NFL Week 1 preview: Buffalo Bills at Baltimore Ravens

NFL Week 1 predictions: Two evenly matched teams?

Both the Buffalo Bills and Baltimore Ravens ended the 2017 regular season with a 9-7 record, but it was the Bills who advanced to the post season, where they were eliminated in an even lower scoring wildcard game in Jacksonville, 10-3. 

Had Baltimore been able to hang onto a 27-24 lead over bitter rivals Cincinnati instead of allowing a touchdown on 4th and long with just 44 seconds remaining in week 17, it would have been the Ravens, rather than the Bills playing in 2017’s wildcard weekend.

So superficially there would appear to be little between the two teams, based on the evidence from 2017. However, randomness can play a huge role in a relatively short 16-game season (as demonstrated by Baltimore’s rapid change of fortunes deep into the final throes of the 2017 regular season). 

We can get a much more nuanced view of a side’s core abilities if we look at their underlying statistics, rather than trusting their win loss records.

Buffalo had a negative points differential, allowing 57 more points than they scored. This was the equivalent of Cincinnati, also from the AFC who missed the playoffs, achieving a final seeding of 9th from just seven wins compared to the Bills’ 6th seed and nine wins. 

There is good reason to believe that Buffalo were fortunate to achieve nine wins. The same mark would have been a minimum expectation for Baltimore given their underlying statistics.

Similarly, Chicago in the NFC had a points differential of -56, one better than Buffalo’s, but were seeded a lowly 15th out of 16 in the NFC with only five wins.

It’s likely that nine wins flattered Buffalo and other indicators support this view. The Bills won five games decided by a touchdown or fewer points, losing just two such games and their Pythagorean win expectation that awards wins based on a side’s cumulative points scored and allowed suggested that just 6.5 wins would have been a fairer reward.

In addition to this, they had a turnover differential of plus nine, meaning they received the ball from turnovers nine more times than they gave it away to their opponents.

Turnovers are always beneficial to the side regaining the ball. They prevent the opposition from scoring points that come with possession and the beneficiaries often receive the ball in very good field position.

However, randomness also plays a role in creating and recovering turnovers and relatively high differentials in one season often give way to less extreme values in subsequent ones.

By contrast, Baltimore had far fewer question marks surrounding their nine wins season. They did have a league high plus 17 turnover differential, which will inevitably fall in 2018.

But they also had two more narrow losses than wins, a points differential of +92, the fourth best in their conference and a Pythagorean expectation of 10.5 wins, one and a half wins higher than their actual total of nine.

NFL Week 1 predictions: What do the stats suggest?

Although both teams won nine games last season, there is good reason to believe that Buffalo were fortunate to achieve those wins. The same mark would have been a minimum expectation for Baltimore given their underlying statistics. 

Offensively, Baltimore’s passing woes failed to improve in 2017. They passed for an average of over a yard per attempt shorter than the defences they faced usually allowed. 

Their drafting of a quarterback in Lamar Jackson in the first round suggests that veteran starter, Joe Flacco, will need to improve his passing efficiency or cede to his younger rival sometime during the 2018 season. He has been given upgraded targets to reverse his declining yardage.

The Ravens’ running game was around par last season, but they did manage to score 2.5 more points per game than their opponents usually allowed.

If we were take both teams’ 9-7 record at face value, the Ravens would be favoured by around three points on the Handicap. However, a closer look at the stats suggests that Baltimore is a superior team.

Defensively, they were a step back from the vaunted defences of the past, but they were above average against the pass, around par on the ground and shaved 1.5 points per game from the averages usually scored by their rival offences.

Buffalo were equally insipid through the air, with a throwing efficiency that was only 88% of the yardage per play allowed by the defences they faced and they have traded last season’s starting quarterback Tyrod Taylor to the Browns for 2018.

They were better on the ground, but offensively they fell a field goal short of the average points per game conceded by the defences they faced. They were slightly above par defending the pass, but below par against the run and conceded one point per game more than the offenses they faced scored on average.

NFL Week 1 odds: Where is the value?

Given the way these two teams play, the Total Points mark is set at a relatively low 40.5 for this game (the lowest across the eight game NFL schedule for this weekend). Baltimore are a fairly short Money Line favourite with the odds suggesting they have a around a 75% of starting their season with a win.

If we were take both teams’ 9-7 record at face value, the Ravens would be favoured by around three points on the Handicap. However, a closer look at the stats suggests that Baltimore is a superior team. The current Handicap is set at 7 with the Ravens priced at 1.961* and still offering potential value.

Matchups favour both passing defences, but Baltimore may do better on the ground and scoring rates stretch the Raven’s likely margin of victory to nine points and a total of around 40 points scored.