Passer Rating, PFF, Stats, and Context

Passer Rating’s Flaws

The formula for passer rating is a little complex, but it is based on just for influencing factors. Passer rating has its flaws because it does not account for sacks, yards after the catch, or any rushing stats. For a contextual look at a quarterback‘s performance, many use QBR, developed in 2006 by ESPN.

While many people use passer rating to compare between era or compare to players within the same era, the purpose of passer rating is not for comparisons. As with many stats, the purpose of passer rating is to get a generalization of good quarterback play and bad quarterback play. For example, every individual perfect passer rating game is not better than a performance such as Peyton Manning’s seven touchdown game or better than particularly outstanding performances in the playoffs. However, the group of 158.3 games, on average, is better than a set of other passer rating games.

This logic can carry over to other statistics, so not every individual season of 1,000 rushing yards is better than a season of 800 rushing yards because it does not include any receiving stats or efficiency on the number. In general, it is more productive to have a 1,000-yard rusher than an 800-yard rusher.

The Importance of Context

When people look at stats, they often look at stats in a vacuum. While stats are fantastic, and they often can tell a story of what happened and what did not happen, stats must be taken with a slight grain of salt based on the context of the situation. For example, it is significantly more impressive to get 500 yards of offense against a great defense as opposed to 550 yards of offense against a bad defense that is playing the bench players in the second half. However, in general, a 550-yard performance is better than a 500-yard performance.

From my perspective, stats need to take a more generalized approach. While we like to gawk at individual numbers and individual accolades, everything should have some context within it. Context can both help and hurt a statistical argument. In general, it improves the quality of the argument because you know not only what you’re discussing but why it’s important to discuss.

Is passer rating flawed?

Passer rating certainly has some flaws, but it would be difficult to fix the flaws without distorting the previous 100 years of NFL history. Passer rating was made for a different time, and the modern era of efficiency has made passer rating more of a novelty stat than anything that should have meaning behind it. When passer rating was introduced, the league average was just a 66 rating. In recent years, the average rating has jumped to nearly 90.

In the modern era, more quarterbacks completed passes at a higher percentage and quarterbacks throw significantly fewer interceptions. Mathematicians should create an updated passer rating metric. However, any adjustments with the system would compromise the careers of quarterbacks in the past. The updated metric would not be compatible over the last few decades of seasons, but the metric would help the modern era.

Speaking of context-based statistics…. Hi, PFF.

Pro Football Focus is a fantastic resource for football fans. It certainly has its shortcomings, especially in the player grading, but in general, PFF is a rabbit hole of information and statistical heaven.

PFF has a problem with player grading. Due to player grading being subjective and liable for error, it can be difficult to take PFF player grades seriously. However, when you look at a larger sample size of PFF player grades, trends do occur. If you look at the elite players and compare them to average, the elite players will have elite grades more often. For every outlier in a section, most other players would be elite.

Conclusion

The system of applying numbers to a sport can be inconsistent. Derived numbers from the sport can be misleading. At the same time, mathematical interpretations can also be misleading. Even PFF’s system of detailed analysis and numerical data can be misleading. Each system has both its flaws and its positive characteristics.

By themselves, stats can show a lot of information. Without context, stats can often be inflated or deflated and not show the entire story. Passer rating is one example of stats being useful, while also having the downside of not having any context.

Be sure to follow me @MrSplashMan19 on Twitter for more NFL thoughts and statistical questions. Also, be sure to check out the Overtime Heroics Forums page to join me and many others in some good discussion. If you are interested in purchasing some Overtime Heroics Apparel, head over to our shop to rock some gear!