The calendar might have changed, but championship mettle still is the name of the game for this week’s combatant. Sitting atop their division, this team is a force to be reckoned with. They’ve already thoroughly embarrassed two of their five opponents, and have been improving steadily as the season near’s it’s midpoint. A crushing running game is complimented by a capable quarterback that can do whatever it takes to win.
How in the world are the Seattle Seahawks going to be able to contain the Dallas Cowboys?
This can’t be life.
Dallas has reeled off four victories in a row for the first time since Weeks 9-12 of the 2011 season. They will try to push the streak to five, something they haven’t accomplished since doing so twice during the 2007 campaign.
We all know the infamous take originally attributed to Henry Du Pré Labouchère, "There are three kinds of lies: lies, damned lies, and statistics." In sports, though, statistics are how we measure performance beyond just the wins and losses. In football, where sample sizes are so small, and the mean so cluttered, it behooves us to try to analyze things that have a higher correlation to winning than previous wins.
Fortunately, over the past several years, many statisticians a million times smarter than the author of this piece have spent countless hours devising formulas that they use to judge the quality of teams beyond the normal statistical measures. These advanced metrics allow us a glimpse into which teams are winning based on quality play, which ones are winning because of inept opponents and which ones have a horseshoe crammed up their chimneys.
So, let’s take a look at some of the statistical explanations for what kinds of team the 2014 Dallas Cowboys and 2014 Seattle Seahawks have proven to be, and what might transpire on Sunday.
|Off Yds / G||Off. Pts / G||Def Yds / Gm||Def Pts / Gm|
|Off Yds / G||Off. Pts / G||Def Yds / Gm||Def Pts / Gm|
We all know that it is a passing league, while running the football undoubtedly has an effect on the passing game, and more importantly how the opposition defends it. In the end, though, passing success is what leads to team success.
There are generally two stats that are easily calculated that seem to have a high correlation to predicting future success; Passer Rating differential and Adjusted Net Yards Per Passing Attempt differential. Both are calculated by simply finding the difference between what your offense achieves in each category and taking away what your defense gives up. Teams with the larger differentials are more likely to have future success (read: wins) than those teams with smaller, or negative differentials.
As my good friend and Stats Sensei One.Cool.Customer over at Blogging The Boys can attest to right here, these stats have shown to be the most closely correlated stat to winning and losing outside of scoring differential. (I think we all know why that one ranks so highly.) This is must read stuff.
Here’s a snippet from last week:
The graph illustrates a fairly simple truth: the higher your passer rating differential, the higher the number of wins a team had last year. Conversely, the teams with the lowest number of wins had some of the worst passer rating differentials (PRD).
In statistics this relationship between wins and PRD is called a correlation. The strength of the correlation is measured by the "correlation coefficient". The correlation coefficient measures the relationship between two variables. This coefficient is often referred to as "r²" and is expressed as a number between 1 and -1. The closer the r² number is to 1 or -1, the stronger the relationship between the two variables. The closer it is to zero, the weaker the relationship.
R² between PRD and wins in 2013 was 0.71, an astonishingly high correlation. It is almost as good as scoring differential (0.89) at predicting the W/L record of an NFL team.
|Rk||Tm||G||Off Passer Rating||Def Passer Rating||Differential|
|1||Green Bay Packers||5||110.6||70.5||40.1|
|3||San Diego Chargers||5||116.3||84.2||32.1|
|5||New York Giants||5||95.6||76||19.6|
|8||San Francisco 49ers||5||90.7||75.7||15|
|12||New England Patriots||5||88.5||82||6.5|
|21||St. Louis Rams||4||92.6||98.5||-5.9|
|23||Kansas City Chiefs||5||88.8||97||-8.2|
|24||New Orleans Saints||5||91.8||103.2||-11.4|
|28||Tampa Bay Buccaneers||5||78.2||103.9||-25.7|
|32||New York Jets||5||66.5||107.5||-41|
Dallas has made major strides here. Two weeks ago, they were ranked 22nd, last week 15th. Now they have moved up to 11th in the league. The Seahawks rank 7th, as they are one of only a handful of teams that have a Passer Rating above 100 on the season.
Adjusted Net Yards Per Pass Attempt, or ANY/A, comes in third when it comes to correlating to victory with an R2 coefficient of 0.69. What exactly is ANY/A?
ANY/A: Adjusted net passing yards per attempt = (Passing Yards + (Passing TDs)*20 - (INTs thrown)*45 - Sack Yards) / (Passing Attempts + Sacks)It’s basically a formula that takes into account sacks and interceptions in qualifying how efficient a team is in the passing game. Here’s how the team’s square up in this metric.
|Rk||Tm||G||ANY/A For||ANY/A Allowed||Differential|
|2||San Diego Chargers||5||9||5.3||3.7|
|3||Green Bay Packers||5||7.6||4.5||3.1|
|8||San Francisco 49ers||5||6.1||5||1.1|
|10||New England Patriots||5||6||5.2||0.8|
|12||New York Giants||5||6.3||5.8||0.5|
|20||St. Louis Rams||4||6.4||7||-0.6|
|25||Kansas City Chiefs||5||5.7||6.7||-1|
|27||New Orleans Saints||5||6.5||8.1||-1.6|
|29||Tampa Bay Buccaneers||5||5.1||7.5||-2.4|
|31||New York Jets||5||4.1||7.2||-3.1|
Dallas continues to move on up the rankings, and finally gets to a positive ANY/A differential of +0.3, ranking them 13th in the league. The difference between them and Seattle is minimal, indicating this should be a much closer game than the spread (currently -8) says it should be.
