The 2017 NFL season has concluded (already?), and as such it’s time to look at how the first season of our PFFELO system went. Like all ELO systems, PFFELO is an iterative system that updates each team’s rating after each game based on two things: how well they played against their opponent and how well they were expected to play against their opponent. The PFFELO rating system measures the former using our unique offensive, defensive and special teams grades, while the latter is determined by the PFFELO ratings leading up to the game and where the game is played.
Sometimes in football the best team does not win, and PFFELO accounts for these oddities by using our grades. Hence there will be teams rated higher or lower in this system than their win-loss record would suggest. That being said, it’s nice to see the world champion Philadelphia Eagles end up with the rating system’s first spot, with the runner up second.
In this article, we provide the last installment of our rankings and elaborate on some examples each of where we went right and where we went completely wrong. The actual PFFELO ratings from Week 1 to the present time correlate at a rate of 0.490 (r-squared of 0.240), which is relatively high considering the backwards-looking nature of the algorithm and the substantial number of injuries we saw at the quarterback position throughout the year. With the noise inherent in teams like Green Bay, Arizona, Miami, Indianapolis and Houston losing their signal-callers for all or much of the season, we’re going to focus on teams that were mostly full strength at that position but still defied our rating system.
Final ranking: 1, Week 1 ranking: 14
Wow. While we’d like to say we saw this coming, we’re all wrong. With only scant (and mostly poor) data on Nick Foles going into our algorithm after Week 17, we gave the Eagles only an eight percent chance of winning the Super Bowl. With our numbers and the market fading the Eagles all through the NFC playoffs, we saw what was one of the best coaching efforts in modern NFL history by Doug Pederson and his staff. Against New England in the Super Bowl, they went for it (and converted) on fourth down multiple times, punted once and overcame a late-game surge by the Patriots to come from behind and win a shootout despite playing without their starting quarterback, left tackle, middle linebacker, ace special teams player and all-purpose running back. The play of Nick Foles (over 300 yards and two touchdowns when kept clean) will be what everyone remembers, but the contribution of mid-season acquisition Jay Ajayi (6.3 yards per carry), bargain free agent LeGarrette Blount (90 rushing yards) and undrafted rookie Corey Clement (100 receiving yards) should receive due praise in the history books.
Final ranking: 2, Week 1 ranking: 1
This is generally a hit for our algorithm, as the Patriots were atop this rating system for most of the season; requiring almost a perfect performance from the Eagles to get knocked from their post. Their defense, 24th in our rating system, finally broke when it mattered the most, allowing more than 10 yards per pass attempt to Nick Foles on his 30 unpressured dropbacks in the Super Bowl. Tom Brady was his fantastic self, again, this time generating outlier statistics (96.6 passer rating) when under pressure, throwing for nine touchdowns and just two interceptions when under duress. With questions surrounding the future of Rob Gronkowski, our highest-graded tight end this season by a mile, and others throughout the Patriots roster and coaching staff, it will be really interesting to see how much long they can sit near the top of a list like this.
Final ranking: 4, Week 1 ranking: 19
We could say many of the things we said about the Eagles about the Vikings. One of the league’s best defenses (they ended the 2017 regular season first in our rating system), coupled with unexpected play at the quarterback position from Case Keenum (not to mention an injury to Aaron Rodgers), and 8-8 turned into 13-3 pretty quickly. The offensive supporting cast of Stefon Diggs (85.4) and Adam Thielen (84.6) was one of the best in the PFF era, while their special teams ranked as the league’s best overall unit this season. It will be interesting to see what they do at the quarterback this offseason, as well as what they are capable of offensively once they get running back Dalvin Cook (82.0) back in 2018.
Final ranking: 6, Week 1 ranking: 23
The old adage of needing a defense and a running game to win in the NFL is likely wrong wholesale. That said, if you have a Hall of Fame quarterback and you add a defense and a running game, you’re probably going to have some success. The Saints nailed the draft, with the two biggest gems, cornerback Marshon Lattimore (90.5) and running back Alvin Kamara (90.2) turning what has long been a laughing stock of a defense to a legitimate contender-worthy group and a potent offense into a downright scary one. We’ll all be left wondering “what if?” regarding this group: What if the Vikings hadn’t stunned the world with a miracle finish in Minneapolis? Could they have gone into Philadelphia and won? We’ll have to wait until next year to see if they can answer some approximation of those questions moving forward.
