15 Fantasy Football Transactions for Week 4

Sometimes, things don't go your way.

That was the case for yours truly on Sunday.

No, I didn't lose my dog. All of my friends are OK, and my bank account didn't get hacked. Don't think things were really bad for me, unless you consider your Twitter mentions being lit on fire as "really bad".

My quarterback picks were not good. They were wrong. I was all into Nick Foles in a matchup against a terrible Steelers' secondary that had allowed 10 top-10 quarterback performances over their last 11 games. And I wasn't into Andy Dalton, who's struggled mightily in divisional games over his career, or Derek Carr, who was a road underdog against a decent Browns' secondary.

It could've been a lot worse, because it's just fantasy football. But when it comes to this column, it wasn't a flawless Sunday. Let's improve on that here in Week 4.

Sell Devonta Freeman

It was pretty obvious that Devonta Freeman was about to lose significant touches to Tevin Coleman when the season was getting started. In Week 1, Coleman out-snapped the second-year back 44 to 29, and if not for an injury in Week 2, we would've more than likely seen the same.

Freeman had a solo audition in Week 3, and he thrived. Against the Cowboys, he added over eight expected points -- per our Net Expected Points (NEP) formula -- on the ground, culminating in a Rushing NEP per rush average of 0.28. For reference -- that's absurd. Of all backs with 30 or more carries last year, Darren Sproles' 0.21 Rushing NEP per rush ranked first. Freeman was even better than that.

I'm not knocking his Week 3 game at all -- he balled out. But what I'm worried about with Freeman is the fact that he was the worst 50-plus attempt running back in the NFL on a per rush basis last season according to our numbers, and that Tevin Coleman, the team's third-round pick in May, is going to return to the lineup in the coming weeks. His timetable to get back on the field is still up in the air, but reports stated it shouldn't be anything longer than a few weeks.

With Coleman out yesterday, Freeman had the backfield all to himself in what was considered to be a plus-matchup. Per FantasyData.com, Freeman's 65 snaps played were the fifth most in a single game by any back in this young season. Volume was his friend.

Freeman will only be the lead back for a week or two more considering all of this. And upcoming for Atlanta is Houston and Washington -- Houston's been fairly average stopping the run so far this year (12th according to our numbers), but Washington, like last season, has been stout (6th). In other words, it's no lock that Freeman performs even without Coleman over the next couple of weeks -- if that's what it comes to -- and when Coleman returns, we should more than likely expect a full-blown running back-by-committee system.

If there's a league member of yours who's buying Freeman after his incredible Week 3 performance -- and I'd be surprised if there weren't, considering he's the second best PPR running back in fantasy football right now as a result -- I'd be selling.

Add Karlos Williams

If Karlos Williams is on the waiver wire in your league, you should more than likely make him your top priority this week. LeSean McCoy has a hamstring injury which should keep him out for at least Week 4, which means Williams, with not much behind him on the depth chart, should get the bulk of the work in the Bills' backfield.

So far this year, no running back with 10 or more carries has a better per-rush NEP average than Williams, who's now scored in each of the Bills' first three games this year. The Bills, too, have the lowest pass-to-run ratio in the NFL, so you know they'll be willing to hand off the rock to their rookie runner. Everything is in place for a week or two of really strong production, with the upside that McCoy's injury lingers all season long.

Sell Joseph Randle

Like Freeman above, Joseph Randle had a great Week 3 performance.

Or did he?

This is a classic case of letting fantasy numbers skew the way a player actually performs. In Week 3, Randle scored three touchdowns, making him one of the best fantasy options at running back. And from an efficiency perspective, he finished with a 0.17 Rushing NEP per rush, which is nothing to scoff at.

But Randle made all of this happen with unsustainable big plays. A deeper look at his 14 carries shows that just 7 of them were for positive NEP gains, good for a Success Rate of 50%. While a Success Rate that high is very good (the average among running backs hovers the 41 to 42 percent range, typically), it's not as great when you consider how efficient he was. In other words, his averages looked good on Sunday because of splash plays.

