During the offseason, we looked at what wins NFL games, using Pro Football Focus grades to look at what correlates to overall success. Now that we are the mid-season point, why not do the same? In this article, we will attempt to answer three questions:
- What has correlated to success in 2020?
- How does this differ from the past?
- Which teams have been the most/least lucky this season?
In our previous study, it was clear that games were being won in the perimeter, rather than in the trenches. As we saw with the way teams coveted speed this offseason, it is clear that the NFL is starting to catch on. Meanwhile, due to limited training camp and the COVID-19 pandemic, the dynamics of this season are far different than seasons of the past. Thus, my hypothesis is that there will be some interesting quirks regarding what has correlated to success this year, and we can use this information to judge which teams are likely to positively or negatively regress throughout the rest of the season.
Let us start with comparing offense to defense. Previously, offense was found to be roughly 1.8 times more important as defense. This year’s results, however, may surprise you:
Here is offensive/defensive Pro Football grades plotted on the y-axis versus average point differential on the x-axis. As you can see, offensive and defensive performance have been nearly identical in terms of importance. Considering that teams are scoring roughly three more points per game, this comes as a surprise. However, it makes some sense. If points are easier to score this season, then the gaps between offenses may not be as strong as they generally are. Meanwhile, if a team can limit the amount of points teams are scoring, that may make more a difference than it has in the past.
Yet, the difference between the best and worst offenses remains stronger than that of the best defenses versus the worst defenses. Thus, in the end, I would label this as a result of a small sample, especially as some of the top defensive teams (Steelers, Colts, Bears) are having more success than expected, the team’s best team (Chiefs) is dominating on both sides of the ball; I’m guessing that the gap in importance between offense and defense will become more extreme over time.
With that in mind, let us analyze the teams that have been the luckiest and unluckiest. Using our previous knowledge of what correlates to a strong point differential, here are the teams with the greatest differences between their actual point differentials and their expected point differential, based on Pro Football Focus offense and defense grades.
- Miami Dolphins
- Arizona Cardinals
- Baltimore Ravens
- Kansas City Chiefs
- Las Vegas Raiders
HM: Carolina Panthers
Currently, the Dolphins have relied on a defense that is first in points/game allowed. However, not only are they relying on defense, but said defense’s overall unit grade ranks just 20th in the NFL. The Raiders, meanwhile, are the league’s second-lowest graded team, while the Cardinals (22nd) have been far from a powerhouse. As for the Ravens and Chiefs, the two premier teams in the AFC, they simply haven’t been as dominant as their ranks in point differential (3rd for BAL, 1st for KC) would suggest.
Is it worth noting that all of these teams benefit from tremendous coaching. That is an outside variable that is going to become even more a theme as we shift towards looking at specific units.
As for the unlucky teams:
HM: Titans, Broncos
The Browns’ overall offensive grade is high, but their main issue has been unstable quarterback play, which is also true for the Jets; I am not buying them as unlucky teams. The Seahawks, Vikings, and Texans, on the other hand, rely on their offenses, so as the gap between offense and defense strengthens, they are the teams that stand to benefit the most.
Previously, we had found that by far, a team’s quarterback and receiving production were the greatest indicators of scoring points.
Is that still the case? Let us take a look at the coefficient of determinations (r-squared values) between average points scored and different Pro Football Focus grade variables. In other words, how much can a team’s offensive success be predicted based their grade in various aspects.
- Offensive Grade= .532
- Passer= .575
- Pass Protection= .269
- Receiving= .36
- Rushing= .004
- Run Blocking= .036
- Expected Offensive Grade= .622
- Passing Offense= .624
- Rushing Offense= .032
In the past, there had been an 83%-17% ratio between passing offense and rushing offense. This year, however, that is around 95%-5%. Given how fun it is to watch teams embrace explosive passing offenses, here is hoping that this a trend that is here to stay! As for more specific variables, a team’s passer is as important as it has been in the past. On the other hand, the gap in importance of receiving grade and pass-blocking grade has diminished tremendously. Considering that one of the top teams in receiving grade in the 49ers have suffered through poor quarterback play, while COVID-19 has had a particular impact on offensive lines, I am not ready to label this as a sign that winning on the offensive trenches is going value; most of the top receiving teams are also ranked near the top in pass protection, for whatever that is worth.
My expected offensive grade is calculated based on a team’s performance in the smaller variables of the offensive grade. So far, I am proud to say that, based on its higher correlation than overall offensive grade, has proven to be a useful tool that will continue to play a major role in my projection models.
Lastly, why is passing offense so important this season? It is hard to tell, but since the lack of training camp has had an affect on simple fundamentals of defense, such as run fits and tackling, I would lean on that as a reasonable explanation.
