/cdn.vox-cdn.com/photo_images/1057190/GYI0061308401.jpg)
In the glory days of my youth, I was quite a formidable baseball player. I spent most of my free time playing or watching baseball. That was some time ago, and these days when I watch the occasional baseball game, I have to keep my laptop close at hand in order to look up all those fancy stats they keep showing online or on TV, and that I'd never heard of in my youth. OPS, ISOP, SECA? Baseball has sure come a long way statistically with Sabermetrics and Moneyball.
Football, on the other hand, is easy to understand. I could watch a football game with my 71-year old dad - who last watched a live football game in the 80s, and still thinks Dan Marino is the greatest football player ever - and my dad would still understand every single stat they put on the screen. And whether that's a good thing or a bad thing doesn't really matter. That's just the way it is.
Football is still very much dominated by volume based stats like total yards, number of completions, number of touchdowns and the like. Perhaps football fans and broadcasters are still traumatized by the introduction of the much misunderstood passer rating 40 years ago, which remains one of the few efficiency measures being broadly used today.
So over the next couple of posts, I'd like to introduce you to some of the stats that are a little off the beaten path and that I'll be keeping a close eye on this season. Today, we start with Adjusted Net Passing Yards per Attempt (ANPY/A).
What is ANPY/A?
Adjusted Net Passing Yards per Attempt is a stat that was introduced in a book called The Hidden Game of Football by Bob Carroll, Pete Palmer, and John Thorn. It was recently amended by Chase Stuart at Pro-Football-Reference.com (PFR).
Here's the exact definition:
(Passing Yards + (Passing TDs)*20 - (INTs thrown)*45 - Sack Yards) / (Passing Attempts + Sacks)
And because ANPY/A is kind of hard to pronounce, from now on I'll call it adjusted passing yards in this post.
What does it measure?
It's basically passing yards minus sack yards divided by pass attempts minus sacks - with a 20 yard bonus for every touchdown and a 45 yard penalty for every interception. The result is a number that resembles the more prevalent Yards per Pass Attempt stat and is usually somewhere in the range between three and eight yards.
It provides a single, easy-to-understand number that encapsulates a team's passing performance using five different passing stats, only leaving out fumbles in terms of what a quarterback does. Best of all, it is a rate stat, and not a simple volume- or counting stat.
Why is it important?
Adjusted passing yards is a souped up version of your basic Yards per Pass Attempt stat that reflects both the reality of today's NFL and the importance of passing efficiency in the modern game. ColdHardFootballFacts.com (CHFF) weighed in last year on why they think passing yards per attempt is the single most important indicator of success in all of football:
* Teams that win the passing YPA battle almost always win the game.
* Teams that lose the passing YPA battle almost always lose the game.
* The winningest teams in history are typically the teams with the best passing YPA average
* The winningest quarterbacks in history are typically the quarterbacks with the best passing YPA average
I recommend you read the full version of their rant, it is really quite entertaining.
Adjusted passing yards and the Dallas Cowboys
To understand whether a given stat is relevant or not, it helps to know whether that stat has a high correlation to, say, the outcome of the game. So I went and looked at the adjusted passing yards for the Cowboys and their opponents in the last four years to understand how accurate the stat is a a predictor for who won the game. As an example, here's the adjusted passing yards calculation for the Saints game last season:
Cowboys - Saints, December 19, 2009 |
|||||||||
Team | CMP | ATT | YDS | TD | INT | SACK | SACK YDS | ANPY/A | |
Cowboys | 22 | 34 | 312 | 1 | 0 | 3 | 18 | 8.5 | |
Saints | 29 | 45 | 298 | 1 | 1 | 4 | 27 | 5.0 |
In the 64 regular season games from 2006-2009, the Cowboys compiled a 42-22 W/L record. In 38 of their 42 wins (90.5%), the Cowboys had an ANPY/A that was higher than their opponents. In 20 of their 22 losses (90.9%) the Cowboys also lost the adjusted passing yards battle. Adjusted passing yards was an accurate predictor in 58 out of 64 Cowboys games (90.6%). Have you ever seen a stat that had such a high correlation to wins?
Adjusted passing yards and the NFL
mgrex03 over at SB Nation's StampedeBlue.com has an outstanding series of posts dealing with different stats (start reading here) and how they ultimately correlate with wins in the NFL. Among many other stats, he also looked at adjusted passing yards. He looked at the last 13 years in the NFL and found that ANPY/A was an accurate predictor in 89.1% of games in which the offense and defense had above average ANPY/A figures. This method is slightly different from the straight up comparison I did for the Cowboys above, but adjusted passing yards is the stat with the highest winning percentage in the NFL that mgrex03 has found to date.
Adjusted passing yards and Tony Romo
Since adjusted passing yards is a measure of how efficient the passing game is, it can be used just as well to measure individual quarterbacks.
For those of you who've read the MSP Cowboys Annual (and for those of you who haven't: order here), you'll know that Tony Romo surpassed 1,500 career passing attempts in the 2009 season. By passing that mark, Tony Romo has now entered the "official" NFL record books, and he enters the record books at the very top. His career passer rating of 95.6 is third all time behind only Steve Young (96.8) and Philip Rivers (95.8) and ahead of Peyton Manning (95.2) and Kurt Warner (93.7).
