TheCD
Rookie
There has been a lot of talk about just how effective Kubiak’s offense is. It’s obvious to all of us watching that it gets off to slow starts, or falters in the second half. I spent this summer compiling data and analyzing statistics to see just how effective this offense has been. This is a summary of my findings. I will avoid inserting my opinions until the very end. My hope is that this data will help to spark a logical debate as to the offenses effectiveness.
For those of you who are not mathematically inclined or just don’t care to read about how the methods used to determine these findings, please feel free to skip the final post where I summarize my findings in the conclusion.
Goals:
I decided to begin this research by becoming as narrow and specific as possible in my goals. My goal is simply the following: How effective is Houston’s offense against opposing defenses in terms of points scored. Bill Belichick famously said that stats are for losers. However, he then went on to say that the most meaningful stats are games won, followed by points scored. I decided that the best way of determining how good the offense has been was to remove all factors other than points scored by the offense, and weighed this against the overall defensive ranking against that team that season (I determined defensive rankings also by points allowed, rather than yards).
Methods:
My first decision was to narrow the offenses performance down to only points scored and weigh that against the defenses points per game allowed in a season. My second decision was to remove all defensive/special teams points scored, though I decided to allow field goals to count towards my numbers. I then had to choose how many seasons to go back for reference to provide a good pool of numbers. I decided to chart all the way back to the 2007 season, as this was our first season with Matt, and what could essentially be called the “modern” era of Kubiak’s offense.
This research also completely ignores wins and losses, focusing solely on offensive performance.
For those statistically/mathematically inclined, it should come as no surprise that the primary formula used to determine these findings was Standard Deviation (S.D.). I apologize for not providing the specific formula, but cannot do so with a standard keyboard.
Briefly, the S.D. for all games from 2007 to 2010 is:
Sa = 3.6
Sb = 4.4
What this means is that any game in which we scored more than 3.6 offensive points than that defense allowed on average in a season, our offense performed better than average. Likewise, any offensive points that resulted in 4.4 points less than what the defense averaged in a season was a poor showing by our team. Any games that fell between 3.6 and 4.4 was an average showing by our team.
For those of you not used to S.D., here is a brief summary of what I did and why I used it: S.D. is a method for understanding how unusual a score is in relation to other scores. What this allowed me to do was chart all of the Texans’ offensive points against defensive rankings (how many points that defense averaged per game) and determine if the Texans performed above, at, or below the average points scored against that particular defense. For example, in 2009 the New York Jets had the #1 defense in terms of points allowed and averaged 14.8 p.p.g. on defense. Our offense scored no points in this game. Since Houston’s S.D. for points scored below average is 4.4, this means that we scored 10.4 points fewer than other teams in the season. This is a very large S.D. and (obviously) a horrendous performance by our team.
The benefit of using S.D. is that you can easily determine what scores do not fall within what defenses would normally give up throughout the season. Any scores higher than the S.D. on either scale means that the offense performed either very well, or very poorly.
All data comes directly from NFL.com.
For those of you who are not mathematically inclined or just don’t care to read about how the methods used to determine these findings, please feel free to skip the final post where I summarize my findings in the conclusion.
Goals:
I decided to begin this research by becoming as narrow and specific as possible in my goals. My goal is simply the following: How effective is Houston’s offense against opposing defenses in terms of points scored. Bill Belichick famously said that stats are for losers. However, he then went on to say that the most meaningful stats are games won, followed by points scored. I decided that the best way of determining how good the offense has been was to remove all factors other than points scored by the offense, and weighed this against the overall defensive ranking against that team that season (I determined defensive rankings also by points allowed, rather than yards).
Methods:
My first decision was to narrow the offenses performance down to only points scored and weigh that against the defenses points per game allowed in a season. My second decision was to remove all defensive/special teams points scored, though I decided to allow field goals to count towards my numbers. I then had to choose how many seasons to go back for reference to provide a good pool of numbers. I decided to chart all the way back to the 2007 season, as this was our first season with Matt, and what could essentially be called the “modern” era of Kubiak’s offense.
This research also completely ignores wins and losses, focusing solely on offensive performance.
For those statistically/mathematically inclined, it should come as no surprise that the primary formula used to determine these findings was Standard Deviation (S.D.). I apologize for not providing the specific formula, but cannot do so with a standard keyboard.
Briefly, the S.D. for all games from 2007 to 2010 is:
Sa = 3.6
Sb = 4.4
What this means is that any game in which we scored more than 3.6 offensive points than that defense allowed on average in a season, our offense performed better than average. Likewise, any offensive points that resulted in 4.4 points less than what the defense averaged in a season was a poor showing by our team. Any games that fell between 3.6 and 4.4 was an average showing by our team.
For those of you not used to S.D., here is a brief summary of what I did and why I used it: S.D. is a method for understanding how unusual a score is in relation to other scores. What this allowed me to do was chart all of the Texans’ offensive points against defensive rankings (how many points that defense averaged per game) and determine if the Texans performed above, at, or below the average points scored against that particular defense. For example, in 2009 the New York Jets had the #1 defense in terms of points allowed and averaged 14.8 p.p.g. on defense. Our offense scored no points in this game. Since Houston’s S.D. for points scored below average is 4.4, this means that we scored 10.4 points fewer than other teams in the season. This is a very large S.D. and (obviously) a horrendous performance by our team.
The benefit of using S.D. is that you can easily determine what scores do not fall within what defenses would normally give up throughout the season. Any scores higher than the S.D. on either scale means that the offense performed either very well, or very poorly.
All data comes directly from NFL.com.