Final ranking: 9, Week 1 ranking: 29
This is a classic case of our model simply not being able to predict how much of a difference a combination of a good head coach and a previously-poorly-utilized young quarterback can have on a team. Jared Goff thrived under Sean McVay, posting a 109.3 passer rating when using play action, and a 112.0 passer rating when clean (both top-seven at his position, with the latter top three). Running back Todd Gurley (91.9) was also a revelation, averaging 2.15 yards per route run (fifth among backs), while increasing his yards per carry from 3.2 in 2016 to 4.8 to 2017. It’s likely that the Seahawks will re-tool in an effort to take back the NFC West in 2018, it will be interesting to see if the Rams are up to the task of holding them off.
Final ranking: 11, Week 1 ranking: 30
While the Jaguars had been stockpiling talent on their defense for a few years now, the combination of Doug Marrone and Blake Bortles finally gave them the offense requisite of a contending team. Bortles still had something of a roller-coaster season overall, having the highest PFF grade in Weeks 13 through 15, only to finish the season with two relatively-poor efforts. Compared to offensive success (and success at the quarterback position specifically), success defensively is not all that stable season-to-season, meaning that if the Jaguars want to remain contenders, they’ll likely need either more out of Bortles or a new quarterback. Like the Vikings, it will be interesting to see where this goes.
Final ranking: 16, Week 1 ranking: 16
Here’s an example where we were not particularly thrilled with our numbers, even if they ended up dead on when it was all said and done. Despite trading their best wide receiver, cornerback and interior defensive lineman, all while trying to sabotage the quarterback position on a weekly basis, the Bills made the playoffs with a 9-7 record. Tyrod Taylor, even in light of having the league’s 14th-highest grade at the position, is likely gone this offseason, leaving a big question mark for the Bills at the league’s most important position. Couple this with the fact that they succeeded defensively in large part due to forcing a lot of turnovers (they were 10th in turnovers forced per possession), rather than being stout against the run or pass, and we’re going to need to see significant movement throughout the Bills’ roster to be confident in a repeat performance come 2018.
Final ranking: 25, Week 1 ranking: 11
Fans bullish on Derek Carr had to eat some crow come 2017, as the fourth-year signal-caller struggled through poor accuracy (16th in adjusted completion percentage) and even poorer play from his supporting cast (second in passes dropped by targeted receivers). The backsliding play of the offense was nothing compared to what was a horrendous performance of the defense at times. Oakland’s defense allowed the sixth-most yards per play and the fifth-most net yards per passing play in the league, grading out as the seventh-worst group by our metrics. We’ll see what Chucky can do to turn this around after a decade-long stint in the TV booth.
Final ranking: 26, Week 1 ranking: 10
While Oakland can partially blame their offense for their 2017 decline, the 2015-16 Denver Broncos didn’t have much of an offense to begin with, leaving their fall from an elite-level defense to a simply-good one as the main culprit. The ninth-best unit in our rating system (they finished 2016 first) allowed the 16th-most net yards per passing play, resulting in the 11th-most points allowed per game. With quarterbacks like Trevor Siemian (53.2), Brock Osweiler (47.7) and Paxton Lynch (42.3), slipping to even the bottom of the top-10 proves insurmountable in a division with the Chiefs and Chargers at the top.
Final ranking: 30, Week 1 ranking: 25
This is an example of what we saw above with Denver (and are worried a bit about with Jacksonville): It’s difficult to consistently win with mediocre (at best) play at the quarterback position and elite-level defense. The Giants’ defense went from third in our rating system in 2016 to 20th this season, showing how difficult it is to sustain that level of play, even with most of their starters returning. Add in the declining play of Eli Manning (28th in quarterback grades), and an injury to Odell Beckham Jr., and it’s not difficult to see how 11-5 can turn into 3-13 in a hurry. It will be interesting to see what new head coach Pat Shurmur does with the quarterback position, and whether he can get a talented defense to play more like they did in 2016 than they did in 2017.
Final ranking: 32, Week 1 ranking: 32
While many (including us) with an analytics bend felt somewhat optimistic about the Browns coming into the season, nothing ultimately came to fruition in what was the second 0-16 season in the history of the NFL. There’s talent on the defensive side of the ball in Myles Garrett (88.4), Briean Boddy-Calhoun (84.5), Danny Shelton (81.5) and Joe Schobert (77.4), as well as a solid offensive line on the other side of the ball. But, we’re unlikely to see the Browns make any real, meaningful progress until they figure out the quarterback position. DeShone Kizer (51.1) was about as poor as it gets, grading badly in basically every subset of throws during his rookie season. With tons of ammunition in the draft, and a decent crop of quarterbacks coming out, there’s no excuse not to not land a new signal-caller for Cleveland this offseason.