The other reason to not feel bullish about Randle is because he's clearly not the receiving back on the team. He saw just two targets to Lance Dunbar's 10 (more on him in a second), and when the game script flipped to a negative one, Randle didn't even see the field. In fact, his last touch in the game against Atlanta was with 4:39 left in the third quarter. That very next drive was when the Falcons took the lead, and when Dunbar took over. At the end of the game, Dunbar actually played 10 more snaps than Randle.

Dallas gets New Orleans, New England, and Seattle in three of their next four games. Even though we may view New Orleans as an easy opponent (they are), Vegas has them as four-point favorites over the Cowboys in Week 4. Meaning at least three of the Cowboys next four games (they face the Giants in between) could feature a negative game script, and not Joseph Randle.

Add Lance Dunbar

Well that's an easy transition.

As I mentioned, Dunbar served -- and will serve -- as the main pass-catching back for the Cowboys. We saw the impact that can have on the fantasy football gridiron this weekend -- Dunbar ended up catching 10 passes on 10 targets for 100 yards against the Falcons. Without even scoring a touchdown, Dunbar was a high-end RB2 in PPR leagues this week.

Not only is Brandon Weeden more than likely going to look to check down to Dunbar and play things safe in order to manage the game, but as I mentioned, the Cowboys are set up to have plenty of games with a negative game flow moving forward. Given that, Dunbar's going to be a potential PPR machine for Dallas.

Sell or Hold Jeremy Hill

Getting rid of Jeremy Hill now feels like the ultimate sell low. And that's exactly what you're looking to not do in fantasy football.

But sometimes situations don't dramatically improve for players who were highly touted entering the season.

Jeremy Hill has a ton of talent, and led higher-volume backs in Success Rate last season, his rookie year. But the Bengals just aren't using him like Hill owners want them to. Through three games, Hill's played 52.9%, 30.3% and 33.8% of the team's snaps, respectively. Giovani Bernard, meanwhile, has seen 44.1%, 68.2% and 66.2% of Cincinnati's snaps.

In other words, the Bengals are using Bernard more than they're utilizing Hill.

Part of this is bad luck and matchups. In Week 2, Hill was benched for his fumbling problem, and on Sunday versus Baltimore, the Bengals were kind of in shootout mode, playing a team that's been much better at stopping the run so far than this year than stopping the pass.

The problem is that the team's upcoming schedule kind of dictates a similar game plan. Over the next four weeks, the Bengals get the Chiefs (2nd against the run according to our metrics, 18th against the pass), Seahawks (3rd, 11th), Bills (27th [skewed by quarterback data], 7th) and Steelers (8th, 23rd). I know I won't be surprised if Hill struggles through that stretch.

And, you know, all of this information doesn't even mention the fact that he's played like doo-doo. Among backs with 30 or more carries this year, only C.J. Anderson has a worse Rushing NEP per rush average. And Hill's Success Rate is a poor 35%, down about 15% from last season. Things aren't exactly looking up. If you can get a reasonable return, I'm for it. Don't go trading him for trash, though.

Add Derek Carr

If you were to tell me that, through three weeks, Derek Carr would be the eighth most efficient passer in the NFL per our NEP metric, I'd laugh at you. Because I've admittedly been a Carr hater, and really didn't see this kind of start coming, even with the weapons he now has.

I'm not ready to dub Carr a great starting quarterback in the NFL, and I'm certainly not ready to call him an every-week fantasy football starter. After all, over the last two weeks, Oakland's faced Baltimore and Cleveland, two teams who are starting to struggle against the pass.

Fortunately, however, Carr gets the Bears' this week in Chicago. While playing a road quarterback who's fairly inexperienced isn't a plus move, the fact that Vegas has the Raiders as favorites with a reasonably high team total against this 27th-ranked secondary (per NEP) makes me bullish about Carr's potential in Week 4. He's a good streamer -- perhaps the best streamer -- at quarterback this week.

