was a trending topic on Twitter that implied that Bills running back C.J.Spiller was being figuratively imprisoned by former head coach Chan Gailey, who refused to give him many touches early in the season despite his dominant play.

Bills fans (they do exist) and fantasy owners alike took to Twitter to urge Gailey to "free" Spiller and let him become a true feature-back. This season, I sense a #FreeEllington campaign on the horizon for Cardinals rookie running back, Andre Ellington, who coaches have still not committed to starting over the thoroughly unimpressive Rashard Mendenhall. They really should just listen to the numbers.

Note: For an explanation of Net Expected Points (NEP), read the intro to this article.

NEP Studs

Andre Ellington (0.30 NEP/Rush, Ranked 1st of 51 Qualified Running Backs)

Elllington got his first start of the season in place of the injured Mendenhall last Sunday, and responded with a 15-carry, 154-yard, one-touchdown stat line. Can he be a viable fantasy starter going forward? It depends.

Ellington clearly has more speed, explosiveness, and overall running talent than Mendenhall. And while Ellington increases the Cardinals' expected point total by 0.30 with every carry, Mendenhall decreases it by 0.14 per carry.

Even before last Sunday, Mendenhall's lead over Ellington in snaps was just 216-180. However, the Cardinals coaching staff has repeatedly said that they want to limit Ellington's snaps so he doesn't get beat up. Even though he got the start, Ellington played only 35 snaps and received just 15 carries. With Mendenhall healthy, Ellington never received more than seven carries in a game.

Ellington has been involved in the passing game, however, drawing 4.1 targets per game. He's been pretty effective too, ranking 17th of 51 qualified running backs in receiving net expected points per target.

Even though Ellington has played many of his snaps on passing downs this season, a back in his mold still has the potential to be a legitimate fantasy asset. Look no further than Danny Woodhead or Darren Sproles.

The big question is whether Ellington will get enough touches. Bruce Arians has already confirmed Mendenhall will be reinserted as the early-down back when healthy, which will likely be in their next game versus Houston in Week 10. Keep in mind Woodhead averages 12.1 touches per game and Sproles just 9.4, which are both higher than Ellington's 8.1 (6.9 before his start on Sunday).

While Ellington certainly has big-play ability, those plays are unsustainable and should not be relied on for fantasy production. Ellington gained 102 of his 154 yards on two carries against the Falcons. If his touches continue to be limited to 8-10 per game, he has the potential to be a major liability in your lineup if he doesn't break a big gain.

Since Mendenhall will likely continue to be the goal-line back, and because the Cardinals suffer from poor run blocking, there will inevitably be games where Ellington won't score you many points. Our algorithms agree, as we still have him projected as an RB4 for the rest of the season. He's a high-upside, low-floor type of player, but you could do worse for a flex, especially during the bye weeks.

Terrance Williams (0.79 Rec. NEP/target, Ranked 13th of 62 Qualified Wide Receivers)

While much of the media was busy (foolishly) demonizing Dez Bryant for his sideline "blowup", Williams has been blowing up on the field. He's scored a touchdown in each of his last four games and has rendered the hamstrung and ineffective Miles Austin a non-factor in the Cowboys' offense.

Can Williams keep this up? While expecting a touchdown every game is unreasonable, Williams has been seeing a lot more pigskins come his way recently. He's averaging 8.5 targets per game in his last two, up from 4.1 in his first five. Tony Romo looks to be trusting Williams more and more lately, too. As long as the targets are there, Williams will produce. He owns an impressive 68 percent catch rate and averages 17.6 yards per reception.

At six-foot-three, Williams has the height to be a factor in the red zone. Romo has already shown proficiency at throwing the end zone fade to Bryant, and Williams could be dangerous on that same throw while facing a team's second-best cornerback.

Williams and Romo have also been deadly on the deep ball, hooking up for four completions, 160 yards, and two touchdowns on only seven attempts. Williams often gets single coverage because of Bryant's presence on the opposite side of the field, so that deep ball will likely be there for the taking all season.

We have Williams pegged as the 26th-best wide receiver for the rest of the season, and his arrow is pointing up.