But perhaps even more impressive is that he also ranks in the top five in career passing yards per attempts with 8.1 YPA behind only Otto Graham (8.98) and Sid Luckman (8.42) and ahead of Norm Van Brocklin (8.16).
So it won't come as a great surprise to find that he also ranks fairly high in adjusted passing yards. But just how high?
NFL Career ANPY/A Pass Attempt Leaders |
||||||||
Top 10 |
11-20 |
|||||||
Rank | Player | ANPY/A | Rank | Team | ANPY/A | |||
1 | Tony Romo | 7.20 | 11 | Dan Marino | 6.55 | |||
2 | Philip Rivers | 7.15 | 12 | Matt Ryan | 6.29 | |||
3 | Peyton Manning | 7.10 | 13 | Ben Roethlisberger | 6.26 | |||
4 | Aaron Rodgers | 6.86 | 14 | Trent Green | 6.24 | |||
5 | Steve Young | 6.85 | 14 | Jeff Garcia | 6.24 | |||
6 | Kurt Warner | 6.71 | 16 | Chad Pennington | 6.08 | |||
7 | Matt Schaub | 6.71 | 17 | Carson Palmer | 6.06 | |||
8 | Tom Brady | 6.65 | 18 | Brett Favre | 5.98 | |||
9 | Drew Brees | 6.61 | 19 | Donovan McNabb | 5.97 | |||
10 | Joe Montana | 6.60 | 20 | Mark Rypien | 5.96 |
Of course, you'll notice quickly that the list above is populated largely by quarterbacks who are still active today. We've detailed extensively how the NFL has turned into a Passer's Paradise and have also looked at how today's crop of QBs isn't automatically better than the Peerless Passers of yesteryear.
But still, to be the career leader in this stat category, that is a mighty fine achievement Mr. Romo. Even the good folks at PFR are beginning to realize that Tony Romo's career stats may not be as misleading as they may have thought.
Adjusted passing yards as a measure of team strength
Another way to look at this stat is by calculating the difference between a team's adjusted passing yards gained on offense minus the same metric given up on defense. This adjusted passing yards differential effectively tells you how many more adjusted yards a team gains per pass attempt than its opponents, and it can be a very strong indicator of team strength.
ANPY/A Differential by NFL team, 2009 (click column header to sort)
Team | Offensive ANPY/A | Defensive ANPY/A | ANPY/A Differential |
---|---|---|---|
Packers | 6.8 | 3.7 | 3.1 |
Saints | 7.4 | 4.5 | 2.9 |
Chargers | 7.7 | 5.2 | 2.5 |
Eagles | 6.4 | 4.2 | 2.2 |
Colts | 6.5 | 4.7 | 1.8 |
Patriots | 6.8 | 5.1 | 1.7 |
Cowboys | 7.1 | 5.4 | 1.7 |
Vikings | 7.1 | 5.6 | 1.5 |
Texans | 6.8 | 5.3 | 1.5 |
Steelers | 6.5 | 5.3 | 1.2 |
Ravens | 5.6 | 4.5 | 1.1 |
Broncos | 5.5 | 4.4 | 1.1 |
Cardinals | 5.6 | 4.7 | 0.9 |
Bengals | 5.0 | 4.4 | 0.6 |
Giants | 6.4 | 5.9 | 0.5 |
Bills | 3.8 | 3.3 | 0.5 |
Jets | 3.7 | 3.3 | 0.4 |
49ers | 4.7 | 4.8 | -0.1 |
Panthers | 4.1 | 4.2 | -0.1 |
Redskins | 5.1 | 5.5 | -0.4 |
Titans | 5.1 | 5.6 | -0.5 |
Jaguars | 5.4 | 6.4 | -1.0 |
Falcons | 5.1 | 6.1 | -1.0 |
Bears | 4.2 | 5.5 | -1.3 |
Seahawks | 4.4 | 6.1 | -1.7 |
Dolphins | 4.3 | 6.2 | -1.9 |
Buccaneers | 3.3 | 5.4 | -2.1 |
Chiefs | 4.0 | 6.2 | -2.2 |
Raiders | 3.5 | 6.5 | -3.0 |
Browns | 2.9 | 6.5 | -3.6 |
Rams | 3.2 | 7.0 | -3.8 |
Lions | 3.0 | 7.3 | -4.3 |
The Cowboys tied with the Vikings for the third best adjusted passing yards offense, but ranked only 17th in adjusted yards pass defense. But the 1.7 adjusted yard differential was still good enough to tie with the Patriots for 6th place in the league.
As you look at the differentials, you'll find that there is a strong correlation between the yards differential and the number of wins per team. The Bills are the only team on the list with a positive yardage differential that had a losing record last season. Conversely, the Falcons are the only team with a negative yardage differential with a winning season last year.
In fact, the correlation (r²) between the adjusted passing yards differential and the number of wins in 2009 was an astonishingly high 0.78.
When all is said and done, ANPY/A may just be the Robitussin of stats. No matter what you've got, Robitussin better handle it.