Dressed to Regress: D’Andre Swift (2021 Fantasy Football)
Regression. It’s a basic statistical concept that most of us understand. If a value is unusually large or small, odds are that the following values will eventually move (regress) towards the average (the mean). It’s also an important concept when evaluating players for fantasy football, particularly in relation to touchdowns.
Like bumping into Andy Reid at a salad bar, a touchdown is a random, unpredictable event. Let’s say it’s 3rd and 10 on the opponent’s 10 yard line and the QB has two receivers open in the end zone. In a split second, he has to make a decision on who to throw it to. If that play gets run ten times, maybe each receiver would be targeted four times, with the QB tucking and running on the other two. This particular Sunday, though, those six points are only going to one of them.
In sports like baseball or basketball, where there are more games, things tend to even out over the course of a season and we can put a little more weight on a player’s full-season stats. In a 16 or 17 game NFL season, though, a few quick decisions among the chaos of the football field can skew a player’s points total and disproportionately impact their position rank at the end of the season. Yet, while touchdowns totals aren’t typically predictive from year to year, the biggest factor in where players go in fantasy drafts often comes down to how frequently they found the end zone the year prior.
On the flip side, touchdown regression is more like Thanos (or me making lame, pop-culture references): it’s inevitable. Sure, some players tend to naturally get fed more of those precious end-zone targets, but regression catches up with everyone eventually. So by isolating touchdowns as a component of a player’s overall fantasy production, and factoring this into our preseason process, we can identify which players are fantasy gold at their ADP and which players are just shiny lumps of pyrite.
In this series of articles, we’ll use some of the excellent (and free) reports that FantasyPros offers to identify players you may want to flag on your cheat sheet come draft day—both those to target and those to avoid. We’ll analyze data from the Touchdown Regression, Stats Leaders, Red Zone Stats, and Advanced Stats reports—along with situational factors that might also influence a player’s production—to determine which players could be in line for positive or negative regression in 2021. We’ll then compare that to the latest ADP to help you mine greater value in your fantasy drafts.
Regression also has many flavors (more like Andy Reid at an ice cream bar), and we’d be doing ourselves a disservice if we limited ourselves to just one. So, while touchdown regression will be the double scoop that forms the foundation of this article series, we’ll also be sprinkling on attempts, targets, and receptions to make sure we’re building a multi-layered regression sundae that’ll have your league-mates eyeing you with jealousy come playoff time.
So, in the words of Andy Reid at a Hawaiian shirt sale, settle in and get comfortable. We have a lot of material to go through. (Editor’s note: Andy, I can only apologize…)
D’Andre Swift: ECR 31, RB16
Prior to last season’s NFL draft, D’Andre Swift was pegged by many analysts as the best running back available. He got the kind of draft capital you hope for, being drafted as the second RB off the board at the top of the second round. However, landing with the Detroit Lions put him in a tough spot. He’d be playing in an anemic Matt Patricia offense and competing for carries with Kerryon Johnson and Adrian Peterson.
Still, Swift showed plenty of glimpses of being a valuable fantasy asset, scoring 10 touchdowns and getting most of the work in the passing game. And with Peterson and Johnson moving on, that leaves a large number of vacated rushing attempts for Swift to lay claim to. A deeper look at some of the stats that led to his RB18 finish in Half-PPR, though, shows cause for concern.
One of the biggest factors that led to Swift’s RB18 finish was, you guessed it, touchdowns. In 2020, 35.97% of Swift’s fantasy points came from TDs. Of running backs who had at least 100 rushing attempts, only three had a higher percentage of points from touchdowns: Alvin Kamara, Nick Chubb, and Jeff Wilson Jr. More concerning is how Swift got those TD points.
Analyzing some red zone data shows us that among running backs who had at least 10 rushing attempts inside the red zone, Swift led the league with a sky-high 40% TD rate, with eight touchdowns on just 20 attempts. Perhaps these stats mean that Swift is a generational talent when it comes to sniffing the end zone. More likely, though, his red-zone TD rate is also in line for regression in 2021.
A quick look at the Touchdown Regression report bears this out. We can see that by moving his 2020 touchdown rate 50% of the way towards the league average, Swift would have dropped three spots, from RB18 to RB21, in both half and full PPR. (As an aside, when I run this report, I like the Based On setting to be Areas of Field. This takes the location of a player’s touches into account.) Three spots may not seem significant, but a couple of other key situational considerations hint that Swift could be in line for deeper regression.
First, Swift faces arguably higher-quality competition this year, with the fresher legs of free agent Jamaal Williams replacing Peterson as the primary competition in the running back room. As well as touchdowns, the other lynchpin of Swift’s game last year was his work in the passing game. He received 57 targets (13th among all running backs) compared to Peterson’s 18. Williams should provide much stiffer competition as a receiving option, which is likely to lower Swift’s ceiling.
Arguably the bigger problem, though, is the overall strength of the Lions’ offense. While we don’t know what Dan Campbell’s offense will look like, it’s hard to look at this roster and argue that they’ll be a stronger unit in 2021. The biggest and most obvious change will be the downgrade at QB from Matt Stafford to Jared Goff. However, the Lions also lost their WR1, WR2, and WR3 to free agency, with Kenny Golladay heading to New York, Marvin Jones going to Jacksonville, and Danny Amendola leaving to follow his dream of becoming a hand model or something (this may not be true). Given the lack of legitimate threats at wide receiver, opposing defenses should be able to focus their attention on stopping the Lions’ rushing game.
That said, there are 208 vacated rushing attempts up for grabs this year, so a solid case can be made that Swift’s share of the rushing attempts should see some positive regression. I’m projecting the Lions to be playing in more negative game scripts, though, which should decrease their overall number of rushing attempts. It’s also hard to see the Lions consistently moving the ball downfield this season, which bodes poorly for Swift given that such a high percentage of his points in 2020 came in the red zone.
Conclusion: Regression is Coming
Looking at the data, we would expect D’Andre Swift to be in line for some touchdown regression in 2021 even if his opportunities remained exactly the same. In all likelihood, though, the signs point to the quality of his opportunities diminishing, and possibly some of the volume, too. Despite finishing as the RB18 last year, his ADP on Underdog Fantasy at the time of writing is the RB16, meaning that drafters are projecting Swift’s production to increase this season (at this time of year, paid best ball drafts are usually the best indicator of how players are being valued). When we factor in the probability of regression, and bake in situational factors, it’s hard to see Swift returning value at his current cost.
While I like the talent of the player, I don’t plan on rostering Swift in 2021.
Whether you’re new to fantasy football or a seasoned pro, our Fantasy Football 101: Strategy Tips & Advice page is for you. You can get started with Starting Your Own Fantasy Football League or head to a more advanced strategy – like What is the Right Amount of Risk to Absorb on Draft Day? – to learn more.