The 2018 NFL draft was historic for quarterbacks. It was the first draft in 69 years that four quarterbacks came off the board in the first 10 picks, and it was just the third time that five or more were taken in the first round. That kind of expenditure of draft capital naturally leads to high expectations. Typically it takes three years before we can judge the success of a draft pick, so with over a third of the 2020 season completed, we wanted to ask: Is it still too early to say which 2018 QB is the best?
Picking the worst is easy. Joshua Ballinger Lippincott Rosen, who was taken with the 10th overall pick, has already been traded and waived in his brief career and is now on the practice squad in Tampa Bay. Picking the best is more difficult. Among the four first-round QBs who are currently starting, three have made a case for consideration through Week 6 of this season — if only temporarily.
- In 2018, Baker Mayfield took over starting duties in Cleveland from Tyrod Taylor after Taylor left a Week 3 game against the New York Jets with a concussion. (A young, first-round QB taking over for Taylor and then never leaving has become a familiar story for the beleaguered Chargers quarterback.) Mayfield went on to set the rookie touchdown passing record en route to a 6-7 record as a starter.
- 2019 was Lamar Jackson’s year. He captivated the NFL with his athleticism, rushing for a single-season quarterback record and earning league MVP honors while leading the Baltimore Ravens to a 13-2 record in games he started.
- This season began with Josh Allen playing near-elite football. And while he’s cooled off since the Buffalo Bills’ 4-0 start to the season, losing two straight games heading into Week 7, Allen has surpassed Mayfield in Total Quarterback Rating (QBR) over their careers. Allen now places second among the four starting QBs from the ’18 draft class by that measure, behind only Jackson.
- Despite being taken with the third overall pick, Sam Darnold of the New York Jets is the only QB in the group to have not enjoyed a moment — however brief — in which he was performing well enough to be considered the top quarterback in the draft class. Darnold currently sits last in career QBR among the four 2018 draftees, with a rating of just 46.2 through Week 6.
We can also track the evolution of each quarterback’s performance over his career so far using FiveThirtyEight’s QB Elo ratings.1
Mayfield’s peak came early on, and he hasn’t matched that level of performance since, settling in as a slightly below-average NFL QB in recent weeks. Meanwhile, Jackson’s ascent was rapid and sustained, with 70 percent of his games earning an above-average Elo value. Allen’s plot is emblematic of the Josh Allen Experience: a roller-coaster ride of ups and downs, with a general upward trend. Finally, Darnold has had his moments, but his peaks aren’t especially high, while his lows are among the lowest. Just 30 percent of Darnold’s games have been above average.
Even with the fluctuations in performance, it’s safe to say that the conventional wisdom was wrong about these quarterbacks. The No. 1 overall pick ranks third among his draft peers in career QBR, and the last pick of the first round ranks first. Not even Baltimore was completely sold on Jackson — the Ravens took tight end Hayden Hurst with the 25th pick before trading up to select their QB with the last pick of Round 1. Pointing and laughing surely isn’t useful, but examining the reasons that coaches and scouts missed the way they did can be instructive.
If we’re going to talk about bad football opinions, it’s hard not to start with people’s thoughts on Jackson. Perhaps the coldest take of the 2018 draft process was the notion that Jackson wasn’t big enough or accurate enough to be an NFL QB. That sentiment was expressed most famously by Bill Polian, former general manager of the Indianapolis Colts, when he said Jackson should become a wide receiver. But Polian wasn’t alone. The evaluation was shared by at least one other coach. “He will not be able to play [quarterback] in this league, mark my words. When he throws, he hopes,” an anonymous offensive coordinator told NFL.com’s Tom Pellissero. Perhaps it’s piling on, but it’s worth mentioning that Jackson won his MVP at quarterback.
But NFL coaches’ misreads didn’t end with Jackson. “If you watch them both, I think it’s obvious how much better an athlete Darnold is than Josh Allen,” one QB coach told Pelissero. “I’m not talking about [forty] times or anything. I’m just talking about his ability to move quickly, to adjust.”