NEP Duds

Tom Brady (-0.01 NEP/Pass, Ranked 23rd of 35 Qualified Quarterbacks)

Answer: Christian Ponder. Matt Schaub. Thaddeus Lewis.

Question: Name three quarterbacks averaging more standard fantasy points than Tom Brady this season.

Answering that question wasn't hard at all. In fact, there are 28 quarterbacks to choose from (minimum three games played) that average more than Brady's 12.1 standard fantasy points per game. Forget stud, Brady has barely been a QB2.

Earlier this week, JJ Zachariason wrote an excellent piece highlighting the inferiority of Brady's supporting cast as compared to last season. While Brady's receivers surely have a hand in his lackluster season, there's also a less-talked about reason he's struggling: he's not throwing the ball particularly well.

More specifically, Brady has often missed the mark on longer types of throws, such as outside passes and deep passes. First, let's take a look at his statistics on passes directed outside from 2012 and 2013:

Tom Brady on Outside Passes Only

Brady has been looking outside at nearly the same rate as last season, but his completion percentage is down 13 points. While I would argue that growing pains with rookie wide receivers Aaron Dobson and Kenbrell Thompkins - who I'll discuss later on - are responsible for some of the decline, they aren't responsible for the entire 13-point drop. Brady simply has not been as accurate.

Still not convinced? Check out how many of Brady's deep balls have been catchable this season and last:

Tom Brady on Deep (20-Plus Yards Downfield) Passes

Brady's supporting cast this season has exposed weaknesses in his game. Without the likes of a dominant slot receiver (Wes Welker), dominant tight ends (Aaron Hernandez, Rob Gronkowski), or good pass-catching running backs (Danny Woodhead, Shane Vereen), Brady hasn't been able to make things happen on his own because of his deteriorating effectiveness at throwing longer passes.

What Brady's -0.01 NEP/pass essentially tells us is that every pass he throws loses his team -0.01 points. He's been a quarterback that, devoid of his studs from last season, has been unable to add points to his team on his own with a lesser group around him. The Patriots are 6-2 because of their defense, which ranks third in Net Expected Points per play allowed.

Because of his effectiveness in the red zone, Gronkowski's return will likely improve Brady's touchdown numbers. It's also quite possible that the Patriots will be forced to throw more after losing key players on defensive side of the ball.

Losing starting right tackle Sebastian Vollmer for the season will hurt in pass protection, however. He's allowed only two sacks in 311 pass blocking snaps.

If I'm in a one-quarterback league, I'd leave Brady on the bench until further notice. If he starts stringing together multiple QB1 games, then you can think about re-inserting him into your lineup. For now, he's a QB2.

Kenbrell Thompkins (0.59 Rec. NEP/target, Ranked 44th of 62 Qualified Wide Receivers)

Thompkins has had some big moments this season, but a discouraging trend seems to have emerged. Let's take a look at his targets by game so far in 2013, as compared to Aaron Dobson, who plays the same position in two-wide sets:

Dobson has drawn more targets than Thompkins for three games in a row, and five of the past seven. Add this to the fact that Gronkowski and Amdendola are back in the picture, and there just won't be enough targets to sustain Thompkins as a viable fantasy starter.

To put the nail in the coffin, check out the number of pass routes run by Patriots wide receivers last week:

Dobson: 24. Danny Amendola: 21. Austin Collie: 19. Julian Edelman: 11. Thompkins: 6

The fact that a midseason signing, Collie, is already running more routes than Thompkins shows that the Patriots do not view Thompkins' involvement as a priority.

Remember, Thompkins was an undrafted free agent. Dobson, on the other hand, was the Patriots second-round pick. While Dobson did struggle early in the season, both receivers' respective draft positions make it obvious that the Patriots were much higher on Dobson than on Thompkins.

Looking at the tape, Thompkins appears to have a hard time gaining separation. Combine this with Brady's aforementioned struggles on longer throws, and it becomes clear that a Brady-Thompkins combination is not one New England's more effective options on offense. In fact, Thompkins' catch rate is sitting at an abysmal 41.8 percent.

I'd argue that Thompkins may be droppable in 12-team leagues and Dobson is the guy to own. Due to Thompkins' inconsistent and declining usage, he's not a viable option unless the Patriots receiving corps get depleted by injuries once again.