Add Tyrod Taylor

After three weeks of play, Tyrod Taylor is a top-five fantasy quarterback. Is it a fluke? Maybe a bit. So far this season, Buffalo's faced Indianapolis, New England and Miami. According to our numbers, those three teams rank 20th, 22nd and 32nd, respectively, against the pass. Let's not pretend things have been incredibly difficult in terms of matchups.

So why add him and not sell him? Well, the Bills get the Giants and Titans over their next two games, and Taylor is still giving fantasy owners a relatively decent floor with his near-100 yards rushing so far on the year. And according to our Passing NEP metric, Taylor's been efficient through the air (ranks seventh on a per drop back basis), which means his numbers aren't being inflated because of the way fantasy football is scored -- he's actually been good.

If you're in need of a quarterback, Taylor's your guy. He's still out on more waiver wires than not, and should produce over the next two weeks before Buffalo faces Cincinnati in Week 6. And after that, the schedule opens up favorably again for him.

Sell Andy Dalton

I may look foolish telling you all to sell Andy Dalton after I was burned by him in Week 3, but logically, it makes sense to.

First off, we all have to remember that the quarterback position in fantasy football is almost completely replaceable. The number of quarterbacks who are for sure, honest every-week starters can be counted on one hand. Maybe even with three fingers. Therefore, if you can get a good deal for a player like Dalton, regardless of the way he's played through a three-week stretch, you make it happen.

But as I mentioned with Jeremy Hill above, the Bengals' schedule gets a lot tougher in the upcoming weeks. Thus far, they've faced the Raiders, Chargers and Ravens defenses. According to our metrics, they rank 31st, 19th and 16th in total schedule-adjusted defense thus far, and 31st, 13th and 19th against the pass.

Dalton could certainly end up as a top-10 or so quarterback this year given the way he's played and the weapons he has. And Week 4 isn't a bad matchup at all for him, so you may even want to wait to sell him. But being a top-10 quarterback doesn't really tell us much. There are always quarterback assets on the waiver wire to play, and it's not difficult to Frankenstein one together through streaming. If you can find a trade partner, especially in a shallower league with lots of passers in free agency, that'd be ideal.

Add Rishard Matthews

Here's a fun game: through three games this season, which Dolphins wide receiver is first in snaps played?

It's Jarvis Landry. Duh.

But do you know who's second? That would be Rishard Matthews, who's somehow, someway started the season by scoring near top-10 wide receiver PPR points.

It's tough to imagine things continue, but Matthews is second on the team with 23 targets (7.67 per game), clearly becoming Ryan Tannehill's second-favorite weapon. As it stands, that's not a terrible thing, because the Dolphins have run 80 percent of their plays (per Rich Hribar, a master of all things football) while trailing. Meaning, you know, they're passing a lot because the defense has been atrocious.

If things continue down this path, Matthews will have some PPR value. Don't expect the world from him, but he's worth a roster spot.

Add Antonio Gates

If you're one of those people who hate rostering "cheaters", maybe this isn't the move for you. Because, as you probably know, 35-year-old Antonio Gates was suspended for four games this year for using performance enhancing drugs.

If you want to be a little ahead of the game -- and, really, it was my mistake for not adding him to this list last week -- Gates comes back in Week 5. Last season, only Greg Olsen and Rob Gronkowski had more top-12, usable tight end performances in PPR leagues than Gates did. And given the way the Chargers' defense has played (19th in the league, per our metrics), there should be plenty of opportunity in the passing game. In fact, Ladarius Green, Gates' replacement, has 12 targets in two games played this year.

Add Ty Montgomery

Davante Adams reportedly re-aggravated his ankle injury last night, which means rookie Ty Montgomery will have a lot of opportunity in Green Bay.

Last night, Montgomery played starter snaps (66, the same as Randall Cobb), and was able to catch a couple of passes for 14 yards and a touchdown. It wasn't the most inspiring game in the world considering Aaron Rodgers' big night, but with more reps, perhaps that changes.

