Wide Receivers Who Should Have Scored More: Positive Touchdown Regression Candidates
June 5, 2018 | By
Michael Dubner
Due to the
variance in wide receiver touchdowns year-to-year and other stats only weakly predicting next-season receiving TDs, regression can be used to identify wide receivers whose 2017 receiving touchdowns didn’t align with their expected touchdown total.
WR Touchdown Regression Equation
You can see
my methodology behind this equation here. But the jist of it is:
Expected reTDS = 0.139 + (-0.015 * Total reRTGS) + (0.008 * Total reYDS)
- For positive touchdown regression: look for WRs with a lot of yards, but few touchdowns.
- For negative touchdown regression: look for WRs with more touchdowns than their total receiving yards would predict, even if they have a lot of targets.
* Note: a WR underperforming his receiving TDs expectation in 2017 doesn’t mean he’s going to overcompensate with even more touchdowns in 2018 to correct that deficit. Rather, he should score at a rate
1 closer to what his opportunity suggests.
2
Positive Touchdown Regression Candidates
1. Julio Jones: Actual: 3; Expected: 9.5; Difference: -6.5
Jones’ -6.5 TD difference is the third largest underperformance over the past decade, behind only
Calvin Johnson‘s -7.9 and
Andre Johnson‘s -6.5 2012 seasons. Scoring just three TDs on 1,444 receiving yards is an even greater outlier considering he saw 19 red-zone targets (eighth among WRs). Additionally, the WR TD expectation equation suggests that when given an equal number of targets, a WR with a higher yards per target (YPT) is expected to score more TDs — Jones ranked eighth in YPT (9.
. If Jones and
Matt Ryan (whose 2017 TD rate of 3.8 percent was well below his career average of 4.6 percent) both positively regress in the touchdown department, Jones could be a massive value at the Rounds 1-2 turn.
2.
Michael Thomas: Actual: 5; Expected 7.9; Difference: -2.9
Thomas’ underachieving touchdown total is highly correlated with
Drew Brees’ low touchdown total. Brees threw his fewest touchdown passes (24) over the last 15 seasons, while still leading the league in yards per attempt (8.1) and completion percentage (an NFL record 72 percent). Even if the Saints continue to have a strong running game, their 23 rushing TDs should negatively regress, as they had five more rushing scores than any other team. And even if the Saints defense remains strong, we shouldn’t expect them to play with a lead during
55.5 percent of game time (third) or a multi-score lead during
28.5 percent of their offensive snaps. More negative or neutral game scripts would lead to an increase in pass attempts, as the Saints’ pass-to-run ratio dipped from 59 percent in neutral game scripts to 42 percent when playing with a multi-score lead.
3. Adam Thielen: Actual: 4; Expected: 8.2; Difference: -4.24
Despite nearly identical targets and yards, Thielen scored just one TD before the Vikings’ Week 9 bye, but found the end zone three times in the second half of the season. While
Case Keenum played well, new QB
Kirk Cousins is still an improvement. Expect Thielen to score at a rate closer to the second half of 2017.
4. Pierre Garcon: Actual: 0; Expected: 3.1; Difference -3.1 and
Marquise Goodwin: Actual: 2; Expected: 6.3; Difference: -4.3
I’m not projecting a 32-year old Garcon recovering from a neck injury with a career high six receiving TDs or a 5-foot-9-inch, 183-pound Goodwin with eight receiving TDs in five NFL seasons to suddenly catch double-digit scores. But the 49ers threw just 15 passing TDs in 2017, and they look to be an ascending offense captained by
Jimmy Garoppolo.
5. DeVante Parker:
ActuaL 1; Expected: 4.1; Difference: -3.1
Parker hasn’t lived up to his first-round draft pedigree, but should see positive touchdown regression
and increased opportunity with
Jarvis Landry‘s 161 targets (27 percent target share) up for grabs.
Bonus: Status Quo Candidate
1. Antonio Brown: Actual: 9; Expected 10; Difference -1
Brown had the highest receiving touchdown expectation (10) in 2017, and of the 14 WRs to score eight or more receiving touchdowns, Brown is the only one that actually underperformed his expected touchdown total. Therefore, Brown (the No. 1 WR in points per game)
should have actually scored more points. Since 2013, Brown has averaged 10.4 TDs per year, and a ridiculous 10.9 expected TDs per year.
Conclusion
Regression isn’t the end all be all, as we must account for opportunity changes, team situation, and Average Draft Position (ADP). But regression is a quick way to identify WRs who over- or underperformed in the touchdown department. Stay tuned for a look at which players overperformed in 2017, and will likely see their touchdown totals decrease in 2018.