Forecasting 2018 NFL Rookie RB Success: 3-Year Model

Forecasting 2018 NFL Rookie RB Success: 3-Year Model

By Kevin Zatloukal (4for4 Contributor), last update Feb 4, 2019

Kevin Zatloukal's picture

Kevin is a Ph.D. computer scientist. His doctoral work at MIT was focused on quantum algorithms. During the fantasy offseason, he teaches computer science at the University of Washington, his alma mater.

Follow Kevin Zatloukal on Twitter: @kczat.

My last article looked at the new class of rookie wide receivers trying to break into the league in 2018. Here, the spotlight shifts to the newest class of running backs. While none of the rookie receivers have a strong chance of becoming fantasy relevant this year, several rookie running backs appear poised to do so. We will discuss a few of them below.

More Rookie ForecastsLB | WR

As with receivers, we will also try to look a bit further into the future and attempt to determine which rookies are likely to be fantasy starters at some point over the next three years, even if not right away. That information is especially important to those readers in dynasty or keeper leagues, who can stash rookie players on their deep rosters for multiple seasons.

By the way, if you’re in a dynasty league, you should definitely check out all the content from our friends at Dynasty League Football. They have much more information on all of the rookies I talk about below.

Forecasting Running Back Success

Two years ago, I built a statistical model that predicted the odds of a rookie running back having a top-24 season in one of their first three years in the NFL. That model uses the pick at which the player was taken in the NFL Draft, along with their age, 3-cone time, market share of team touchdowns, and number of receiving touchdowns. Last year, I added a machine learning (ML) model as well. Combining different models with similar individual accuracy typically leads to more accurate forecasts due to a wisdom-of-crowds effect. The combined odds I give below are a weighted average of these two different models.

Without question, the top prospect in this year's class is Saquon Barkley. Drafted second overall, with stellar college production and athleticism, he crushes both of the models. Unfortunately, unless you have the 1.01 pick in a rookie draft or are willing to trade your first born to get it, you won't be able to draft Barkley.

Below, I'll focus my attention on running backs who can be acquired later in rookie drafts. In particular, those who can be drafted at a discount to their true value.

Rashaad Penny, Seahawks

Let me start off by saying that, for me, Rashaad Penny is clearly the 1.02. The only way Pete Carroll could be any clearer that the Seahawks want to run the ball more next year would be to get it tattooed on his face. In every interview, he's made clear that fixing the running game has been the main priority all offseason. They fired the entire coaching staff under Carroll from last year and have made moves in free agency to improve the offensive line. Then, they used their first-round pick (which they usually trade away) to draft Penny. Carroll has since said they want Penny to be a three-down back. This is as close as we're going to get to a giant neon sign flashing "opportunity".

The consensus 1.02 rookie pick still seems to be Darius Guice. Guice was the consensus 1.02 pick before the draft, and most drafters do not seem to have properly adjusted their views to the fact that he was the seventh running back taken in the draft, falling to the end of the second round. Six RB-needy teams passed on Guice. He also falls into a worse situation than Penny since Chris Thompson has the third-down role locked down.

The combined model still sees Guice, even with his lower-than-expected draft pick, as a Tier 1 running back, but he's at the bottom of the first tier.

Continue reading for five more running backs and complete forecast of all rookie backs...

4for4 Premium Content

There's more, but it's available only for 4for4 subscribers.

Subscribe Now - Get Premium 4for4 Content

If you are a current subscriber, please login here

Filed Under: