NFL

Fantasy Football: Carlos Hyde Is the Running Back to Draft in Kansas City

Hyde -- considerably cheaper than his backfield mate in early drafts -- may not offer much as a receiver, but don't be surprised if he out-rushes Damien Williams in 2019.

At this point, most fantasy players are well aware of the value of a running back on a winning team and in a high-scoring offense. Playing with leads can mean more volume on the ground, and an offense moving the ball well will generate plenty of scoring opportunities.

This isn't an article designed to highlight the value of running backs on winning teams, though. Nor is it an article to outline why the Kansas City Chiefs are likely to once again be a winning, high-scoring team. In 2018, their offense was the most efficient we've ever seen going by our Adjusted Net Expected Points (NEP) per play metric (which you can read more about in our glossary). They finished first among the 606 offenses in our database, which dates back to 2000, schedule-adjusted NEP per play. FanDuel Sportsbook also gives them the second-best odds (+800) of winning the 2019 Super Bowl. We can pretty safely operate under the assumption that KC is going to be good.

And so far fantasy football drafters are doing exactly that, buying high on the running back that closed 2018 atop the Chiefs' depth chart -- Damien Williams.

Per Draft.com's average draft position (ADP) data, Williams is going, on average, in the early third round, with an ADP of 29.2, making him the 16th running back off the board. In the month of May, he has gone as high as third overall in drafts, too, which is a higher "minimum pick" than you'll find for any other player going 18th or later.

That's a heck of a premium price to pay for a guy whose career-best single-season fantasy finish was as the RB48 last year in PPR formats (building on an RB73 finish in 2017). Will Williams be worth that price in 2019, or should you opt for Carlos Hyde at a much cheaper cost?

Let's dig into one of the most fantasy-friendly backfields in the league.

Williams' Late-Season Emergence

Granted, that overall rank of RB48 doesn't capture why Williams is going so high. Once he took over as the team's top back last year, he closed the regular season with 15-plus PPR points in three of his final four games, also notching 25-plus points in both his playoff games.

A per-game average of 24.4 fantasy points in that six-game stretch is some serious damage, and extrapolated to 16 games, that would have him pushing 400 points, which would have made him the clear-cut RB1 last year.

That's about as incredible a rate of production as you're going to find, but it was also only a six-game sample. It was also a sample that saw him score 10 touchdowns -- averaging 1.67 scores per game. That pace would give him 26.7 touchdowns over a 16-game stretch, which would blow out the highest marks we've seen from any running back in the last three seasons (21, 19 and 20).

Yes, part of the appeal of being in a good offense is the increased touchdown potential, but in today's NFL, you simply can't project any running back to come close to that kind of a touchdown rate.

But that's why he's going as the RB16 and not RB1, right? A little touchdown regression makes him lose some luster, but the top back in the KC offense is still going to be worth a ton -- or so the thinking goes.

What if he's not the top back, though?

Top Backs' Track Records

Volume is the absolute top priority when analyzing running backs for fantasy football, and in general, the market has been good at finding it.

Of the 13 backs who were drafted in the top 30 picks last year, five saw at least 300 opportunities (carries plus targets), another three saw at least 275, and only three saw fewer than 200. Those three included Le'Veon Bell, who did not play all year in a contract holdout, and Leonard Fournette and Dalvin Cook, who were hampered by injuries.

Over the last few years, though, we haven't exactly been going out on a limb in projecting these top-volume backs. Excluding rookies, there isn't a single back that has had a top-30 ADP in the last three seasons without having at least one previous season with at least 200 opportunities to their name.

That makes Williams a serious anomaly. He hasn't even racked up 200 opportunities in the last two seasons combined -- seeing 74 each in 2017 and 2018. His six-game sample as a starter did have him on pace to crack that mark, but there we're dealing with less than half a season of data, giving us a far less reliable sample than we've seen for any of these other top picks.

The Chiefs also have a running back who they are paying more than they are paying the recently-extended Williams in 2019, and it's a back who has twice eclipsed 200 opportunities in a season, with 250 in 2016 and 328 in 2017. We're talking about Carlos Hyde, of course.

Hyde bounced between two teams last season, but he still saw over 2.5 times as many opportunities on the year as Williams did, with 188 over 14 games.

Is Carlos Hyde Bad? Recency Bias and "Skill"

I'm not here to take shots at anyone, but one thing that got me interested in putting this article together was seeing Hyde's poor yards per carry from the last two seasons (3.7) cited as the reason that he isn't a threat to take work from Williams.

Yes, Williams crushed Hyde in that category last year, with 5.1 yards per carry. But that was a 50-carry sample, while Williams' first four seasons (133 carries) saw him average only 3.6 yards per attempt -- fewer than Hyde has even in his worst two-season stretch.

The 2018 Kansas City offense is by far the most efficiency-friendly environment either of these two have ever played in (and only Williams played in it), so it certainly makes sense that Williams' 2018 was the most efficient season we've seen from either of these two. And just because it's the most recent data doesn't mean we can completely throw away his early-career struggles or that we can crown him as definitely more skilled than Hyde, who has him beat in yards per carry in three of the previous four seasons.

Plus, yards per carry isn't a great efficiency metric anyway, so instead let's look at Rushing NEP per carry, which adjust for things like field position and down-and-distance.