The reason you want to add Montgomery, in general, is simple: if Adams misses a decent chunk of time, Montgomery will have a few weeks playing with the best quarterback in the game. And, because of Adams' inconsistent and inefficient play in the past (his 0.58 Reception NEP per target was below average last season, which is frightening considering who's throwing him the ball), there's always the off chance Montgomery ends up taking his job.

Hold Justin Forsett

Regression was coming for Justin Forsett. For a player with such a high Rushing NEP per rush, his Success Rate was far too low. In other words, long runs and big plays were skewing his numbers a bit.

I'm not saying that's the sole reason for his slow start in 2015. There are a lot of things you can point towards -- the Ravens haven't gotten much push up front, and there's really no reason for defenses to try and stop a passing attack that has a 36-year-old Steve Smith and a plodding Crockett Gillmore leading the way.

Forsett's played poorly, and fantasy owners are frustrated. The silver lining here, though, is schedule-driven. The first three games of the season saw Denver, Oakland and Cincinnati. Denver and Cincinnati have two of the best defenses in the NFL, and game planning around Oakland involves targeting their secondary. And in PPR leagues against Oakland, Forsett still ranked as the 18th best running back in the league for the week.

The Ravens get the Steelers, Browns, 49ers, Cardinals, Chargers and Jaguars over their next three games. None of these matchups, aside from the Cardinals, are bad ones.

Add Joe Flacco

I just mentioned the Ravens' upcoming schedule, but really, you could even take this particular recommendation one week at a time.

Joe Flacco has been a middle-of-the-road QB2 in fantasy thus far, but he'll face a Pittsburgh secondary on Thursday night that hasn't been able to stop anyone since they opened up the offense in Week 7 of last year. Well, anyone outside of Nick freaking Foles.

Over their last 12 games (This was 11 before Foles' disaster in Week 3. I'm not bitter, I promise.), the Steelers have surrendered 10 top-12 quarterback performances in fantasy football. Per our numbers, the team ranks 23rd against the pass, and that's with Foles' mediocre game on Sunday.

The Ravens are favorites on Thursday night with a decent team total, too. You could do worse than stream Flacco.

Add Leonard Hankerson

I guess Leonard Hankerson is a thing now. Roddy White saw exactly zero targets in Week 3, and he's now been held catchless over his last two contests. Hankerson, meanwhile, has seen 17 targets over the last two weeks while hauling in 9 of them for 122 yards and a touchdown. Seems to be a pretty obvious change going on in Atlanta, no?

Hankerson played 67.5% of snaps in Week 1, 66.7% in Week 2 and Week 3 saw him on the field for 64.9% of the Falcons' snaps. Those aren't great numbers, which should give us a little hesitation moving forward, but it's clear that he's producing more than Roddy is.

Buy Dion Lewis

After recommending Dion Lewis as a "sell or hold" candidate in this very column last week, my thoughts shifted on Friday, and I analyzed the situation a little further. If you don't want to read the article, here are the cliff notes: (1) Lewis will, at the very least, have the Shane Vereen role in the Patriots' offense this year, (2) that Vereen role produced RB2 numbers last year, (3) Lewis still has the upside of being a better runner than Vereen, because Vereen wasn't a strong player on the ground.

The reason I'm mentioning Lewis in the column this week is because his owners may be turned off by the three touchdowns LeGarrette Blount scored on Sunday against Jacksonville.

It's true that Blount will see touches when there's an ultra-positive game script, as we saw against the Jags. But it's also true that Blount didn't touch the ball until the second quarter, and that Lewis started the game and saw action early. Lewis, too, ended up seeing 35 snaps versus Blount's 31, so that's a huge plus in a game with positive game flow.

I'm not assuming your league mate is an idiot and is ready to bail on Dion Lewis. Why should he? All Lewis has done is score fantasy points, and most owners who have him in the first place are the ones who believed in him from the beginning. But if you can go to the Lewis owner and talk up LeGarrette Blount -- which would then make Lewis look worse -- there's a possibility you could get a nice deal for a good (at the very, very least) RB2 in PPR leagues.