We’ve reached the lull between free agency and the 2014 NFL Draft. Not much will be happening in the next couple weeks; the big fish have all been reeled in and free agency is winding down. I'm going to break from discussing draft prospects and take a closer look at what makes a winning team in today’s NFL.
If you've ever met me in the comments section of a FanPost, you know that I love statistics. The problem I've found with statistics is that they don’t always tell the truth. Numbers can mislead if they aren't put in the proper context and professional football is a complex team sport… where every player on each side of the ball has a specific assignment on each play and his teammates depend on him to execute. Football can’t be described completely by statistics, but in this article I will try to "define" the elements of a winning football team in the 2013 NFL… What do winning teams have in common? What are the factors that really make a difference? Can power-football still win today’s ‘pass-happy' NFL? Should some positions be viewed as 'more valuable' than others?
Here's How I Went About It
I used Pro-Football-Reference.com as my source for all of the statistics. I started by plugging all of the 2013 statistics by team totals into a spreadsheet (Wins, Points, Turnovers, QB Rating, etc…). I then used a statistical software to calculate the Correlation Coefficent, which I’ll refer to as CC from here on.
Correlation in a Nutshell
Correlation is basically the relationship between two variables. Think of driving... you step on the gas and you're RPMs go up and so does your MPH. These two variables have a relationship that if one increases the other also increases (if you're driving on a nice flat highway with no wind).
It's easier to sum it up in graph. Here are some good examples:
Positive Correlation: The CC (correlation coefficient) gives a measure between +1 and -1 of how strongly one statistic is correlated to the other. A CC of 1.0 means 100% positive correlation, so there is a direct relationship between these two statistics and as one gets larger the other also gets larger. For examples, Pass Completions and Passing Yards have a very strong correlation of .88.
Negative Correlation: A CC of -1.0 means that there is a 100% negative correlation, so as one statistic goes up the other goes down. For example, Int's and QB Rating have a strong negative correlation of -.75, so as Int's increase the QB Rating tends to decrease.
No Correlation: A CC of 0 means that no correlation can be found between the two statistics. So whether one statistics goes up or down has no effect on the other statistic. An example of this is Total Plays and Yards/Completion… they have a CC of -.01, meaning they have no detectable relationship.
After the statistical software calculated all of the CC’s, I plugged this into another spreadsheet and sorted them by their correlation to Total Wins. This is the basis of my article. I’ll lay out and discuss which statistical categories tend to have the most correlation with Winning. I’ll also touch out a few surprises and areas of interest.
Disclaimer!! Correlation does not imply Causation?!
I understand that correlation does not imply causation. Just because two statistics tend to increase together does not mean that one causes the other. But I think that it is equally important to point out that when there is causation, there is a likely correlation. Looking for distinct and repetitive trends in the data will help us identify the difference.
Let’s Begin Now.
First, here is a summary that shows all of the statistics with a CC greater than +.5 or less than -.5.
|1||DEF Total Pts Against||-0.75|
|5||OFF Pass TD %||+.67|
|7||DEF Pass TD% Against||-0.67|
|9||OFF Start Field Position||+.65|
|10||TEAM Turnover Margin||+.65|
|17||DEF Sack Yards||+.51|
Many of these statistics are similar. For example, OFF Score% is nearly identical to OFF Total Points and OFF Points/Drive. To simplify the analysis, I removed these duplicates (shown with a
That brought the list down to the Final 7.
|1||DEF Total Pts Against||-0.75|
|3||OFF Pass TD %||+.67|
|4||DEF Pass TD% Against||-0.67|
|5||OFF Starting Field Position||+.65|
|6||TEAM Turnover Margin||+.65|
|7||DEF Sack Yards||+.51|
1. Defensive Points Allowed: CC = -.75
|Top 5: Def Points Allowed||Wins|
|3||San Francisco 49ers||272||12|
|4||New Orleans Saints||304||11|
|Bottom 5: Def Pts Allowed||Wins|
This is the team statistic with the highest correlation to Winning, and it is obviously a negative correlation. The more points your defense gives up the fewer games you win.
Chicago is a bit of an anomaly on this list with 8 wins, but that is mainly because they had a fairly high-powered offense in 2013 ranking 2nd in the NFL in offensive points scored.
