In this piece, I am looking for what stats may have been outliers for the NFL in 2020. The NFL tends to have a way of adjusting to trends. Whether it was because of empty stadiums, disappearing holding calls, or offenses getting ahead of defenses in scheme, NFL offenses ran the show in the 2020 season. This article will look at how often this level of growth has occurred in a single season and whether or not there was a corresponding correction in the following seasons. When we find items that have grown faster than historically, we are going to label them as “outliers” that can be expected to regress to the mean.
In order to find outliers, I am looking across 15 years of data (2006-2020) to find typical growth and regression rates in the NFL averages. For the purpose of this piece, I am considering any stat that has not had at least equal growth or regression at least twice over the 15-year period to be an outlier (with some exceptions made based on diving deeper into the numbers, explained below).
Additionally, for each chart below there is some color-coding involved. The below key should help in explaining the color-coding:
Team offense for the average NFL team has skyrocketed over the past 15 years. That means that team defense for the average NFL team has suffered. We all know that the NFL has become much more of an offensive game than a defensive game. The growth in offense and regression on defense in 2020 was staggering. The below table summarizes the findings:
- Points per game grew at a rate of 8.7% in 2020 (increase of 1.98 points). There are no instances of similar growth in the dataset, with the closest growth occurring in 2018 (increase of 1.63 points). After the growth in 2016, there was a decline in 2017 (decrease of 0.53 points). Points per game appears to be an outlier for the 2020 season, with regression potential for 2021.
- Yards per game has grown at a rate of 3.2% in 2020 (increase of 11.2 yards). There are two other instances of similar growth (2011, 2018). In those instances, one of the following seasons saw marginal growth and one saw regression. Yards per game is a weak outlier, but with the mixed evidence for the following season, there is not much support.
- In 2020, yards per play grew at a rate of 1.9% (increased 0.10 yards). There are three other instances of similar growth (2008, 2011, 2018). Yards per play does not represent an outlier.
- Touchdowns per game grew at a rate of 12.8% in 2020 (increased 0.31 touchdowns). There is only one other instance matching this level of growth (2018, increased 0.32 touchdowns). In 2019, there was marginal regression in touchdowns per game (decrease of 0.08 touchdowns). Touchdowns per game is an outlier for the 2020 season, with regression potential in 2021.
Now that we know how offenses exploded in 2020, it is worth looking into the passing game versus the run game. Below, I have embedded a table detailing passing offense for the average NFL team over the past 15-years:
- Net passing yards per game grew at a decent clip of 2.2% (increase 5.19 yards per game), but this is on par with typical growth for any given year. Five other instances in the past fifteen years have seen similar growth. Net passing yards per game does not represent an outlier.
- Net passing yards per attempt grew at a rate of 2.1% (increase 0.13 yards per attempt), also not unheard of in the last fifteen years. In four of the last fifteen years, the NFL has seen similar growth. Meaning that this category is not an outlier.
- Touchdowns per dropback were at their highest rate of the past fifteen years in 2020. This category showed year-over-year growth of 9.1% (increase of 0.38%). There is one other instance of similar growth (2018) and sharp regression followed the next season (decrease of 0.30%). Touchdowns per dropback does appear to be an outlier for the 2020 season, with regression potential for 2021.
- Interceptions per dropback was the lowest it has been over the fifteen-year period in 2020. It showed year-over-year regression of -3.8% (decrease of 0.08%). However, there have been eight other instances of this level of regression. In the following season, the interception rate grew twice but further regressed six times. The trends do not point to interceptions being an outlier.
- Pressures per dropback fell in 2020, declining -8.5% (decrease of 4.09%) from the previous year. This level of regression has occurred one other time in the past fifteen years (2008) and growth followed. Growth was minor in these instances, but pressures appear to be an outlier for the 2020 season with probability for growth in 2021.
- Sacks per dropback declined in 2020 at the largest rate seen over the past fifteen years, declining at a rate of -11.2% (decreasing -0.74%). This one is interesting, because the NFL had been near its fifteen year highs in this category in each of the previous three years. The closest regression level in the past occurred in 2007 (decreased 0.56%). In 2008, the sack rate decreased again, but at a more marginal rate. The next closest decrease occurred in 2016, and was followed by the largest growth in sack rate seen in the past fifteen years (increase of 0.64%). Sacks certainly represent an outlier for the 202 season as they decreased significantly despite a 5-year growth trend. They likely have a high chance for growth in 2021. However, with the influx of mobile QB’s entering the league a lower sack rate may become a reality.
Below, I have embedded a table detailing rushing offense for the average NFL team over the past 15 years:
- Rush yards per game grew at a strong rate of 5.3% (increase of 5.9 yards per game). In the past 15 years, the NFL has never seen this level of growth in a single season. The closest growth came in 2008 when rush yards per game increased by 5.04 yards. In the following season, the figure increased again by 0.72 yards. In 2018, rush yards per game increased by 4.75 yards, but the figure regressed in the following season (decrease of 1.55 yards per game). Rushing yards per game represents an outlier for the 2020 season with potential for decline in 2021. However, this is another instance where the influx of mobile QB’s into the league may be creating a new normal.
