Adam Steele is back for another guest post. You can view all of Adam’s posts here. As always, we thank him for contributing.
Positive Yards Per Attempt: 2017 Update
If I could only share one thing from my time doing football analytics, it would be the following principle: Positive plays carry more weight than negative plays in determining the winner of a football game. I’ve already written a couple of articles on this subject and hope to further the cause with this update.
Overview
For those of you who don’t feel like reading the previous two posts, I’ll give you the basic gist. Since passing has a far greater impact on winning than running, I’ve focused my research on quarterbacks, but the principle applies to the entire offense (defense, not so sure). Despite everyone constantly harping on turnover avoidance, a potent passing offense is usually able to overcome giveaways. Conversely, avoiding turnovers is normally not enough to overcome a weak passing game. Furthermore, turnovers are highly random and situation dependent, so it follows that turnovers are a very poor method of gauging quarterback performance. Even though sacks are largely the quarterback’s fault, they are also very context dependent and only contribute a small amount in determining game outcomes. More importantly, the majority of signal callers trade sacks for interceptions or vice versa, so it’s no really fair to include one but not the other.
Here is my new formula for Positive Yards Per Attempt (PY/A):
(Air Yards + YAC / 2 + Passing First Downs x 10 + Passing Touchdowns x 10) / Pass Attempts
This metric captures the frequency and magnitude of a quarterback’s positive plays while ignoring the negative ones (sacks and interceptions). I only give half credit for YAC because I think it’s more a function of the receivers and system than the QB, and I bump the traditional first down bonus up to 10 yards (it’s usually nine yards) because I consider first downs very important and I prefer the nice, even number. Also, touchdowns are still worth 20 yards, as usual: but all touchdowns are first downs, so the bonus here needs to be lowered to only 10 yards.
Results
I have the necessary data to calculate PY/A back to 1992. Using the standard cutoff of 224 attempts to qualify, this sample yields exactly 800 quarterback seasons. The table is default sorted by Value, which equals (PY/A – League PY/A) x attempts. If you’d prefer to ignore volume, Relative is the quarterback’s PY/A compared to league average. Win % represents the team’s win percentage in games started by the QB.
I won’t bother telling you what you can see with your own eyes, but there are a few things worth noting. For one, this list dovetails closely with traditional rankings of QB seasons that do include INT’s and sacks; the best seasons were generally great regardless of how many negative plays the QB had. The other main takeaway from PY/A is how strongly it correlates with winning. The correlation between Relative PY/A and Win % is a robust .56. This is pretty remarkable considering PY/A not only excludes interceptions and sacks, but also ignores fumbles, the running game, defense, and special teams. I feel confident in stating that the most vital factor in winning games is consistently producing big plays in the passing game.
The relationship between PY/A and winning holds true from the best passers to the worst and everyone in between. In the table below, I split Relative PY/A into ten bins from highest to lowest. As you’ll see, team Win % drops in lockstep with a drop in PY/A. Teams featuring an elite PY/A offense can expect to win more than three quarters of their games, while PY/A failures barely win one quarter of their matchups.
In data analysis, sometimes a picture really can tell a thousand words. Below are three visualizations of PY/A in relation to Win %. The first two charts show each of the 800 seasons in the sample (with different filters), and the third depicts career numbers for QB’s with 1500+ attempts since 1992.
