Regression Might Hit the Oakland Raiders Hard in 2017

The Oakland Raiders make headline after headline during the off-season. The team that finished 12-4 a season ago signed QB Derek Carr to a record-setting 5-year, $25M-deal and signed one of the most fascinating Running Backs over the last decade – Marshawn Lynch – out of retirement to put him behind a top-10 offensive line. The media hype cannot be higher on a certain team. The fan base talks about a serious Super Bowl contender in the AFC. As I have written in 2016 (Oakland Raiders GM Reggie McKenzie is turning this team into a serious playoff contender), I love the work Raiders GM Reggie McKenzie has been doing for the past few years, upgrading this team tremendously. One season after, are they really a serious Super Bowl contender?
Let us start with a review of 2016. As overwhelming as the 2016 season was for Raiders fans, as suspicious it was for an objective observer like me, who also loves describing the beautiful game of Football by math. The Raiders achieved a point differential of +31. This is the lowest value any team with 12+ wins has ever produced since 2002. Even if you take away the week 17 game in which Derek Carr did not play, their +49 differential on a record of 12-3 would still be the lowest value since 2002. Next to the Raiders, there have been 10 other teams since 2002 who produced a point differential lower than +100. The next fewest belongs to the 2015 Broncos (+59). That team had one of the best defenses in NFL history and played almost eight games with Brock Osweiler. With Peyton Manning, the Broncos had a +53 until week 9. Overall, the average for 12-win-teams is over +100. That means, based on the point differential, the Raiders were by far the worst team with 12+ wins since 2002.
Let us take a more detailed look into the season. They went 9-2 in close games. Close games are one-possession games within 8 points of negative or positive winning margin. The record in close games is a good indicator when predicting a team’s performance going forward, especially in the playoffs and into the next season. You can find the reasoning in my 2016 article regarding seasonal regression methods: Seasonal regression methods in the NFL. Winning close games is not a quality. In most of the games, it requires key plays that can go either way. A quality is when teams dominate their opponents and do not require key or random plays in the fourth quarter. Casual NFL fans tend to say: “Does not matter. A win is a win”. That is simply not true. It matters how you win your games. Here are 7 of the 9 close wins:
Week 1 at NO: Saints were leading 24-13 in the fourth quarter when Will Lutz missed a 50-yard FG to make it 27-13. The Raiders made a crazy comeback (Kudos to this) and had a 4th & 5 at the NO 18 with 1.41 minutes to go. Derek Carr missed a throw to the outside to Jalen Richard, but Saints-LB Craig Robertson committed pass interference. Instead of ending the game, Carr and his offense got another shot they converted for six. Instead of going into overtime, Jack Del Rio had the balls and decided to go for two and Derek Carr hit Michael Crabtree with a defender at his body. What was supposed to be a loss, turned out as a win.
Week 4 at BAL: Raiders were outgained by 151 total yards, but they got two touchdown drives off a start within the Ravens 29-yard line due to a fumble and a long punt return. They won 28-27.
Week 5 vs SD: It was a wire-to-wire game. With two minutes left in the game, the Chargers attempted a game-tying FG at the OAK 18 that holder Drew Kaser fumbled. Raiders were not forced to try a game-winning drive or go to overtime.
Week 8 at TB: Sebastian Janikowski missed a field goal at the end of regulation and on the first overtime drive that would have put the Bucs into a pressure situation. However, the Raiders won with 1.45 minutes to go in OT by a 4th down conversion.
Week 11 vs HOU: Oh boy. Do you remember that crazy blown game by the refs? If not, please watch it again. On one play, the Texans were denied a TD pass to DeAndre Hopkins, but the referees called him out of bounds. There is still no clear evidence that his foot touched the sideline as his toes were inbounds. At 20-20, the Texans faced a 3rd & 2 and 4th & 1 within the Raiders’ red zone. Both times, Lamar Miller and Akeem Hunt converted a run for an obviously sure first down, but both times the referees called it short. Bill O’Brien even challenged the  4th down play and it was obvious that Hunt was over the line for a first down, but the refs denied it. Raiders got the ball back and scored the game-winning touchdown.
Week 12 vs CAR: Raiders were leading 24-7 at halftime, but all of a sudden, Derek Carr brought the Panthers back into the game, because they scored touchdowns off a fumble and interception by Carr. At 32-32, the Raiders executed a very good game-winning drive to win the game. That would not have been necessary if they would have been able to control the game.
Week 15 at SD: Raiders won 19-16 off two fourth quarter field goals. One off a fumble by San Diego in their red zone.
What do these games tell you? The Raiders fans do not care, because their team won. Nevertheless, anyone else should see seven games against six non-playoff teams that could have gone either way. These games even out in the long run and big close games discrepancies regress in almost every case from one season to another. If the Raiders have gone 8-8 or 9-7 by going 3-4 or 4-3 in the games above, no one could argue with that. People would be saying “Hey, it is a young and developing team that goes through a progression”.
Pythagorean win expectation is a method to calculate how many games a team should have won based on their offensive and defensive scoring. The Raiders had a pythagorean win expectation of just 8.7 wins. That means they won 3.3 games more than the average team should win based on how much they score and how many points they concede. Even by considering only pure offensive and defensive scoring (no defensive or special teams scores), the Raiders should have won 8.7 games. Football Outsiders calculated 8.8 estimated wins for the Raiders based on DVOA efficiency in important phases of the games. Statistically, teams with high discrepancies between their pythagorean win expectation and actual records tend to regress the other way in the upcoming season. There are only few exceptions in the past.
Even though having a bottom-10 defense overall in terms of efficiency, the Raiders managed to produce a +16 turnover differential which ranked first in the league along with the Chiefs. They had 30 takeaways (2nd) and won 14 fumbles (T-2nd). The offense produced only 14 giveaways (4th) whereas Derek Carr only threw 6 interceptions. According to Cian Fahey, who charted each pass of each QB in 2016, Derek Carr threw 24 interceptable passes – passes that were off target and could have been intercepted by a defender. But only 6 of them were have actually been caught. Carrs low 25% rate of interceptable passes caught ranked 3rd in the league. The average turnover differential is around 0. Crazy outliers tend to regress heavily from one season to another, as you can read in my regression article. It is impossible to sustain that turnover differential of +16. It is more likely to regress towards 0. Also Derek Carr might not have that same luck of defenders not catching his interceptable passes. If he throws 13 interceptions in 2017, it might not be because he will be playing worse than in 2016, but just that variance will hit him hard. In 2017, he can be an improved QB whose bad passes just have a worse outcome.
If you combine all that, you will come to the same conclusion as I do: The 2016 Oakland Raiders were an improved team and fun to watch, but their 12-4 record is somehow misleading. It was guided by a lot of positive variance that is not sustainable in the long run or from one season to another.
2017 Outlook:
The offense has changed quite a bit. The RT position might be a weak spot with Marshall Newhouse projected to start. Other than that, Marshawn Lynch will boost this offense with his phenomenal running ability. But Lynch has turned 31 in April, so we shouldn’t expect him to see 280+ carries like he saw in his last few full Seahawks seasons. Amari Cooper and Michael Crabtree are still on of the best WR-tandems in the NFL at this point, but the Raiders downgraded their TE position. Most of the people remember Jared Cooks great sideline catch against Dallas, but other than that, he was a huge liability, because he has hands like stone. He drops too many passes and does not make good adjustments to the ball. He will take care for some 3 & outs this year.
Derek Carr is a promising young Quarterback who improved a lot since his terrible rookie campaign. He has great arm strength and can make every throw:

But he still has some major flaws, especially in tight pockets or under pressure. Bill Musgrave did a phenomenal job in letting him drop deeper than any other QB in the league and make quick reads most of the time. He is not a QB who wants to hold the ball too long in the pocket:

That is why the Chiefs defense owns him every time. It is gonna be interesting how new OC Todd Downing schemes this offense to make Derek Carr feel comfortable. Unlike media pundits, I believe that Derek Carr needs to improve a lot to put this offense on a championship caliber level. The huge contract means very little to me, because the Raiders simply had to sign him to a longterm deal. Otherwise 5-8 other teams would have paid the same amount of money. The offense is very talented and get a new dimension with Marshawn Lynch. But a highly possible regression by Derek Carr in terms of turnovers, the questionmark on RT and Jared Cook as the #1 TE make me predict the offense not to be as productive as in 2016.
In 2016, the Raiders had a bottom-10 defense. Let’s assume that CB Gareon Conley has a standard developmental phase, the CB group of him, David Amerson and Sean Smith might be a solid one by the end of the season. Karl Joseph needs to make another step forward, but as a whole, this secondary is still young and does not look to improve vastly in 2017. The LB corps (Jelani Jenkins, really?) is a major weak spot, I think it is not debatable. With Khalil Mack they got a freak on DL, but the rest cannot be considered good enough to make a big leap in 2017. Mario Edwards played only 42 snaps due to injury last season, he might improve the line. Actually, I do not see an improved defense.
The Raiders have a very fair first half schedule, but after their bye week they might have the toughest schedule of all teams:

I can really see the Raiders losing 5+ games after their bye. At this point, I only have them favored over the Broncos at home based on my personal Power Rating.  They finish the season with three road games during the last four. They simply have to go into the bye with a very good record to have a shot at making the playoffs. They always struggle against the Broncos and get killed by the Chiefs. This year the Raiders will face a highly improved Chargers team whose injured 2016 version they barely beat by three points each. Before their bye they have three divisional games and a tough early road start at Tennessee. This schedule as a whole is anything but easy.
Prediction for 2017:
Regression might hit the Raiders very hard in 2017. It does not mean that they are not a talented team that will fight for the playoffs though. Combining that with the strong schedule, especially after the bye, it will shock me if the Raiders win more than 10 games this year. Even winning 10 games would be a lot to ask for in my opinion. I predict the Raiders to finish the season somewhere around 8-8. 9-7 is possible, but so is 7-9 if some small things go wrong. Oddsmakers will inflate the Raiders’ spreads in 2017 and bettors will take advantage from the public overhyping them. We might see a lot of good opportunities to bet against the Raiders with solid spreads. Even if they manage to leave the first nine games with a good record and good performances, bettors will take advantage in the second half.

Share on facebook
Share on google
Share on twitter
Share on whatsapp
Share on email