2. Offensive Scoring Percentage: CC = +.73
(percentage of drives ending in an offensive score)
|Top 5: Scoring %||Wins|
|2||San Diego Chargers||44.9%||9|
|3||Green Bay Packers||40.9%||8.5|
|4||New England Patriots||40.8%||12|
|Bottom 5: Scoring %||Wins|
|28||Tampa Bay Buccaneers||27.1%||4|
|31||New York Giants||25.5%||7|
This is basically the opposite of the last category. As much as great defenses are related to winning, great offenses have almost equal impact. There were several other categories that were very similar to this that had similar correlation: Total Points (CC= .71) and Points/Drive (CC= .70). Essentially, we are talking about offensive efficiency. Not surprisingly, winning teams score on a higher percentage of their drives than losing teams.
Interestingly Jacksonville was ranked in the bottom 5 of both statistics so far. Ouch. They have a long way to go to become competitive… Notice Miami has not been at the top or bottom of either of these categories.
3. Offensive Pass TD %: CC = .67
(Pass TD’s / Pass Attempts)
|Top 5: Passing TD %||Wins|
|4||New Orleans Saints||6.0%||11|
|5||San Diego Chargers||5.9%||9|
|Bottom 5: Passing TD %||Wins|
|31||New York Jets||2.7%||8|
This category deals with passing efficiency. Teams that can move the ball and score efficiently through the air tend to win more games. Offensive Pass TD % was the most highly correlated of several passing statistics. Others that I won’t discuss in detail due to the similarity include QB Rating, Int’s and Yards/Pass.
Again there are some anomalies on this list. Both Baltimore & NYJ had very poor Pass TD % but came out of 2013 with 8 wins. Here are possible explanations: The Jets had a solid defense and a commitment to the run. This kept them in games and made up for their lack of firepower. The Ravens were barely in the top half of the league on defense and were in the bottom half of the league in both Passing TD’s and Rushing TD’s. Their saving grace was Justin Tucker who led the NFL in Field Goals Made.
4. Defensive Pass TD % Against: CC = -.67
(Defensive Pass TD’s Allowed/Attempts)
|Top 5: Defensive Pass TD %||Wins|
|5||San Francisco 49ers||3.2%||12|
|Bottom 5: Def Pass TD %||Wins|
|28||Green Bay Packers||5.6%||8.5|
A trend seems to be developing. Pass efficiency seems to have a very strong correlation with winning… whether it be a team’s passing attack on offense or their ability to stop the pass on defense. Both are equally correlated with winning.
Miami’s first appearance on a list indicates that the strength of their defense (and really, the entire team) was the ability to lock-down the airways to the end zone. Tennessee and Miami are both a little out of place here with 7 & 8 wins, but keep in mind both were in the bottom half of the league in offensive scoring. We’re beginning to see that a strong defense can offset a poor offense and vice-versa... enough to make you a play-off contender.
5. Starting Field Position: CC = +.65
|Top 5: Starting Field Position||Wins|
|1||Kansas City Chiefs||33.6||11|
|2||San Francisco 49ers||31.9||12|
|4||New England Patriots||30.2||12|
|Btm 5: Starting Field Position||Wins|
|29||St. Louis Rams||26.3||7|
|31||New York Jets||26.0||8|
6. Team Turnover Margin: CC = +.65
|Top 5: Turnover Margin||Wins|
|2||Kansas City Chiefs||18||11|
|5||San Francisco 49ers||12||12|
|Btm 5: Turnover Margin||Wins|
|30||New York Jets||-14||8|
|31||New York Giants||-15||7|
Both Field Position and Turnover Margin are true team statistics and are an indicator of a team’s offensive and defensive effectiveness. We always hear that the turnover differential is a key to winning, and this supports it. Another old football cliché is to say that it is a game of field position… which is also true. This isn’t ground –breaking but it does support good ol’ fashioned football logic. Hang on to the football and move the ball downfield.
7. Defensive Sack Yards: CC = +.51
|Top 5: Sack Yards||Wins|
|4||New Orleans Saints||336||11|
|5||St. Louis Rams||326||7|
|Btm 5: Sack Yards||Wins|
|28||New York Giants||214||7|
The Panthers and the Bills really dominated that category? They are way ahead of the Cardinals and the rest of the NFL.
This category supports the importance of line play, both offensive and defensive. It isn't as highly correlated as the other categories we've looked at, but it still suggests that protecting the quarterback is nearly as important as the ability of the quarterback. It’s interesting to note that Sack Yards showed a higher correlation to Winning than Total Sacks or even Sack %.
Surprises and Points of Interest
How much does the running game matter anymore?
