I had an interesting discussion where I argued a point regarding the Ravens' performance last season. Technically, it's only a theory, and the more that I thought about it, the more I began to question it. Obviously, many aspects of football are arbitrary and random. Guys can wake-up and have a good or bad day on the football field for no apparent reason. This is the human factor that can never be perfectly described by any one theory and what makes football so fun to watch. That being said, many theories can help to discover and illustrate what may be happening in reality. My theory is that the Ravens' struggles versus quality opponents last season was a result, in large part, to quarterback play on both sides of the ball. This would seem obvious, but my theory is a little more specific.
Basically, I believe that Joe Flacco's quarterback play was worse toward the end of the season and the defense's play was better at the end of the season. The fall of Flacco and the rise of the defense created a unique situation that affected the entire season. Since Flacco was having less success versus opposing defenses and wasn't able to score as efficiently, pressure was put on the defense. Luckily, the defense improved, but the weaknesses in pass coverage and pressure could only be covered up. As a result, the Ravens continued to struggle versus teams with good quarterbacks because Flacco was unable to keep scoring-pace with quarterbacks who had success versus his defense. Obviously, teams with good quarterbacks and good defenses were the hardest for the Ravens to overcome. In my mind, Flacco's drop-off can be attributed to the injuries that he suffered in the Vikings game, and the defense's poor play in the beginning of the season can be attributed to the new system under Greg Mattison.
The argument was basic and based on general feelings. After I made the argument, I began to think about its plausibility. There are so many different and valid arguments that can be made against it. I watched every snap last season, and I definitely got the feeling that Flacco was "on" early in the season but seemed to slow down as the season progressed. I also felt like the defense started out weak but finished strong. Many other fans have expressed the same feelings, and the statistics for Flacco and the defense seem to agree. However, it's hard to make an argument connecting both Flacco's play and the defense's play to wins and loses. For every fact that I could point out, there is one that contradicts it, so I decided to discover the plausibility of the argument for myself.
What exactly was I trying to prove? I wanted to show that as the season progressed, Flacco slowly cooled down and the defense heated up. I also wanted to discover if there was any connection between that argument and wins and losses. The argument hinges on the fact that the Ravens never reached their full potential last season. Most teams never do, but in the case of the Ravens season, the success of the quarterbacks, both the opponents and Flacco, versus the respective defenses should help us to understand the Ravens success and failure last season.
Football is a team game, and both sides of the ball effect the game's outcome and effect each other. If Flacco's success versus an opposing defense matches the success of an opponent quarterback versus his defense, the game should be close... theoretically. If Flacco is having more success than the opposing quarterback, the Ravens should be winning. Obviously, quarterback play isn't the only factor in winning football games. Flacco only threw for thirty-four yards versus the Patriots in the playoffs, and the Ravens still won. That's why I decided not to analyze a single game or even a few games. I decided that the entire season needs to be taken into account and that I would be looking for trends.
Trend analysis is a statistical technique that uses data analysis to discover what is happening in reality. It's like reverse-engineering with numbers. The NFL is great because statistics are well documented and easily accessible. I decided that I would analyze several statistics as the 2009 season progressed and would compare Flacco to his opponents in different games. The idea behind comparing Flacco to his competition is to understand how Flacco's performance versus opponent quarterbacks' performance contributes to wins and losses over a season.
Passer rating should give me a good understanding of how much success a respective quarterback is having versus the defense he is facing. Yardage will tell me the same thing but should be checked versus the other statistics since yardage can be racked up when a team is playing from behind. Touchdowns and interceptions should be the most telling statistic in my opinion. Touchdowns and interceptions can be arbitrary, but when you consider that quarterbacks are trying their best to score touchdowns and defenses are trying to force interceptions, the ability to achieve them is an indicator of success, especially over time. Interceptions are useful when studying opposing quarterbacks since they are typically a good indicator of pressure and coverage by the defense. The more games analyzed, the more accurate the results will be, especially with regard to touchdowns and interceptions.
The trend analysis provided intriguing insight into the Ravens season. From the "QB Rating" graph, it's clear that both the opponent quartbacks' rating and Flacco's rating fell as the season progressed. This is exactly what I expected if my argument is true. If Flacco's rating is declining as much as the opposing quarterbacks', that means his decline is happening at the same time as the defense's emergence. The fact that they're parallel is also a huge support for my theory. This means that the defense was increasing in success versus opposing offenses at nearly the same rate as Flacco was declining in success versus opposing defenses. The "QB Rating" trend line gives us a good overall view of the season in terms of the success of the quarterbacks versus the respective defenses.
