The NFL's passer rating system is the Dow Jones Industrial Average of football: You hear about it all the time, but you probably don't know how it's calculated.
Actually, the NFL's formula isn't
too complicated
– not if you have a few minutes to sit down and work it out. Basically, it computes a passer's performance based upon his completion percentage, TD percentage, INT percentage and yards per pass attempt. You can
find the formula here, along with many other places.
Passer rating is nice. But there are better, easier and more accurate ways to meaure the ability of quarterbacks to throw the ball.
We have two main complaints with the passer rating system:
PROBLEM ONE: Passer rating is difficult to understand
In baseball, if Derek Jeter hits .342, you don't have to be a math whiz to know that he hit safely in 34 percent of his at-bats. But what does Damon Huard's lofty NFL passer rating of 105.2 mean? It's really just a number that exists in its own little world, like a statistical bubble boy.
PROBLEM TWO: Passer rating overestimates the value of completion percentage
Ask yourself this simple question: Would you rather complete 2 of 3 passes for 9 yards or 1 of 3 for 10? Of course, you'd rather pick up those 10 yards and a first down. The truth is that the percentage of passes a quarterback completes matters little. It only matters how far and effectively he moves the ball downfield.
If you want a good example of the uselessness of completion percentage, consider that midway through 2006, Houston's David Carr has completed 70.4 percent of his passes, while Indy's Peyton Manning has completed 65.1 percent. Does anyone believe Carr is having a better season than Manning, just because he's completing a higher percentage of shorter passes?
No.
A BETTER WAY: Simple Passer Rating (SPR)
There's a much better way to rate passers. We call it Simple Passer Rating. It's so easy to understand that even you can figure it out in minutes – before your busy day of collecting empty beer cans begins.
SPR is based upon just two factors, but they're the most important measures of a passer's skill: interceptions and yards per pass.
Factor One: interceptions
In the 1987 book "Football by the Numbers," which was coauthored by George Ignatin, we calculated the value of interceptions by looking at every NFL game played from the inception of the 14-game schedule in 1961 through 1986. We figured the yards gained on an average possession, as well as yards gained on an average punt and punt return. For good measure, we also added in the number of yards gained on all interception returns.
We determined that the average interception in the NFL cost an offense 49.6 yards (and, conversely, was worth 49.6 yards to a defense). Four years later, with the assistance of
Coach T.J. Troup, we analyzed the seasons from 1987 to 1991 and found that the value of an interception had scarcely changed. It was now 49.7 yards.
In 1999, we calculated all games for the four seasons from 1995 to 1998, and the average was still 49.7. For ease in calculating our SPR, we rounded it to 50.
Factor Two: yards per pass (YPP)
We'll give it to you straight: Yards per pass may be the single most important statistic in pro football, the most reliable measure of a team's strength on offense and defense. Every winning team in the modern NFL has finished near the top of the food chain in yards per throw on either offense or defense – and usually in both.
The Super Bowl champion Steelers led the league in yards per pass on offense (8.19) last year, and were fifth on defense at 6.34 YPP. This year, the undefeated Colts are second in yards per pass on offense (8.15) and, despite their struggles on the other side of the ball, remain in the top 10 in this category defensively (6.42 YPP).
Yards per pass is simply the yards a team gains through the air divided by the number of attempts (not completions, but attempts).
It tells us how well a quarterback is moving the ball downfield. To use the Carr-Manning example again, we get a better look at the effectiveness of each by looking at yards per attempt. Carr averages 6.75 yards per pass. It's not bad, but not in Manning's class of 8.15 YPP.
By our estimation, around 80 percent of all NFL games since 1961 have been won by the team that, in the course of the game, posted the highest average per pass attempt.
Don't take our word for it. Amaze your friends with this simple trick: On Monday mornings, have someone open the paper to the sports section and, without identifying the teams or the score, tell them to give you just two simple stats: the number of yards both teams passed for and the number of times they threw the ball. You'll be able to determine the winner four out of five times simply by looking at these numbers.
Yards per pass is probably the most underrated and underutilized stat in pro football. Nearly every sports Web site includes it somewhere, but you have to seek it out because practically no one points out its significance. The biggest flaw in the NFL's passer rating system is that it gives equal weight to completion percentage, which is almost meaningless, and yards per attempt, which is vital.
You can watch an entire season of NFL telecasts without hearing a commentator mention yards per pass, and you can flip through even the most detailed books of football analysis without seeing it. Its simplicity often eludes even the sharpest fans.
Calculating Simple Passer Rating
Yards per pass can be refined even further.
For instance, the Cold, Hard Football Facts look each week
at yards per pass attempt, which is figured almost the same way as YPP, but which adds sacks to the number of passes and subtracts sack yards from gross yards passing. That formula is meant to evaluate a team-wide ability to pass the ball.
Our SPR calculation is more about the individual performance of passers. We feel sacks have more to do with the team itself than with the passer's ability.
Yes, we're aware that all passing statistics are in large part team statistics, and it's impossible to say with certainty that any passing stat doesn't reflect the ability of a quarterback's blockers and receivers, not to mention the quality of his defense and the field position it gives him.
Nonetheless, we're going to make the assumption that, for the most part, a quarterback's passing numbers reflect his individual talent, and we're not going to figure in sacks, if only because the NFL's rating system doesn't consider them either, and we're trying to compare their system with ours.
So here's what we did: We ranked all NFL quarterbacks using our Simple Passer Rating formula. Calculating it is easy. Take gross passing yards, subtract 50 yards for each interception and divide by the number of attempts. The formula looks like this:
SPR = (Passing yards – (INTs*50))/Attempts
Here are the Top 20 passers for the first half of the 2006 season ranked by SPR (only passers with at least 120 throws were included in our evaluation):
Look who else didn't make our cut: Dallas QB Drew Bledsoe. He's 25th in the NFL, averaging just 4.52 SPR. Considering how bad those numbers are, the Cowboys would have been better off running the ball on every play, where they average 4.2 yards per carry.
If you're looking to figure out your team's chances of making the playoffs, here's a simple rule of thumb: If your quarterback isn't in our Top 20 after the first eight games, start planning for the draft. If he's not in the Top 12, plan to spend weekends in January working on the house or taking up hockey.
