For NFL fans seeking strong statistical analysis, mainstream progress has been very slow. Sports networks still rank the best offenses and defenses by total yards. Even a casual fan knows yards are not more important than points.
But general scoring stats also do not paint a detailed picture of how a team's game or season went. A lot of the numbers in a game summary suffer the same problems. We need deeper context.
Fortunately, there is a better alternative. One that hopefully will become a permanent fixture in even the most general NFL analysis, and that is to look at the drive stats.
Football is a series of drives
At its core, football is a game primarily about offense versus defense, sprinkled in with some special teams.
Every team may play four quarters and 60 minutes, but what matters are the possessions; the drives that decide the outcome. Those are not constant. How the game was played - different paces and tempos - dictates the number of drives, and all of this is very important when analyzing a team's stats.
You may have seen some drive stats recently as the networks showed how quarterbacks Mark Sanchez and Tim Tebow have equally led the New York Jets to zero touchdowns on 15 drives each this preseason.
Last season, the worst offense in the league at scoring touchdowns - the St. Louis Rams managed just 16 - scored a touchdown on 8.4 percent of their drives. The Green Bay Packers led the league at 37.5 percent (63 touchdowns on 163 drives).
Fortunately, a website like FootballOutsiders compiles drive data regularly, and you can view it for the 1997-2011 seasons. You can look at offense or defense, and the bottom table includes net drive statistics that take both into account.
Why drive stats work better than traditional stats
Let's apply drive stats to the pair of 2011 games between the New England Patriots and New York Giants. First it was a 24-20 Giants' victory in New England in Week 9, and of course a 21-17 win in Super Bowl XLVI.
By traditional stats, one may say the Giants did better offensively the first time. Having 24 points is better than 21, and if you remember the Super Bowl starting with Tom Brady's bizarre intentional grounding for a safety, then the disparity is actually 24 to 19.
But how many people remember that the Week 9 game was 0-0 at halftime? Drive stats tell a much different story of how well the Giants played offensively. In each game, the Giants had a kneel-down drive before halftime, which is excluded.
- In Week 9, New York's offense had 13 possessions, scored 24 points (1.85 per drive), and gained 362 yards (27.8 per drive).
- In Super Bowl XLVI, New York's offense had eight possessions, scored 19 points (2.38 per drive), and gained 397 yards (49.6 per drive).
If you take both totals and drive stats, this is where the Giants would have ranked in the 2011 regular season:
- Week 9 (totals): 24 points (10th) and 362 yards (14th).
- Week 9 (drives): 1.85 points per drive (15th) and 27.8 yards per drive (19th).
- Super Bowl XLVI (totals): 19 points (26th*) and 397 yards (5th).
- Super Bowl XLVI (drives): 2.38 points per drive (4th) and 49.6 yards per drive (1st).
*20th using the Giants' full 21 points for an apples-to-apples comparison. The NFL scoring rankings do not separate for non-offensive scoring plays. Drive stats do.
The Giants had an elite performance on offense in the Super Bowl when you look at the drive stats. On eight possessions, they scored two touchdowns and two field goals, while also moving the ball well on four drives that ended in a punt.
Yet that would never be recognized if you only looked at their below-league-average 19 points on the scoreboard. But an offense can only score when they have the ball, and on eight possessions, they produced like an elite offense.
Manning's highly efficient offense limited his opportunities, as did a Colts defense that gave up many completions and plentiful gains on the ground. It was the perfect storm for games with few possessions, and ultimately it means the Colts' offensive stats are better than they look, while the defense was actually worse.
Drive stats properly credit offenses and defenses for scoring plays they were actually on the field for, while giving us a much better indicator of efficiency and performance than the totals ever will.
Current and future applications of drive stats
There are many interesting areas of study one can do by analyzing drive stats.
During last season's playoffs I analyzed Bill Belichick's Super Bowl defenses in New England based on drive stats, and went through 3,679 postseason drives for quarterbacks to find some fascinating results.
The research I have spent years on for fourth-quarter comebacks and game-winning drives is essentially looking at drive stats. From that alone we could use drive stats to study and quantify the two-minute drill, the four-minute offense, and which defenses protect the lead best in the clutch.
Field position is a significant factor, especially now with the touchback-heavy league following last year's rule change on kickoffs. Scoring a touchdown when starting at your own 20-yard line is not as likely if you start at the 35. Drive stats quantify such things.
Which offenses are the best at avoiding three-and-out drives? What impact does an excessive amount of three-and-out drives have on a defense? Analyze the drive stats.
If someone researched drive stats for past decades and found that teams average fewer possessions (and plays) in today's game, then that would be the proof that scoring is indeed higher than ever. Once again, offenses can only score if they have the ball, and those opportunities are never constant game to game.
Hopefully we will continue moving forward with football analysis by breaking down games by drive. Anything that gets people to start looking at drives and play-by-play instead of just the league standings and volume stats is a step in the right direction.
Scott Kacsmar (@CaptainComeback) writes for Cold, Hard Football Facts, Bleacher Report, Colts Authority, and contributes data to Pro-Football-Reference.com and NFL Network.