- Yards per attempt increased at a more modest rate of 2.2% (increase of 0.10 yards per attempt). However, this level of growth has happened twice (2008 and 2018). In the following seasons, yards per attempt regressed once and grew once, so there is no clear trend here either. I feel as though rushing yards per attempt could only represent a weak outlier with potential for marginal regression in 2021 despite meeting the stated criteria. This is because it has not grown often in the past 15 years, with large spurts in single seasons followed by a few years of modest decline. This too could be a changing environment due to QB mobility.
- Touchdowns per rush attempt grew 15.5% year-over-year. This is a large increase, but this level of growth has occurred twice in the past 15 years (2008, 2016). In the following seasons, the rate declined twice. Due to the historical evidence of regression, rushing touchdowns represent an outlier for the 2020 season with potential for regression in 2021. The same mobile QB disclaimer applies here as well.
Turnovers and Sacks
Below, I have embedded a table detailing giveaways (or takeaways) per game as well as sacks per game for the average NFL team over the past 15 years:
- Since we have already discussed sack rate and interception rate and the results on a per-game basis are very similar, I will focus on fumbles and total turnovers here.
- Turnovers per game regressed at a rate of -6.5% year-over-year (decreased by 0.09 TO’s per game). This level of regression has occurred four other times in the past 15 years (2008, 2011, 2014, 2016). In the following seasons, the rate grew twice and declined twice. Turnovers do not represent an outlier for the 2020 season but bear watching in 2021 due to the luck factor associated with turnovers.
- Fumbles per game regressed at a rate of -7.9% in 2020 (decreased by 0.10 fumbles per game). This level of decrease has occurred two other times in the past 15 years (2008, 2011). In both of the following seasons, the rate grew. Due to the historical evidence of growth in the following seasons, fumbles represent an outlier for the 2020 season with potential for growth in 2021.
Unfortunately, I only have the details on most penalty stats dating back to 2009. Therefore, all the stats below except for total penalties are limited to a 12-year window rather than a 15-year window. Below, I have embedded a table detailing penalties for the average NFL team:
- Penalties per game regressed at a rate of -16.7% year-over-year (decreased by 1.13 penalties per game). This level of decrease has never occurred in the past 15 years. The closest decline occurred in both 2007 and 2016, as both years experienced regression of 0.22 penalties per game. The 2020 figure wiped out two consecutive years of growth in penalties per game. Total penalties represent an extreme outlier for the 2020 season with potential for growth in 2021. It is an evident outlier based on the historical year-over-year changes. The 2020 figure changed five times faster than any other individual year.
- Drive-killing penalties are defined as penalties that occurred on either third or fourth down, regardless of distance or the result of the play(s) after the penalty. Simply put, these penalties had backbreaking potential for the offense.
- Similarly, drive-extending penalties are defined as penalties that resulted in a first down on any third or fourth down play, regardless of distance. These penalties saved the offense from having to run a play on third or fourth down.
- In 2020, drive-killing penalties and drive-extending penalties do appear to be outliers. However, looking at the numbers, the results appear to be in the neighborhood of historical results in recent years. I am getting the feeling that both of these stats are random in nature, and there is not much worth reading into here.
Offensive Penalties vs. Defensive Penalties
Below, I have embedded a table detailing offensive penalties and defensive penalties for the average NFL team:
- Offensive penalties in 2020 were the lowest they have been in the past 12 years. They declined at a rate of -23.6% (decreased 0.76 penalties per game). Not a single season analyzed approaches the decline seen in 2020. The next closest season was 2012, when offensive penalties decreased by 0.19 per game. Offensive penalties represent an outlier for the 2020 season, with potential to increase in 2021.
- Holding penalties declined 10.5% in 2020 (decreasing 0.50 penalties per game). This level of regression has not happened within the dataset. The next closest decline (2013) shows a decline of 0.11 penalties per game. Holding penalties were definitely an outlier for the 2020 season and show potential for growth in 2021.
- Defensive penalties declined 5.3% in 2020 (decreasing 0.14 penalties per game). This level of regression is marginal on a per-game basis but is matched only once in the dataset (2016). Defensive penalties are technically an outlier, but a weak outlier that I would not expect to see change much in 2021.
- The remainder of the penalties showed above changed at a marginal rate in 2020. Some of these might meet the criteria for being an outlier, but they are weak outliers. I would not expect to see large change in any of these categories, as their baseline is low to begin with (less than one penalty per game).
Outliers expected to Regress:
- Points per game: Regression potential
- Touchdowns per game: Regression potential
- Touchdowns per dropback: Regression potential
- Rush yards per game: Regression potential
- Touchdowns per rush attempt: Regression potential
Outliers expected to Grow:
- Pressures per dropback: Growth potential
- Sacks per dropback: Growth potential
- Fumbles per game: Growth potential
- Penalties per game: Growth potential
- Offensive penalties per game: Growth potential
- Holding penalties: Growth potential
I am utilizing this section to dump off some data I accumulated while figuring out how I wanted to write this piece:
Pressures Allowed per Pass Block Snap by OL Position (excluding spikes and kneel downs):
Sacks Allowed per Pass Block Snap by OL Position (excluding spikes and kneel downs):
Penalties per Pass Block Snap by OL Position (excluding spikes and kneel downs):
Entire dataset for this piece:
Listen To Bears On Tap Podcast
BET WITH BETRIVERS ILLINOIS
More Bears News
/ 4 days ago
Identifying trends and outliers from the 2020 NFL season and examining how we can...
/ 1 week ago
There have been conflicting reports regarding the Bears' starting QB. Here is a breakdown...