I love statistics. Absolutely love them. I've always been a numbers guy, dating back to when I was a kid.
I spent much of my childhood poring over numbers in my Total Baseball Encyclopedia, memorizing Hank Aaron's year by year home run totals or the top-10 all-time single-season stolen base leaders. I am a huge fan of baseball statistician Bill James, who has completely revolutionized the game of baseball.
While casual baseball fans know a player based on his batting average, home runs, and runs batted in, I found myself more interested in the ratios and percentages.
James and the guys at Baseball Prospectus have invented countless useful statistics, such as offensive winning percentage, batting average on balls in play, and power speed number. Those little-known stats are the ones that fascinate and intrigue me more than any others.
I wish the NFL had statistics like these.
I know football isn't as much of a numbers game as baseball is. Stats don't always tell the whole story in the NFL. There's more to the game, like intangibles, leadership, and so on. In my eyes though, stats can do a pretty good job of measuring a specific player's talent.
Nearly half the offensive starters (the five linemen) have no official statistics by which to judge their performance.
The only way to know how well your team's left guard did this past season is by taking the word of the NFL experts who watch and break down the film of each and every game. It's tough to know for every player though.
Take Jon Runyan as an example. The right tackle of the Eagles for the past nine seasons, Runyan has been a Pro Bowl-type player for his whole career. He has been one of the best right tackles in the NFL and instrumental in paving the way for All-Pro back Brian Westbrook. That said, I don't have a clue how he did in 2008.
I have no access to the few stats that exist for offensive linemen (sacks allowed) and no basis for which to make my judgment—and I watch and dissect every single Eagles game. There needs to be a way that we as fans can have a more accurate understanding of the effectiveness of offensive linemen in the NFL.
I compiled a list of 10 NFL statistics I would like to see. These are statistics I want to see on ESPN.com, next to the household numbers like passing yards, tackles, and sacks.
Some of these stats are probably in use by Elias Sports Bureau or the Pro Football Prospectus writers, but they're not readily available to the general public. I want to be able to go on NFL.com or ESPN.com and find these stats to use in my arguments.
Running back rating
Everyone knows about quarterback rating. No one knows how to calculate it, but football fans like myself know 80 is average, 100 is great, and 158.3 is perfect. It's a measure of a quarterback's efficiency in a given season or game.
How about running back rating?
This stat could be a measure of a running back's success carrying the ball. It could somehow factor in yards per carry, rushing yards per game, touchdowns, and fumbles. I have played around with the numbers trying to figure out the formula for running back rating but have been unsuccessful.
Admittedly, it would be difficult to calculate. And there would definitely be some flaws, just like QB rating, which doesn't account for a quarterback's rushing yards or lost fumbles, as well as the ability of his supporting cast.
I would place the most emphasis on yards per carry and touchdowns, with fumbles also playing a big role.
For a running back like Brandon Jacobs, who gets all the team's goal-line carries, his touchdown rate would be higher, which would thus boost his RB rating. In the same way, this would hurt a player like Derrick Ward, who is not normally given the ball in goal-line situations.
RB rating wouldn't factor in receiving yards, which again would be a flaw, because a player like Brian Westbrook receives no extra credit for his receptions compared to a player like Ricky Williams.
However, I think this would be a good way to measure the performance of running backs, just like that of quarterbacks.
It could use the same numerical scale as QB rating (although I don't know why whoever invented QB rating didn't just make it so the maximum was 100, instead of an obscure number like 158.3), and the statline for running backs could now include the position's own sort of rating system.
I love a big, power back like Brandon Jacobs, Marion Barber, or LenDale White, who can break tackles and pick up the tough yards.
I think the NFL should keep track of broken tackles, average broken tackles per rush, average broken tackles per game, and average broken tackles per attempted tackle as official statistics for running backs.
I remember watching a run by Marion Barber in '07 against the undefeated Patriots in which he gained just two yards, but broke six tackles before the seventh defender finally pushed him out of bounds.
For a run like that, Barber would be credited with six broken tackles. His six broken tackles divided by seven tackle attempts would give him an astounding broken tackle rate of 0.857.
For that game, Barber rushed for 49 yards on nine carries. Let's assume as a hypothetical example that Barber broke an additional three tackles in that game, giving him nine broken tackles for the day.
Barber would be credited with nine broken tackles. His average broken tackles per rush would be the number of broken tackles divided by the number of rushes—exactly 1.0.
And his average broken tackles per attempted tackles would be the nine broken tackles divided by however many attempted tackles—say there were 16, which would give him a rate of 0.625.
Rush success rate
I wouldn't be surprised if this stat exists somewhere, but I have never seen it in use. This stat would break down every single carry by a running back and grade it as either successful or unsuccessful.
Then the number of successful carries would be divided into the number of total carries for a successful rush percentage.
It's tough to classify a carry as successful or not, and a lot of it would depend on the situation. Here are some generic examples:
- Rushes for a touchdown
- Rushes for a first down
- Rushes that gain at least half of what is needed for a first down (Example: A three-yard run on 2nd-and-6 or a five-yard run on 2nd-and-10. Obviously if the three-yard run came on 3rd-and 6, it would not be considered successful, because the team would then be faced with a 4th-and-3, and most likely have to punt).
