As many of you may know, and some of you may not, I'm a guy who likes stats. I like digging through them, looking for patterns, looking to see what seems to leap out, and then open up discussions about what that might mean. I also am the type of person who, when he hears sweeping generalizations about a player or a team, will go to the stats pages for that team or player to see if the generalization is supported by the statistical record. Sometimes it is, sometimes it's not.
As someone who enjoys looking at stats, and how they can be used to better understand the game of football, and how they can be used to compare on team to another, I have found myself irritated over the years by the NFL's insistence upon ranking offenses and defenses simply by the number of yards they pile up or allow. After all, it seems to me that the purpose of the offense is to put points on the board, and the defense is tasked with the job of limiting the number of points the other team puts on the board. I have followed a number of discussions about how offenses and defenses should be ranked: yards, points, wins, losses, take-aways, give-aways and a large number of combination's of those different aspects of the game. So, this season, I'm going to engage in a sort of grand experiment to see if I can find a different way to look -- statistically -- at offenses and defenses.
Join me after the jump.
Now, I have to admit that I can see the logic of the NFL's using yards as the measuring stick for offenses and defenses. It's simple, easy to track and calculate. Plus, there's an underlying assumption -- which may, or may not be valid -- that the more yards a team piles up, the more likely they are to score. Personally, I found that concept called into question by Denver's 2008 offensive performance. If you will recall, Denver amassed a very impressive 6333 yards in total offense that year. This ranked 2nd in the NFL. However, that same offense only scored 370 points. This ranked 16th in the league. So was the 2008 offense a top five offense, or just an average one? The issue was further called into question by the Broncos 2009 defensive performance. In 2009, the Denver defense held opponents to 5040 yards (7th in the league). Yet, that same defense gave up 324 points (or 12th in the league). So was the 2009 defense a top ten one, or a barely above average one? Discussions on these topics have gone around and around, and I'm inclined to believe that we will never find a definitive answer, nor even come to any kind of consensus on them.What I'm planning to do with this series is look at the concept of Yards Per Point. If I'm not mistaken, this is not a new concept, nor is it original with me. The concept in and of itself is simple: take the total net yards earned by an offense or surrendered by a defense, and divide it by the number of points scored by the offense or given up by the defense. This will give us the average number of yards an offense has had to travel for each point that they have scored. It will also give us the avearge number of yards a defense has forced opposing teams to travel for each point scored against that defense. The hypothesis being examined here is three-fold:
(1)The fewer yards an offense has to travel to score, the more points they will put on the board,
(2)The farther the defense forces the opposing team to travel, the fewer points the opponent will be able to put on the board, and
(3)The better the averages in #1 & #2, the more likely it is that the team will win.
Whether or not the results of the season will bear this out, remains to be seen. Each week, I'll be updating the yards/point data for each of the 32 NFL teams and inviting a discussion of that data. I will also be looking at one or more teams from the past -- most likely the past Super Bowl participants -- to see what kind of yards/point data may be found in their performances.
Before we get into the data itself, I would like to acknowledge that the data is restricted to yards/points either generated by the offensive unit or surrendered by the defensive unit. Return yards are not included in the yardage, nor are points scored by the defense or by kick/punt returns. This would mean that, for Week #1, six teams will show fewer points than the final score:
(1)Miami and Carolina each scored 2 points on a safety.
(2)New England and Seattle each scored 6 points on interception returns (PATs count as offensive points).
(3)Kansas City scored 6 points on a punt return.
(4)Washington scored 6 points on a fumble return.
This week, I am going to focus simply on the offensive performances of each team, as a way of introducing the method I will be using.
The Traditional Way of Ranking an Offense
Team |
Yards |
W/L |
Team |
Yards |
W/L |
|
Chicago |
463 |
W |
Jacksonville |
299 |
W |
|
Indianapolis |
463 |
L |
Green Bay |
299 |
W |
|
Cincinnati |
428 |
L |
Miami |
296 |
W |
|
San Diego |
389 |
L |
Atlanta |
295 |
L |
|
Dallas |
380 |
L |
Tampa Bay |
288 |
W |
|
Arizona |
378 |
W |
Oakland |
286 |
L |
|
New England |
376 |
W |
Baltimore |
282 |
W |
|
NY Giants |
376 |
W |
San Francisco |
263 |
L |
|
Denver |
363 |
L |
Minnesota |
253 |
L |
|
Houston |
355 |
W |
Washington |
250 |
W |
|
Pittsburgh |
354 |
W |
Seattle |
242 |
W |
|
Tennessee |
345 |
W |
Carolina |
237 |
L |
|
Cleveland |
340 |
L |
Kansas City |
197 |
W |
|
St. Louis |
325 |
L |
NY Jets |
176 |
L |
|
Philadelphia |
321 |
L |
Detroit |
168 |
L |
|
New Orleans |
308 |
W |
Buffalo |
166 |
L |
I found it interesting that in this traditional method of ranking offenses, Denver had the ninth ranked offense. I was not left with the impression that the Broncos had a top ten offense after watching the game against Jacksonsville. Further, I found it somewhat surprising that four of the top five offenses all lost. However, five of the top ten teams did win. Four of the bottom five teams lost, as did six of the bottom ten.