If Darnold actually is more athletic than Allen, he hasn’t shown it yet. So far in his career, Allen is averaging 37.8 rushing yards per game on scrambles and designed runs, nearly four times as many as Darnold’s 10.6 yards per game, through Week 6 of 2020. And their career sack rates are nearly identical, at 7 percent for Darnold and 7.1 percent for Allen.
Teams take great pains to try to quantify prospects’ general athleticism each year at the combine in Indianapolis. That Jackson’s obvious athleticism was held against him as a QB prospect and Allen’s relative athleticism compared to Darnold was missed entirely suggests that even the most straightforward evaluations of QB play are fraught with error.
But the most common fault scouts and coaches found with these QBs was their accuracy. One coordinator summed up the feeling of many in the draft analytics community when he pointed out that you can’t teach accuracy, and this truism was especially damning for Allen’s pro prospects. “You’re not going to fix that in 10 weeks or two years,” the coordinator said of Allen. “If you’re not accurate, you’re not accurate. There’s an element of processing and anticipation that I think he lacks. But I think he’s got a big arm and somebody’s going to over-draft the guy because of that.”
Every bit of evidence-based analysis we have about pro football supports that notion. Yet this year, we have data indicating that Allen may actually be getting more accurate. His early-career returns were terrible, and he sometimes went stretches where more than 1 in 4 of his passes were off-target, but he’s recently matched the lower off-target shares of his peers.
It’s a remarkable development — and quite unexpected. The number of bad balls a QB throws is one of the more stable and predictive performance measures we have for the position, and at least so far this season, Allen has made a mockery of the idea that players are incapable of substantially improving this skill once they’re in the NFL.
Jackson was also thought to be particularly susceptible to inaccurate throws over the middle of the field because of mechanics one analyst described as a “narrow base [that] forces his elbow to drop.” Yet the middle of the field has turned out to be a strength of the Ravens’ passing attack, with tight end Mark Andrews and wide receiver Marquise Brown both averaging over six yards per route run since 2018 on deep inside patterns. When you look at the distribution of his throws, Jackson’s proficiency over the middle is obvious.
Between the hashes, Jackson’s off-target heat map shows almost no pink from 0 to 25 yards downfield, indicating that his throws to that area are on time and on target. Like the other QBs, he’s been inaccurate to the right sideline at intermediate depths, but his accurate throws — represented by the teal areas of the heat map — bulge out into the middle of the field deeper than those of his peers.
So what explains all these failures of projection? How could so many people — and the numbers — be so wrong in so many important ways? And why is it still so difficult to say which QB is the best from the 2018 class? One possible answer is team fit.
“Give [Baltimore offensive coordinator] Greg Roman a ton of credit for making [Jackson] more accurate and putting him into plays where he can succeed,” an anonymous scout recently told USA Today.
An example of what the scout may be referring to, and an obvious explanation for Jackson’s sudden ability to accurately throw over the middle, is Baltimore’s use of the play-action. Play-action passes are especially effective when targeted 10 to 30 yards downfield, and in the three years that Jackson has been in Baltimore, the Ravens have run play-action passes on at least 28 percent of dropbacks each season, ranking fifth in the league in 2018, first in 2019 and seventh this year.
These team effects are likely benefiting Allen as well. This season, Buffalo has surpassed Baltimore in play-action percentage, and about 47 percent of Allen’s passing yardage has come off those plays. While play-action surely isn’t the entire explanation for Allen’s improvement in accuracy, it points toward something more pervasive. It appears that teams with organizational structures in place to maximize talent tend to dominate the analysis of a prospect. What is seen as a liability can suddenly become a strength — on the right team. The Tennessee Titans’ reclamation of Ryan Tannehill — third in the league in play-action dropback percentage and fourth in the league in QBR this season through Week 6 — adds further evidence.
So while it’s still too early to answer the question of who’s the best QB drafted in 2018, we at least have a framework for answering the question — and for handicapping QB prospects moving forward. Bet on players who land on smart teams, and discount those who are drafted into chaos and dysfunction.
Check out our latest NFL predictions.