Get Active!
Previously, this space discussed the statistical activity of Defensive Ends through review of their "Elemental Statistics" (or ES) which is explained at length
here, but for the sake of simplicity, they are the compiled defensive statistics which include sacks, interceptions, forced fumbles, passed defended, QB Hits, Tackles for Loss, fumble recoveries, safeties and touchdowns. All are given equal weight in this exercise (although I am in the process of weighting them based on their correlation to both wins and Points Allowed...stay tuned) and measured for frequency to find out at what rate they are gathered.
Divide and Conquer
Because each position player's role in the defense results in different opportunities to generate ES, we will compare players within their position groups on an 'apples to apples' basis. The positions are presented based on the most elite groupings in terms of the fewest players which grade out to an A- or better.
The cut off for inclusion was a player had to have snaps at least equal to 1 standard deviation below the mean for the entire survey of over 700 players.
Color coding for Snaps/Elementary Statistic (Snaps/ES) is: Green = greater than 1 standard deviation above the mean; Yellow with Bold font = between 1 standard deviation above the mean and the mean.
Cornerbacks
Corners are the smallest group given their ES opportunities are fewer than positions like Linebacker. Based on the group average, we assign a grade of 100% to the top performer and then base the rest of the field on overall score distribution for the position.
- The group above is headed up by Alexander of Minnesota who averaged one ES for every 22.77 snaps.
With just 11 players earning "A" grades, this was the most exclusive position in the survey.
- Alexander's margin of victory compared to the other positions was 4% which is 2.2x greater than the next closest position race!
- Notwithstanding the large grade margin for Alexander, the top CBs were all within 1 standard deviation above the mean so the group as a whole was quite similar.
Linebackers
Much longer list, obviously given you have 3 or 4 LBs on the field at one time.
Von Miller is the most statically active LB with one ES every 13.72 snaps.
DEN, KC and BAL each have 2 LBs in the top 10...Frightening. (If you take it out to the top 12, CHI has 3 players).
Again, the group is fairly similar as a whole as there are no players performing beyond 1standard deviation above the mean.
Safeties
The safety group is the first to provide "Green" players who performed beyond 1 standard deviation above the mean.
Adams, James and Levine are all in the top tier of safeties in terms of statistical activity.
Levine is the one name you may not know but he has been a solid contributor on the Ravens for some time now.
BAL is the only team to boast two safeties in the top 10.
Defensive Tackles
Defensive Tackles had six players perform at a higher level than the rest of their position group.
Donald is not surprisingly the best gatherer of ES in the league, gaining one every 10.91 snaps.
The distribution of "A" talent is diverse as no team has more than one top Statistically Active DT.
This group has a lot of old timers including recent league retiree Kyle Williams, the oldest A rated player in our survey for 2018.
Defensive End/Outside Linebacker
Edge is the most Statistically Active position with nine players ringing up stats like Dave Allen, greater than 1 standard deviation from the mean.
PHI had 3 in the top 3 in 2018; the retirement of Chris Long will obviously have an impact on the Eagles performance for 2019.
With the addition of Frank Clark, KC also has 2 in the top 5 performers.
Clowney is now in SEA so we will see if he feasts like Frank Clark did in 2018 or if he will provide above average ES generation like he did last year.
Summary Table
The above table details the averages for the players presented in the positional tables.
The bottom two rows measure the correlation between 2018 team wins and points scored against each team's defense.
The strongest correlation of the areas examined is between Average Player Grade and Points Against (which has a negative relationship), that is a moderate correlation exists so we can expect the higher the Grade for players from the positional ES tables, the fewer points we can expect the opponent to score (PA total).
There is also a moderate correlation between Snaps/ES as expected where the higher the rate the lower the win total (again, a negative relationship) and the higher the PA.
So What?
OK, so we sort of expected increased statistical activity for defensive players would result in some sort of correlation with wins and points allowed but they was only moderate by statistical standards.
It was interesting to see which positions and which teams are most active. The ongoing challenge will be to continue to monitor player movement to see how much of an impact certain players had on the win totals for prior years and teams? If Player X was on Team Y last year and we see, year over year after he left they had a decline or increase in Snaps/ES, can that tell us anything about his numerical value to his particular team when comparing performance to wins and losses? What about his new team; will their gain in ES be in line with that lost by his former team?
No idea, but it will be interesting for us to keep asking and evaluating right here so keep an eye out for future updates.
Thanks and feel free to leave a comment!
****************************************************************
What defensive stats have the most impact in terms of Wins? Glad you asked...check it out right here!