Many football analysts and fans have different thoughts on what teams and players are better than others. Often, their judgment is based on recent statistics to drive their argument. Statistics are the premier method of analyzing the NFL, but they aren’t perfect. Some are better than others, and it is crucial to know which ones to use to build an argument successfully. This piece will cover specifically quarterback stats, and there will be a few examples of which ones are good and which are not.
Good Statistics to Use
Yards Per Attempt (Y/A)
A lot of people bring up yards or completion percentages when discussing quarterback play. However, each of these is useless without the other. A quarterback who throws the ball a lot will naturally have more yards than most others. And a quarterback who throws many short passes will have a very high completion percentage because of the types of throws they attempt. Luckily, Y/A combines these two numbers into one. This number is very useful because it gives the average yards gained for a quarterback’s pass attempts. It is also simple and easy for fans to understand, which can be beneficial for newer fans. Adjusted net yards per attempt (ANY/A) tries to improve on Y/A by weighting touchdowns, interceptions, and other events. However, it comes out to be a worse version of EPA/play.
Expect Points Added Per Play (EPA/Play)
EPA/play is the closest thing out there to an all-in-one quarterback statistic. It is founded on the concept of expected points (EP), which is a bit complicated. Essentially, for every down and distance at any given yard line, there is an EP value that represents how many points the team with the ball is likely to score on their current drive. The difference in EP from one play to the next gives EPA. Plays like turnovers or plays that result in a loss of yards will yield negative EPA. EPA/play has a very high correlation with wins, especially compared to most other QB stats. It is arguably the best single metric that currently exists for QB analysis. That being said, it is a difficult idea for some people to understand and is relatively new, so not everyone will recognize it.
Bad Statistics to Use
TD:INT ratio is something that is referenced extremely commonly when comparing quarterbacks. However, it is not a useful statistic. While touchdowns and interceptions are extremely high-impact plays, focusing on solely these plays is not a good idea. These plays make up such a small portion of quarterbacks’ throws that they do not give a complete analysis of the QB. Stats that cover all of a quarterback’s passes do a much better job of painting the full picture.
The quarterback is the most important player on the field during any pass play. He decides what to do with the football, and a perfect decision coupled with a perfect throw will beat any defense. However, there are 10 other players on offense who can make these jobs easier for the quarterback. And, there are 11 more players on defense who can make it so that the quarterback doesn’t have to score 60 points to secure a win. Football is a team game. And pinning the win-loss record on the quarterback is the wrong statistic to use when trying to analyze them. The best quarterbacks give their teams the best chance to win, but if the rest of the team is horrible, it won’t always be enough.
Knowledge for the Future
There are many more QB statistics that exist beyond the ones listed in this article. Some are good, some are bad. There is a general guideline to keep in mind when weighing whether a stat is useful or not: a good metric will include all (or almost all, with a select few plays removed) of a quarterback’s throws. These are most strongly correlated with winning, which the best quarterbacks give their teams the highest chance of doing.
“From Our Haus to Yours“