True Fantasy Points: 2018 Wide Receiver Projections - Footballguys

A stats-based guide for separating skill from luck in wide receiver projections.

This is the third installment in the True Fantasy Points (TFP) series for 2018. To read the article on quarterbacks, click here. To read the article on running backs, click here. And if you want to learn more about the statistical methodology underlying the series, click here or here.

Today's article will identify two overvalued wide receivers and two undervalued wide receivers. One of this latter pair is the most undervalued player in fantasy drafts according to the combination of his TFP and Average Draft Position.

Below is a table showing David Dodds' projections, TFP, and their difference for 59 wide receivers who meet the following criteria:

  • They're on the same team as they were last season.
  • They're projected for at least 300 "opportunities," which is defined as rushing attempts plus routes run.
Dodds' Stats True Stats Points
Wide Receiver Team ReYds ReTDs ReYds ReTDs Dodds TFP Diff
Stefon Diggs MIN 931 7 864 5 137.0 119.8 +17.2
Tyreek Hill KC 1015 7 918 6 156.0 139.1 +16.9
Antonio Brown PIT 1350 10 1329 8 195.0 179.2 +15.8
DeAndre Hopkins HOU 1321 11 1337 8 195.1 182.3 +12.8
Julio Jones ATL 1336 7 1273 6 174.6 164.2 +10.4
Julian Edelman NE 655 5 625 3 94.0 84.4 +9.6
Marquise Goodwin SF 961 6 969 4 134.5 125.3 +9.2
Adam Thielen MIN 1098 6 1079 5 147.4 138.5 +8.9
T.Y. Hilton IND 1140 8 1155 6 159.6 150.8 +8.8
Mike Williams LAC 549 4 519 3 78.9 70.9 +8.0
JuJu Smith-Schuster PIT 911 7 900 6 132.4 124.6 +7.8
Golden Tate DET 862 6 883 4 122.1 116.0 +6.1
Ted Ginn Jr NO 562 4 552 3 81.8 76.1 +5.7
Corey Davis TEN 756 5 773 4 107.4 101.8 +5.6
Kelvin Benjamin BUF 759 5 739 4 105.9 100.3 +5.6
Cooper Kupp LAR 820 5 796 5 112.6 107.2 +5.4
Keenan Allen LAC 1218 8 1222 7 168.2 163.2 +5.0
Chris Godwin TB 595 3 578 3 79.3 74.5 +4.8
Michael Thomas NO 1125 7 1116 6 154.9 150.1 +4.8
Dede Westbrook JAX 611 5 622 4 88.1 83.6 +4.5
Tyrell Williams LAC 684 5 685 4 97.8 94.5 +3.3
Jamison Crowder WAS 854 5 837 5 118.3 115.0 +3.3
DeVante Parker MIA 647 4 673 3 91.1 87.9 +3.2
Davante Adams GB 1021 8 990 8 151.9 149.1 +2.8
DeSean Jackson TB 630 4 612 4 86.7 84.0 +2.7
Pierre Garcon SF 763 4 777 4 100.9 99.0 +1.9
Robby Anderson NYJ 832 6 848 5 118.4 116.7 +1.7
Tyler Lockett SEA 617 5 664 4 96.4 95.9 +0.5
Kenny Golladay DET 745 5 755 5 104.5 104.2 +0.3
Amari Cooper OAK 980 6 993 6 136.2 135.9 +0.3
Josh Gordon CLE 868 6 922 5 120.4 120.3 +0.1
Jermaine Kearse NYJ 559 4 567 4 78.1 78.2 -0.1
Robert Woods LAR 797 4 784 5 106.4 107.1 -0.7
Quincy Enunwa NYJ 560 3 589 3 76.4 77.3 -0.9
John Ross CIN 625 5 669 4 91.5 93.0 -1.5
Sterling Shepard NYG 690 5 710 5 99.8 101.9 -2.1
A.J. Green CIN 1131 7 1134 7 155.7 158.3 -2.6
Mohamed Sanu ATL 627 4 646 4 88.0 90.8 -2.8
Geronimo Allison GB 452 4 528 3 68.6 71.8 -3.2
Will Fuller V HOU 724 5 732 6 103.8 107.3 -3.5
Alshon Jeffery PHI 751 6 751 6 108.1 111.8 -3.7
Keelan Cole JAX 840 6 910 5 117.0 120.9 -3.9
Taywan Taylor TEN 585 4 619 4 81.9 86.1 -4.2
Nelson Agholor PHI 690 5 712 5 100.4 104.6 -4.2
Doug Baldwin SEA 975 7 1002 7 137.8 143.4 -5.6
Emmanuel Sanders DEN 806 5 877 5 110.6 116.2 -5.6
Mike Evans TB 1112 8 1166 8 156.8 162.6 -5.8
Rishard Matthews TEN 671 4 702 5 92.0 98.7 -6.7
Odell Beckham Jr Jr NYG 1256 9 1283 10 179.6 187.7 -8.1
Terrance Williams DAL 445 4 554 3 67.0 76.7 -9.7
Devin Funchess CAR 824 6 864 7 119.0 128.7 -9.7
Zay Jones BUF 440 3 512 4 62.0 72.8 -10.8
Josh Doctson WAS 500 4 565 4 72.8 83.8 -11.0
Randall Cobb GB 632 5 741 5 96.7 108.5 -11.8
Chris Hogan NE 777 6 889 6 116.1 128.7 -12.6
Marvin Jones Jr DET 930 6 1043 7 130.8 143.9 -13.1
Kenny Stills MIA 827 6 894 7 117.5 132.4 -14.9
Demaryius Thomas DEN 916 6 1048 7 127.0 144.8 -17.8
Larry Fitzgerald ARI 959 6 1098 7 134.3 154.0 -19.7