The "Yards" and the "Touchdown" graphs were also what I expected. As the season progressed, Flacco threw for less yardage and touchdowns. The opposing quarterback also declined but at a much slower rate. In fact, the slope of Flacco's "Touchdowns" trend line causes it to overtake the opposing quarterbacks' trend line. The fact that Flacco was finding the end-zone at a drastically different rate than the opposing quarterbacks means that even though he was competitive statistically (which is evident due to the parallel "QB Rating" trend lines), he was not as efficient at scoring. If Flacco is capable of competing statistically but not scoring consistently, what does that tell us? Obviously, the offense wants to score on every play, so this tells us that Flacco can move the ball but not score when needed. Now, this isn't saying that Flacco is incapable. In fact, the analysis shows that Flacco was more efficient than the competition early in the season but wasn't as efficient as the competition later in season. This is evident because Flacco's "Touchdowns" and "Yards" trend lines start higher and finish lower and therefore, have a greater negative decline. The opponent quarterbacks' "Touchdowns" trend line is almost flat but does decline slightly. Therefore, Flacco became increasingly more inefficient while the competition remained at an even scoring efficiency as the season progressed. This weakness would obviously be magnified versus a quarterback that is having success versus Flacco's defense and scoring when needed.
The "Interceptions" graph is an indicator of the defenses' success. From the graph, it's clear to see the dramatic rise in the opposing quarterbacks' trend line as the season progressed. This basically shows that the defense was playing more aggressive as the season progressed. A bad defense can force interceptions, but it takes an aggressive and effective defense to force a trend line up like the graph above. Flacco's "Interceptions" trend line is almost flat and may decline a little. When a quarterback's play declines, they typically throw more interceptions. In other words, interceptions typically increase as a quarterbacks statistics decrease. With Flacco, every area of his game declined, but he didn't really have a huge increase in interceptions. This tells me that the Ravens were playing it smart with Flacco and protecting him with easy, max-protect game plans. This is technically a subjective point of view because it can just as easily be argued that Flacco's decrease in numbers without an increase in interceptions were a result of a run-first focus in the running game. That argument has weight, but if Flacco is throwing for career highs and the offense is effectively running and passing, why would they change anything? That's what leads me to believe that the coaches were protecting Flacco later in the season after the injury, presumably.
I'd love to be able to say that my analysis proves my argument and subsequent theory beyond any doubt. However, I don't think that I can objectively say that. There are just too many opposing arguments that also have weight. However, I have shown that Flacco was less effective versus opposing defenses as the season progressed and the defense was more effective versus opponent quarterbacks as the season progressed, and if you connect that fact with the fact that the Ravens continued to win games throughout the season, the conclusion is that the Ravens were winning games with their offense early and with their defense late.
At the very least, the analysis shows that my argument is very plausible if not likely. It doesn't necessarily prove anything about the Ravens success and failure during the season since there are so many other factors, but with trend analysis, the idea really wasn't to prove the theory. The idea is to prove the plausibly and likelihood of it, and in that regard, I believe that the analysis has been very successful. Using trend analysis, it is possible to construct a model which is independent of anything known about the physics of a process of an incompletely understood physical system. Therefore, this type of analysis can circumvent the billions of other factors to completely describe any situation given an infinite time frame. Since we can't use an infinite time-frame and only have relevant data from a single season, I cannot say that the analysis completely describes the Ravens season. However, it does give us an useful model from which an accurate estimation of reality can be made.
The fact that Flacco's "QB Rating" trend line is parallel with the opponent quarterbacks' trend line tells us that the average games were competitive between the Ravens and their opponents. In other words, Flacco's decline didn't affect the competitiveness of the Ravens because the defense picked up the slack. The fact that Flacco's "Touchdowns" trend line significantly decreases and the opponent quarterbacks' trend line stays relatively steady tells us that Flacco's efficiency severely decreased as the season progressed. The decreased efficiency was a likely reason for the Ravens struggles versus quality opponents. Even though Flacco and the Ravens were competitive, they weren't able to consistently put games away when it counted, and the analysis shows that. Using the analysis, we have effectively seen that my original argument aligns with statistical data and ultimately, is a very good approximation of what happened during the Ravens' 2009 season.