A week ago I posted analyzing statistical correlations, that relate strongest to winning, to help guide personnel decision making in the National Football League. The premise was made by the study of Brain Burke’s article in Advanced NFL Stats – What Makes Teams Win?
Seeing “the forest for the trees” – Stats that correlate to winning in the NFL
You gotta throw
As you might recall, Burke surmised that “passing efficiency” was a better overall indicator of team success than merely total passing yards. Passing efficiency takes into account not only average per pass attempt, but factors in yardage lost as a result of sacks.
The study also shows the ability to pass the football efficiently, and stop the pass on defense, are the most statistically significant factors in projecting wins. Even more important than being able to run the football and stop the opponent from doing so.
The results when you can’t
To further highlight Burke’s work, the following (0-3) teams rank respectively 28th through 31st in offensive pass efficiency; St. Louis Rams, Minnesota Vikings, Kansas City Chiefs, Indianapolis Colts. My former team, the Denver Broncos, sit (24) in Net Yards Gained Per Pass Attempt (NY/A). This factor alone doesn’t account for each team’s early futility, but it is THE strongest indicator of not being able to put a “W” in the left hand column. *(21) total sacks allowed, (20) in sack percentage.
That’s also why you’ll find Sam Bradford, Donovan McNabb, Matt Cassel, Kerry Collins and Tavaris Jackson littered about the bottom third of the NFL in most passing categories. Oh yeah, Kyle Orton is right there with them.
Gotta stop it too
Conversely, a team’s ability to stop the pass is THE second most statistically important factor predicting wins. The Carolina Panthers are dead last, surrendering 8.7 NY/A and all but negating the admirable start of rookie QB Cam Newton. This probably best explains their own 1-2 start.
You’ll find Kansas City, Indianapolis and Miami at or worse than (25). Denver sits at (20) with 6.8 NY/A. In 2010 the Broncos finished (30) in Defensive NY/A at 7.2. It’s still early to compare 3 weeks to 16, but there isn’t a big difference in overall performance to date. 6.8 equates to Dallas in 2010 (28).
Linked with defensive pass efficiency are Denver’s 5 sacks vs Oakland, Cincinnati and Tennessee; a 4.7% sack rate as compared to the NFL average of 6.4%. Those 5 sacks are tied with 6 other teams and place Denver (25) in the NFL. By comparison the Colts (24) have 5, the Dolphins (29) 4, and the Chiefs (30) 3; again, all winless to date.
Manipulating the numbers
At first glance, Bronco fans point to the improvement of the defense. Traditional ranking (by total yards) has them positioned (15) and their two losses are by a total of 6 points. But deeper indicators have Denver looking more like the 0-3’s than the 3-0’s.
Don’t believe in Burke’s statistical correlations? Then turn to Football Outsiders and Aaron Schatz. His sophisticated formulas have Denver’s pass defense at (29), while traditional total yards “mask” the Broncos at (13). That’s down with Carolina (23), St. Louis (28), and Miami (32).
So while the fans continue to clamor for Tim Tebow to replace Kyle Orton at QB, the inefficiency of the secondary and pressure upfront must be addressed by football scouting from the Personnel Department and Coaching Staff as well.
What’s the fix?
Certainly Denver has missed the play of DE Elvis Dumervil and DC Champ Bailey. Injuries have hampered or sidelined both for most of the beginning of 2011 and their return to the lineup might very well provide a boost. But their replacements and the rest of the defense can’t seem to generate the necessary production to lower an opponent’s pass efficiency and severely impedes the Broncos’ ability to win in the long run.
By the way, THE third highest correlation to winning? Defensive Interception Rate. The Broncos (32) have yet to record a pick in 2011. The Rams (29), Panthers (27), Dolphins (25).
Combined, (1-8).
Umm… The Panthers are 1-2.
You’re dang right they are, my mistake!