As an example of how to read this table, focus on the row associated with Kenny Golladay. Dodds projects 755 receiving yards and 5 touchdowns, whereas the regression-to-the-mean-based TFP system projects 745 total yards and 5 touchdowns. One probably doesn't need to do the math given how close these projections are to the naked eye. But for the sake of exposition, applying Footballguys' standard scoring system results in a Dodds projection of 104.5 points and a TFP projection of 104.2 points. Based on that tiny 0.3-point difference, readers can conclude:

  1. Dodds has optimally accounted for regression to the mean in wide receiver stats.
  2. Golladay is valued properly in the context of fantasy football drafts.

With everyone on the same page, we can proceed to a discussion of overvalued and undervalued wide receivers. Along the way, (arguably) the most positive expected value selection in 2018 fantasy drafts will be revealed.

Overvalued: Stefon Diggs

Dodds' projection of 70 receptions implies Diggs will run 458 routes this season given his 15.3% True Receptions per Route Run with the Vikings. And if that's the case, then Dodds' projection of 931 receiving yards implies 2.03 Yards per Route Run (YPRR) Here's the problem with that. Diggs' True YPRR is only 1.89 across a large sample size (1,317 routes run). Furthermore, to put 2.03 YPRR in perspective, this figure is reserved for only the best of the best at wide receiver. In 2017, it would have ranked ninth among wideouts targeted 100 or more times, with the Top 8 being a veritable who's who at the position: Julio Jones, Antonio Brown, Keenan Allen, DeAndre Hopkins, Michael Thomas, Tyreek Hill, A.J. Green, and Diggs' own teammate, Adam Thielen. Furthermore, in terms of "true" skill, only 10 wide receivers in the above table have a True YPRR of 2.03 or greater: Jones, Brown, Green, Hill, Theilen, Allen, Odell Beckham Jr Jr, Demaryius Thomas, Josh Gordon, and Mike Evans.

In short, Dodds is asking Diggs to enter a level of elite yardage efficiency that three seasons-worth of evidence -- plus the math of regression to the mean -- doesn't deem likely.

Overvalued: Tyreek Hill

The situation is similar with respect to Hill's TFP. Given his 16.9% True RPRR, Dodds' projection of 72 receptions implies 427 routes run this upcoming season. However, if that ends up being correct, then Dodds' projection of 1,015 receiving yards implies 2.38 YPRR, which is astronomical even for Hill, who was just listed as having a True YPRR already in the stratosphere (2.15).

The only difference with Diggs is that Hill's Actual YPRR in 2017 was 2.36, so at least we already have evidence that he can produce yardage efficiency commensurate with what Dodds' projects for 2018. That said, it remains the case that producing 2.38 YPRR over 427 routes this season would be no small feat and would put Hill in overall WR1-WR3 territory, which the TFP math says is unlikely.

Undervalued: Kenny Stills

Stills has been with the Dolphins for three seasons, running 1,506 routes in that time. Across this large sample size, he's produced a below-average 1.50 True YPRR but an above-average 1.20% True TDPRR. And yet, if we do the same kinds of calculations as above, Dodds' projection implies a sudden efficiency nosedive for Stills in 2018: 1.38 YPRR and 0.96% TDPRR.

One would think that such a decrease would be concomitant with losing a starting quarterback, not getting him back. A rebuttal might be that Stills has never seemed to develop a "special relationship" with Ryan Tannehill, so he's actually better off without him. However, that can't be the case because, remember, Stills' "true" stats cake has two seasons' worth of routes with Tannehill -- nearly 800 in total -- already baked into it.

Regardless of where you come down on this argument, Stills' current Average Draft Position (ADP) of 122 is downright criminal. Even selecting him in the mid-90s, which would appear to your league-mates to be a colossal reach, would still -- pun intended -- represent colossal value.

Undervalued: Chris Hogan

The reason for Hogan's difference in the table is identical to Stills' with respect to receiving yardage. Hogan's True YPRR in New England is a Stills-esque 1.52, and Dodds' projection implies a Stills-esque drop to 1.33. It's true that Hogan's sample size of 823 routes run is smaller than Stills', so one can expect more regression to the mean (i.e., more variability in what his Actual YPRR might be in 2018). The thing is that, like Stills, Hogan's True YPRR is actually below the 1.68 league average, so a lower sample size suggests it's even more likely to increase towards 1.68; not drop all the way down to 1.33.