One Alternative to the Traditional Way (Points)
Team |
Points |
W/L |
Team |
Points |
W/L |
|
Tennessee |
38 |
W |
Kansas City |
15 |
W |
|
Houston |
34 |
W |
New Orleans |
14 |
W |
|
New England |
32 |
W |
Detroit |
14 |
L |
|
NY Giants |
31 |
W |
Cleveland |
14 |
L |
|
Green Bay |
27 |
W |
San Diego |
14 |
L |
|
Seattle |
25 |
W |
Miami |
13 |
W |
|
Indianapolis |
24 |
L |
Oakland |
13 |
L |
|
CIncinnati |
24 |
L |
St. Louis |
13 |
L |
|
Jacksonville |
24 |
W |
Buffalo |
10 |
L |
|
Philadelphia |
20 |
L |
Baltimore |
10 |
W |
|
Chicago |
19 |
W |
Minnesota |
9 |
L |
|
Carolina |
18 |
L |
Atlanta |
9 |
L |
|
Denver |
17 |
L |
NY Jets |
9 |
L |
|
Tampa Bay |
17 |
W |
Dallas |
7 |
L |
|
Arizona |
17 |
W |
Washington |
7 |
W |
|
Pittsburgh |
15 |
W |
San Francisco |
6 |
L |
When looking at points, we find that all of the top five teams won, and that seven of the top ten teams won. We can also see that four of the bottom five teams lost, as well as eight of the bottom ten. It may be noticed that all of the teams that scored 25 or more points won. All of the other point totals were scattered as to whether or not a given score would bring victory. It would appear that looking at an offense in terms of points serves as a better indicator of the likelihood of winning vs losing. When we add the defensive analysis, we'll see if this is true, or more of a reflection of the type of defense against which the given offense scored.
The Yards/Point Approach
Team |
Yds/Point |
W/L |
Team |
Yds/Point |
W/L |
|
Tennessee |
9.1 |
W |
Denver |
21.4 |
L |
|
Seattle |
9.7 |
W |
New Orleans |
22.0 |
W |
|
Houston |
10.4 |
W |
Oakland |
22.0 |
L |
|
Green Bay |
11.1 |
W |
Arizona |
22.2 |
W |
|
New England |
11.8 |
W |
Miami |
22.8 |
W |
|
Detroit |
12.0 |
L |
Pittsburgh |
23.6 |
W |
|
NY Giants |
12.1 |
W |
Cleveland |
24.3 |
L |
|
Jacksonville |
12.5 |
W |
Chicago |
24.4 |
W |
|
Kansas City |
13.1 |
W |
St. Louis |
25.0 |
L |
|
Carolina |
14.8 |
L |
San Diego |
27.8 |
L |
|
Philadelphia |
16.1 |
L |
Minnesota |
28.1 |
L |
|
Buffalo |
16.1 |
L |
Baltimore |
28.2 |
W |
|
Tampa Bay |
16.9 |
W |
Atlanta |
32.7 |
L |
|
Cincinnati |
17.8 |
L |
Washington |
35.7 |
W |
|
Indianapolis |
19.3 |
L |
San Francisco |
43.8 |
L |
|
NY Jets |
19.6 |
L |
Dallas |
54.3 |
L |
In this table, we find that four of the top five teams all won, and that eight of the top ten teams won. Nine of the top sixteen teams won. Three of the bottom five teams lost, along with seven of the bottom ten posting a losing record.
The Pairings From Week 1
Minnesota |
28.1 |
Denver |
21.4 |
Cleveland |
24.3 |
Arizona |
22.2 | |||
New Orleans |
22.0 |
Jacksonville |
12.5 |
Tampa Bay |
16.9 |
St. Louis |
25.0 |
|||
Miami |
22.8 |
Cincinnati |
17.8 |
Oakland |
22.0 |
Dallas |
54.3 |
|||
Buffalo |
16.1 |
New England |
11.8 |
Tennessee |
9.1 |
Washington |
35.7 |
|||
Detroit |
12.0 |
Carolina |
14.8 |
Green Bay |
11.1 |
Baltimore |
28.2 |
|||
Chicago |
24.4 |
NY Giants |
12.1 |
Philadelphia |
16.1 |
NY Jets |
19.6 |
|||
Indianapolis |
19.3 |
Atlanta |
32.7 |
San Francisco |
43.8 |
San Diego |
27.8 |
|||
Houston |
10.4 |
Pittsburgh |
23.6 |
Seattle |
9.7 |
Kansas City |
13.1 |
(winning team is in bold)
We can see from this pairing table that in thirteen out of the sixteen games, the offense that had to travel fewer yards to score each point won. The lone exceptions were Miami over Buffalo, Chicago over Detroit, and Baltimore over the NY Jets.
It must be acknowledged that a single game cannot be considered to be a definitive sample. This is why I will be compiling data throughout the 2010 season, and bringing in examples from previous years. One of the ways I am going to use this data is in making my weekly picks for the MHR staff prognostications. We'll see if it helps me at all. I look forward to hearing other people's thoughts on the subject.
Next week, we will not only update the data, but add in a look at the defensive side of the ball.