NFL Week 1 outliers: Which surprising trends will continue?

Like a lot of other commentators and groups in the football world, Football Outsiders makes a series of predictions before every NFL season. Then Week 1 of the season comes and a lot of those predictions suddenly seem silly. On the other hand, it's only one week. A lot of things can happen in that kind of small sample size.

You might remember the year Kevin Ogletree was everyone's hot fantasy waiver-wire pick after he had 114 yards and two touchdowns in Week 1. Ogletree had just two more touchdowns the rest of that 2012 season. Or maybe you remember all the way back to last year, when Ryan Fitzpatrick threw for 417 yards with four touchdowns while beating the New Orleans Saints in Week 1. In the long run, the Saints were fine and Fitzpatrick was back on the bench.

Let's go through some of the outliers of Week 1 and other results that contradicted Football Outsiders' preseason projections. Should we expect the rest of the season to look more like Week 1, or more like the forecast from the first week of September?

Baltimore dominates

Baltimore's win over Miami was legitimately one of the biggest wins we've seen in quite some time. How big? Based on Football Outsiders' DVOA, it was the fifth-largest win in Week 1 since 1986. The offense, expected to be historically run-heavy, instead passed the ball for 379 yards and six touchdowns. We had projected Baltimore as a borderline wild-card team with a below-average offense. Are the Ravens better than that?

What's remarkable about Baltimore's huge win is that this is the third straight year the Ravens have blown out their Week 1 opponent. Two years ago, it was a 20-0 shutout in Cincinnati. Last year, it was a 47-3 dismantling of the Bills. But this year's win is a little different because the offense was so much better. The Ravens had more net offensive yardage in this year's win (643) than in the 2017 and 2018 Week 1 wins combined (637). Only Dallas had a more efficient performance by DVOA.

However, we also have to consider just how bad the Miami Dolphins are likely to be this year. This is not a run-of-the-mill bad team. The Dolphins ranked dead last in both our offensive and defensive projections before the season. They were the first team to fall below 5.0 mean wins in our preseason forecast since the 2010 Detroit Lions. Miami's weakness doesn't erase the impact of Baltimore's huge Week 1 victory. But it does dull it a little bit.

Verdict: Baltimore's offense is likely better than expected but not as good as it looked on Sunday.

Pittsburgh struggles

The Steelers, not Baltimore or Cleveland, were the AFC North favorites, according to both Football Outsiders and ESPN FPI. They went out on Sunday night and got destroyed 33-3 by the defending champion Patriots. The defense allowed Tom Brady 9.0 net yards per pass, while Ben Roethlisberger had just 5.8 net yards per pass and finished 28th in QBR for the week.

There were two particularly worrying issues for the Steelers. First was the difficulty of finding a go-to receiver for Roethlisberger other than JuJu Smith-Schuster. Smith-Schuster was draped by Stephon Gilmore, one of the game's best cornerbacks, and had only two targets (and catches) in the first half. Roethlisberger found Smith-Schuster more often once the Patriots' defense relaxed a bit in the second half, but he had trouble making strong connections with his other receivers, especially Donte Moncrief. For the game, Moncrief caught only 3 of 10 passes for 7 yards and no first downs. Moncrief had four dropped passes; the Steelers as a whole led all teams in Week 1 with six drops.

Other incomplete passes were on Roethlisberger rather than his receivers. He had eight passes defensed, plus a late-game pick. Nine of his passes were marked by ESPN charting as overthrown or underthrown. Roethlisberger was terrible despite being under pressure on only 18.4% of snaps, 25th in the league, according to ESPN Stats & Information. All quarterbacks play better without pressure than they do with pressure. So for Roethlisberger to play that badly without much pressure is not a good sign.

The Steelers' defense perhaps has a bit more of an excuse; after all, the Patriots are one of the best offenses in the NFL year after year. But they also struggled to get pressure; after ranking fourth in pressure rate last season, they were just 23rd in Week 1. Some of that is the effect of the Patriots' offense, but still, the Steelers will need to bring more pressure to cover up the weaknesses in their secondary. At least their run defense was solid, holding the Pats to just 3.4 yards per carry and a 34% success rate.

Verdict: Worry, especially about the offense.

Chicago's defense vs. regression

One of our strongest predictions in the preseason was that the Chicago Bears' defense would regress toward the mean. Well, it certainly didn't look like it got any worse on opening night. The Bears held the Packers to just 10 points and 3.7 net yards per play. The Bears ended up sixth for the week in Football Outsiders' DVOA ratings.

Except, that's about where the projection had them. Our forecast was not that the Bears would regress into being an average defense, but that they would decline from "historically awesome" to just "one of the top four or five defenses in the league." One big reason for that was turnovers. The Bears led the league last year with takeaways on 19.1% of opponent drives -- in particular, interceptions on 14.8% of opposing drives. And in the last 10 years, the team that led the league in interception rate ranked just 13th on average the next season.

And on opening night, while Chicago limited Green Bay's yardage, the Bears didn't get a single turnover. Yes, some of that is Aaron Rodgers' ability to avoid interceptions (and his preference to throw the ball away instead of risking one). But the Bears also had only one forced fumble, and they didn't recover it. There's a lot of luck in the bouncing balls that become turnovers, and it's still very likely the Bears won't have as much of that luck as they did a year ago.

Verdict: Our projection still looks solid.

Kansas City's offense vs. regression

Like Chicago's defense, Kansas City's offense last season was one of the 10 best that we've ever measured (since 1986). And like Chicago's defense, we expected some regression from the Kansas City offense just because it's very hard to play that well two years in a row.

Offense is more consistent from year to year than defense, which means offense tends to regress toward the mean less than defense. Still, the top 20 offenses in DVOA history ranked, on average, just sixth the following season (fifth if we include only offenses with the same quarterback).

But so far, so good for the Kansas City offense in 2019. The Chiefs finished third in Week 1 offensive DVOA behind Dallas and Baltimore. That's particularly impressive because this early in the season, we don't yet include opponent adjustments in our numbers. And the Chiefs scored 40 points on a Jacksonville defense generally regarded as one of the best in the league. In fact, the Jaguars were Football Outsiders' No. 1 projected defense before the season began.

Verdict: Full speed ahead, as the Chiefs probably will regress a lot less than our numbers expected.

Malcolm Brown's big day

We all knew that the Los Angeles Rams were likely to lessen Todd Gurley II's workload this year. The surprise in Week 1 was who took those carries. Instead of hyped third-round rookie Darrell Henderson, the carries went to veteran backup Malcolm Brown. Brown had 11 carries for 53 yards and, after scoring just two touchdowns in his first four NFL seasons, scored another two on Sunday.

What was even more interesting was how Brown was used. It was Gurley who iced the game when the Rams ran out a lead in the fourth quarter; all of Brown's carries came in the second and third quarters. But Brown was the red zone back, getting five of L.A.'s six red zone carries (Henderson got the other -- his only carry all game).