Very few rushing statistics (offense or defense) have a significant correlation with winning. On offense the strongest was Rushing Attempts (.42). Shockingly, Yards/Rush (.07) had almost ZERO correlation with winning. On defense, again we see that Rushing Attempts Against (-.47) had the strongest negative correlation with winning (meaning that the more opponents run the ball against you, the more likely you are to lose). And again Yards/Rush Allowed (-.01) had almost no effect on how often teams won.
via www.sportstalkflorida.com (Aaron Rodger, the NFL's All-Time QB Rating Leader)
A passing efficiency league?
While almost all of the Offensive and Defensive passing efficiency statistics have a very strong correlation with winning… Defensive Pass Yards/Game had almost no impact on winning. The likely explanation is that when a team is behind they are more likely to throw the ball, which results in more pass yards but not necessarily better efficiency. Passing volume nearly as important as Passing Efficiency. Teams must be able to Pass and stop the Pass to win consistently... the running game has become secondary.
On a somewhat related note…There were just over twice as many Passing 1st Downs (6210) in the NFL as Rushing 1st Downs (3059). Simply put, most NFL teams move the chains by throwing the ball.
- I typed "Seattle" a lot in this article. The Seahawks were dominant in nearly every statistical category that mattered. Wow.
- Pass Offense & Defense were the keys to winning in the 2013 NFL. Specifically passing efficiency. It appears that teams pass the ball to get first downs, field position and points.
- The running game has become less critical to winning. Yards/Carry shows almost no effect on winning… As I stated above, Total Rushing Attempts (for or against) has the most significant impact on winning of all the rushing statistics. The conclusion I draw from this is simple, teams run the ball more when they have a lead. And it is starting to become clear that leads are built through an efficient passing game. Think about the 2008 Dolphins: efficient passer (Pennington), great turnover differential and solid pass defense. That is still a winning formula.
- Looking at what winning teams have in common begins to show us trends about how to build winning football teams. Some people refer to this as "positional value"… the idea that some positions are more valuable to the football team than others. These are the positions that finding an All-Pro can make the most direct impact on a football team’s ability to win.
- Obviously Quarterback is the keystone position.
- Below quarterbacks, we will find positions that directly affect the passing game: Pass Rushers (LB/DE/DT), Pass Catchers (WR/TE), Pass Defenders (CB/S) and Pass Blockers (OL).
- The third tier are the positions that are oriented towards running the football: Runners (RB) & Run Stuffers (Nose Tackles, Two-Down LB, In-the-box Safety).
- Not far behind are the Kicker and Punter, since they can have a direct affect on Scoring % and Starting Field Position. The argument could be made that kickers and punters have already become more important than running backs. This will become even more apparent if the Extra Point is moved out to 30 or 40+ yards.
- The 2013 Miami Dolphins were an 8-8 team. Miami’s ranking in key statistics helps explain why.
- Strengths: Pass Defense and Points Allowed
- Weaknesses: Offensive Scoring and Passing Efficiency
- Miami’s inability to score points consistently allowed their opponents to run the ball against them. Alot. Miami had the 2nd most rushing attempts against them in the entire league. Since Miami didn’t score many points teams were able to run the ball pretty much whenever they wanted. Either they were running to maintain a lead or they were running because it was a close game and Miami doesn’t have a high-powered offense that would allow them to pull away. Running the ball shortens games, which probably contributed to Miami’s defense not giving up many points. Miami’s biggest hurdle to making the leap into the playoffs appears to be improving their Offensive Passing Efficiency. So let’s hope we draft a WR, TE before a RB this year. Can anyone remember the last time that happened? (Patrick Turner & Brian Hartline in 2009)
via t0.gstatic.com (Jeff Ireland, former Miami Dolphins GM)
Finally, if you made it this far…
A Sincere Thank You! I really enjoyed doing this and I feel like I came away with a better understanding of what REALLY makes a winner. I hope you learned something too, or found a new perspective, or confirmed your suspicions. Either way, I’d like to hear everyone’s thoughts and criticisms of this article. Could I have presented information more clearly? Charts? Graphs? Additional Info? Any follow up questions?
Please let me know and I will make updates to this post to improve it’s clarity. Again, Thank You for taking the time to read and hopefully you got something out of it. Please REC if you did. I’d like to write more of these and if anyone else enjoys reading them. I’d also appreciate any suggestions from my fellow Phinsiders on other areas you would like to take a look inside the numbers.