Similar to what we did for overall team performance, we will use expected offensive grades to look at which teams have overachieved or underachieved; who is scoring more/less points that they should?
HM: Cardinals, Broncos, Raiders
The Steelers, Ravens, and Dolphins are all benefitting from turnover luck, so it makes sense why they are on this list; Miami, for example, scored multiple times on defense and special teams last week, while the Steelers scored 28 points against Baltimore with the help of great field position from turnovers. The Eagles, meanwhile, have not gotten productive quarterback play at all from Carson Wentz, and the Saints have run their offense exclusively through running back Alvin Kamara- that is not stable.
As for the unlucky offenses:
This helps explain why the Texans and Vikings have unachieved overall this season. Both have top-ten graded quarterbacks, while Minnesota has the best-graded receiving corps in the NFL. As for the Patriots, my model ranks them higher than others may, as their turnover problems have hurt them, as is the case with the Giants. I wonder if the Bills’ participation in multiple bad-weather games has something to do with their appearance on this list.
In our previous study, pass defense was far superior in importance to run defense. Also, pass coverage was far more important than rushing the passer. Similarly to the offenses, let us re-check the coefficient of determinations for each defensive variable, plotted against points allowed:
- Defensive Grade= .322
- Run Defense= .023
- Pass Rush= .12
- Coverage= .392
- Expected Defensive Grade= .376
- Pass Defense= .407
Similarly to offense, the ratio in significance between pass defense and run defense has shifted towards 95%-5%, which I would believe to be the case for the same reasons I suggested in the offense section: the weirdness of a shortened offseason.
Coming into this study, I expected pass rush to play a bigger role. After all, we’ve seen the Steelers, Bucs, Rams, and Bears get after the quarterback at a high level.
Do you know else those defenses have? Above-average coverage units. Here are some other defenses that rank in the top-half of pass-rush grade: Eagles, Chargers, Cowboys, Falcons. None of them rank in the top-ten in points/play allowed.
The importance of coverage versus pass rush has remained the same this season, and that explains why the expected defensive grade is having success this season. Overall, though, I recommend looking at these r^2 values compared to offense. Notice how much smaller they are! This is evidence that defense is far more reliant on the opponent than is the case for offense, which is why trying to win through defense is generally a poor idea.
With defenses being reliant on who they are playing, the “luck” factor here should be more evident. Let us see if that is the case:
All of these teams rank in the upper-half in takeaways/game, which explains the “luck” they me from benefiting from; we already went into how Miami and Arizona aren’t particularly grading well defensively. The Raiders are the team I want to highlight here, though. As much as their offense has benefitted from unstable production, their defense has yet to be exploited the way it should. Despite a 47.5 defense grade, they only allowed six points to the Browns, as drops were an issue for Cleveland in a poor-weather game. Overall, when you’re by far the lowest graded defense, you probably should rank lower than 24th in points/game allowed.
Now that we’ve gone over the process of expected offense and defense grades based on past correlation, and how it is holding up this season, let us look at which teams are most lucky; the difference between expected point differential, based on adjusted offense and defense grade, and actual point differential.
HM: Steelers, Panthers
This is roughly the same list as our first examination of the luckiest NFL teams. The Steelers are now included here, though, as they have not received proper production from quarterback Ben Roethlisberger.
As for the unlucky teams:
- Washington Football Team
HM: Vikings, Texans
Slightly different list here. The Broncos have generally been unlucky on both sides of the ball, while Washington’s offense has graded better than it has performed. The Eagles, meanwhile, have unlucky defensively. It is funny that the Bears have been seen as a fluke because of their point differential. Overall, they have been lucky offensively, yet their defense has been hurt tremendously by poor field position.
So, what should we take away from all of this?
This season, offense and defense have been equally important. I see this as more of a small-sample issue, though, so teams that have won through defense should not get too comfortable.
Also, there are far more overachieving offenses than there are defenses. That is a result of points being easier to produce than ever, which can be attributed likely to a shortened offseason, similar to the greater significance of pass protection.
Is the boost of reliance in passing offenses and defenses related to the shortened offseason or a sign of things to come? Given the way front offices are starting to build their rosters, I do not believe this is completely a fluke, though I believe it has been escalated by the lack of fundamentals, which is more present in the run game than anywhere else.
Overall, the Dolphins, Cardinals, Ravens, and Raiders have overachieved this season, which is why my model likely rates them lower than you’d expected. The inverse is true for the Broncos, Seahawks, Vikings, Texans, and patriots, who should have a better point differential than they currently have.
Is this due to a strange season, small sample, coaching, or all of the above? The answers to that will become more evident as the season plays out, as we’re all living in a new world right now. At the end of the day, though, I am going to trust previous trends over what has been the case this season; doing so ought to allow us to better predict how the rest of the season will play out!