DVOA is a metric, devised by Football Outsiders, which measures success on each play as compared to league average based on "a number of variables including down, distance, location on field, current score gap, quarter, and opponent quality.” The metric is outputted in “percentage points better than average”.
|COWBOYS||Offense DVOA||Defense DVOA||ST DVOA||Total DVOA|
|SEAHAWKS||Offense DVOA||Defense DVOA||ST DVOA||Total DVOA|
DVOA clearly has the Seahawks as the stronger team, across the board. The widest margin is easily that between the Seattle defense and the Cowboys, a total of over 19% points between the two
Expected Points Added (EPA)
The value of a football play has traditionally been measured in yards gained. Unfortunately, yards is a flawed measure because not all yards are equal. For example, a 4-yard gain on 3rd down and 3 is much more valuable than a 4-yard gain on 3rd and 8. Any measure of success must consider the down and distance situation.
Suppose the offense has a 1st and 10 at midfield. This situation is worth +2.0 EP. A 5-yard gain would set up a 2nd and 5 from the 45, which corresponds to a +2.1 EP. Therefore, that 5-yard gain in that particular situation represents a +0.1 gain in EP. This gain is called Expected Points Added (EPA). Likewise, a 5-yard loss on 1st down at midfield would create a 2nd and 15 from the offense’s own 45. That situation is worth +1.2 EP, representing a net difference of -0.8 EPA.
Basically what your looking at in the above interactive chart is that teams with Good Offenses and Good Defenses exist in the upper right quadrant. The further to the right, the better your offense, and the further “north” the better your defense.
Here we see a much closer relationship between the teams, similar to what the differential metrics showed. Seattle’s advantage in defensive efficiency is slightly larger than the Cowboys advantage in offensive efficiency; but it is close. Hover over each team in the interactive table above from AdvancedFootballAnalytics.com. For more information on this metric and it’s friend, Win Probability Added (WPA), I strongly recommend heading over to their website and checking out the glossary.
Success Rate simply determines the percentage of plays that lead to a positive EPA. In other words, the Cowboys “averageness” isn’t a result of being average play by play, it’s that they make crushing mistakes, also known as turnovers.Note: because of the delay in posting the Notebook, Thursday night’s games are included in the rankings for Indy and Houston, skewing things a bit.
|COWBOYS||Offense EPA||Success Rate||Run EPA||Pass EPA|
|SEAHAWKS||Offense EPA||Success Rate||Run EPA||Pass EPA|
What you are seeing here is how well each team is doing on a play by play basis by increasing their chances to score. One thing that jumps out immediately is that despite having the league’s leading rusher, and by a large margin, Dallas is barely Top 10 in Run EPA. The explanation is simple. Standard yardage stats don’t take into account DeMarco Murray’s fumbling habits, but Run EPA clearly does. That’s how detrimental turning the ball over can be, especially when one is returned for a touchdown like it was against San Francisco. Even moreso than any Romo interception, these fumbles are killing the Cowboys scoring outlook on the games. Without them, Dallas would likely lead the league in EPA
The Seahawks clearly aren’t getting what they want out of their passing game, but the combination of Lynch and Wilson is dominating the league.
|COWBOYS||Defense EPA||Success Rate||Run EPA||Pass EPA|
|SEAHAWKS||Defense EPA||Success Rate||Run EPA||Pass EPA|
Dallas' numbers dropped for a second consecutive week, indicating that there is still plenty of work to be done. The rushing defense in particular, took a beating and it directly coincided with Rolando McClain leaving the field. Houston scored all 17 points with him sidelined and it’s no surprise the D parted like the Red Sea with him out of the lineup.
The fun stat here? Dallas’ passing game D has a better ranking than Seattle’s does. Combined with the fact that the Cowboys passing O is far superior to the Seahawks and you can see the outline for the upset.
Dallas appears to have it’s work cut out for it this coming Sunday. Although Seattle isn’t the juggernaut they had become by the end of last season, they are a far more formidable foe than Dallas has faced this year, and are playing in the ultimate home field. If Dallas is able to somehow neutralize the running threat of Russell Wilson, there is a chance that their own running game can lead to passing success. The Seahawks have yet to face a rushing attack that resembles the Cowboys, yet they’ve still been average against the pass. What does that defense look like when they have to commit men to the box? We’ll find out on Sunday.