Year Carlos Hyde Damien Williams
2018 -0.12 0.06
2017 -0.09 -0.19
2016 -0.02 0.01
2015 -0.03 -0.42
2014 -0.09 -0.17
Total -0.07 -0.10


Neither of these guys have been terribly inefficient (a negative Rushing NEP per carry isn't unusual, since rushing is inherently more inefficient than passing), but Hyde has a significant edge over Williams in this stat in their respective careers. Granted, the biggest gap between the two came last season, when Williams had a huge edge of 0.18. Again, though, let's not let recency bias erase the rest of their careers -- especially when that was a 50-carry sample that came while Williams was playing in what might be the best offense in NFL history.

To attempt to capture the influence of their surrounding situations a bit more, we can also compare their numbers to how their teammates have fared in the same seasons, helping us adjust some for offensive line strength, the effect of the passing game quality and so on.

Looking at Hyde's numbers, things seem kind of bleak.

Year Team Rush NEP
per Carry
Teammates
RNEP/C
Difference Success Rate Teammates
Success %
Difference
2018 JAX -0.25 -0.10 -0.14 32.14% 36.57% -4.42%
2018 CLE -0.06 0.06 -0.13 33.33% 41.81% -8.48%
2017 SF -0.09 -0.01 -0.07 35.83% 42.37% -6.54%
2016 SF -0.02 -0.22 +0.20 40.09% 34.85% +5.24%
2015 SF -0.03 0.09 -0.12 42.61% 31.12% +11.49%
2014 SF -0.09 -0.08 -0.01 39.76% 38.15% +1.61%
Total
-0.07 -0.04 -0.03 37.70% 37.42% +0.28%


If you're looking for a reason to confirm why Carlos Hyde isn't good, there's one for you. He hasn't lived up to the average Rushing NEP per carry set by his backfield stablemates over his career. It is worth noting, though, that he holds a very slight edge over his teammates in Success Rate, which measures the percentage of a back's carries on which positive NEP is generated.

If that edge is confirmation that Hyde is not good, though, then you should also be pumping the breaks on Williams:

Year Team Rush NEP
per Carry
Teammates
RNEP/C
Difference Success Rate Teammates
Success %
Difference
2018 KC 0.06 0.04 +0.01 54.00% 44.35% +9.65%
2017 MIA -0.19 -0.09 -0.10 23.91% 36.92% -13.00%
2016 MIA 0.01 0.05 -0.05 34.29% 41.49% -7.20%
2015 MIA -0.42 -0.02 -0.40 25.00% 38.32% -13.32%
2014 MIA -0.17 0.03 -0.20 38.89% 48.83% -9.94%
Total
-0.10 0.00 -0.11 37.16% 42.02% -4.87%


Over his career Williams' production relative to his teammates' production has been ever worse than Hyde's, finishing 0.08 Rushing NEP per carry further behind than Hyde did, while also posting a significantly worse Success Rate than his teammates.

Interesting to note, certainly, is that even in that elite KC offense in 2018, Williams did actually outperform his teammates. This was the only time we've seen him do that in his career though, and it was over a 50-carry sample.

Skill is a pretty nebulous concept, and frankly, I don't think it's worth much in fantasy football. If it's captured by usage and production, then we can lean on usage and production stats, and if it's not reflected in those, then it doesn't mean much of anything for fantasy output.

But even if you are leaning on skill as a reason that Williams will be the lead back over Hyde, it takes a little bit of mental gymnastics to justify these numbers. Can skill appear and disappear out of nowhere? Because while Williams certainly out-played Hyde in 2018, even Hyde's down year in 2017 was better than Williams' 2017 campaign. Can it show through in only 50 carries? Because Williams preceded his 2018 breakout with a 133-carry sample of being inferior to Hyde and his past teammates.

My Concession: They Can Co-Exist

What Williams does have going for him -- without question -- is a big edge in the passing game.

Player Targets Receptions Reception NEP Rec. NEP
per Target
Rec. NEP
per Reception
Williams 139 109 58.71 0.42 0.54
Hyde 169 119 17.73 0.10 0.15


Hyde hasn't been nearly on Williams' level in the receiving game over their NFL careers. This is relevant not only because targets are worth a fair bit more than carries in fantasy football, but because passing-game usage will also play a part in dictating snaps.

Hyde has seen only 39 career opportunities on third down with more than five yards to go. Williams has seen 28, accounting for 8.7% of his career total, compared to 3.9% for Hyde.

In the days of the I-Formation, this wouldn't matter, we'd laugh at Williams for a lack of toughness, pat our fullbacks on the back and draft Hyde in the top-10. We're not living in 2004, though, and even even in the midst of this pro-Hyde piece, I will gladly concede that Williams is going to be fantasy-relevant as the better pass-game option in what should be an elite passing attack.

Cost Considerations

Co-existence is another box checked in Hyde's favor, though, because you're not simply being asked to bet on which one is going to produce more fantasy points, you're also forced to decide what kind of draft capital you want to spend to make that bet.

A top-30 pick is the kind of capital that's typically reserved for backs who have a track record of big workload and a ceiling for true top-end volume. A pick in the 90s, where Hyde's 96.4 ADP puts him (on average the 40th running back off the board) reflects nothing of the sort.

For Williams to pay off this ADP, he needs to do more than just contribute in the passing game with some additional rushing value on the side; he needs to be a true workhorse. For Hyde, simply getting the lion's share of the rushing volume can leave him offering some interesting value, especially with the touchdown upside that comes with being in KC.

Their current draft prices don't leave much room for uncertainty in their workloads, and with Hyde's track record as the superior rusher, it's not wise to completely write off his potential -- especially this early in the offseason -- to command a significant share of the team's carries.

The opportunity cost in being wrong on Williams is very high and can be potentially crippling to your team. The cost of being wrong on Hyde in the seventh or eighth round is much easier to stomach, while he also has the potential to return some serious upside on his draft spot.