- Any run of positive yardage that comes with a lead and under five minutes to play, because it runs the clock down
- A big run on third-and-long (Example: It is 3rd-and-23, and a team hands off the ball. They're obviously just trying to pick up some yards to get better field position, so if the back gains 17 yards, it would be considered successful, even though he didn't pick up the first down).
- Any loss of yards
- Any fumble, regardless of how many yards are gained on the play
- Any failed attempt on third or fourth down
- Any run of three or fewer yards on first down, since four yards is expected for a run
- Any rush that gains less than half of what is needed for a first down (Example: A three-yard run on 2nd-and-7 or a five-yard run on 2nd-and-11 because if that run were reproduced, the team would not get a first down).
I think this would be one of my favorite statistics to see the NFL use. I would love to see a hypothetical scenario like Adrian Peterson led the league in '08 with a 72 percent rush success rate.
There would be some flaws in this system for sure. A goal-line back would do better than a regular back, because there are fewer opportunities and a greater chance of success. And there are times when it would simply be difficult to measure the success of a run.
Say it's 3rd-and-23, and the back gains 17 yards. That is successful, as I stated. What about 14 yards though? What about 10? What about seven? At what yardage does a run become successful? It would take some time to create a realistic working formula and it would always be debatable, but I would love to see the results.
Offensive line yards per carry
Offensive linemen are by far the most underappreciated men in all of professional sports. Football games are won in the trenches, but the five guys blocking for your team's big stars are often unknown by everyone but the hardcore fans.
I would like to see a statistic that would measure an offensive lineman's average yards per carry. Say we want to look at Joe Thomas, the All-Pro left tackle for the Cleveland Browns, as an example.
This stat would take every run designed primarily for the running back to run behind Thomas, and calculate the average yards per carry for those runs.
There is no denying that this would be a difficult, confusing, and sometimes unfair statistic.
What if Jamal Lewis is called to run behind Joe Thomas on a play, but upon taking the handoff sees the defensive tackle get past the left guard? Lewis then takes off to the right and follows the block of his right guard to pick up seven yards.
Who gets credit for the yards per carry? Would it be the left tackle, who was designed to be the lead blocker, or the right guard, who actually was the lead blocker? And what if the running back is designed to run between the left tackle and left guard? Who gets the credit for the yards?
I think for a play in which the back is expected to follow the paths of two blockers, each of the linemen should be rewarded for the yards. So if the back picks up nine yards, both the blockers will receive credit for the yardage.
Cornerback defensive stats
I'm sure this exists somewhere, but I have never been able to find it on any website.
Wide receivers are graded primarily based on their statistics—receptions, yards, touchdowns, yards per catch, and fumbles. I would like to see cornerbacks evaluated based on what they allow to the receivers to which they are assigned to cover.
Here's a fictitious example for this stat: In 2008, Champ Bailey allowed 43 catches for 527 yards and three touchdowns to opposing wide receivers, while intercepting two passes and knocking down 17 passes.
This would show which cornerbacks are most vulnerable to the big play. I would also like to see a breakdown of the percent of passes thrown to the receiver that are caught.
For example, Champ Bailey last year allowed his wide receivers to catch 65 percent of the passes that were thrown in their direction.
This stat would be pretty easy to track, as it simply involves keeping track of every time a defensive player drops an interception. I can't remember how many times I see a defender drop an easy interception, and I think to myself, "I wonder how many times he's dropped an interception this season."
This stat would tell us. And it would allow us to properly measure the effectiveness of a cornerback making interceptions by dividing interceptions into total attempts (interceptions plus dropped interceptions).
Ever watch a game and the quarterback hits a receiver in stride but the ball bounces off the receiver's chest, up into the air, and it's picked off by a defender? That counts in the stats as an interception, even though it was in no way the fault of the quarterback, who simply did his job by hitting the receiver with a perfect pass.
I don't like that.
The quarterback should not be penalized simply because he has an inept group of receivers. I would like to see interceptions thrown by a quarterback broken into subcategories: real interceptions and total interceptions.
The total interceptions would obviously be the number that we as fans see on the stat sheet, but the real interception total could be the number of interceptions that were actually the fault of the quarterback.
To focus even more on the topic of interceptions, we could throw in two more stats: dropped interceptions for the quarterback and possible interceptions, which would include total interceptions thrown plus dropped interceptions. Here is a fictitious example of a quarterback's stats in '08:
Jay Cutler: 18 total interceptions, 14 real interceptions, and 24 possible interceptions
This would offer a better understanding of which interceptions were actually the fault of the quarterback and which interceptions he simply could not help.
Real passing yards and extra passing yards
This past year, Drew Brees threw for 5,069 yards, narrowly missing out on breaking Dan Marino's single-season NFL record of 5,084 yards.
Of those 5,069 yards though, how many did Brees actually earn and how many came courtesy of his receivers picking up yards after the catch?
Passing yards has always been a deceiving statistic, in the fact that a five-yard screen pass that a running back turns into an 80-yard touchdown will count in the books as 80 passing yards for the quarterback, even though the running back gained 75 yards after the catch.
Meanwhile, most websites fail to mention yards after the catch along with their receiver statistics, so the back would be denied a hard-earned 75 yards after the catch.
I would like to see passing yards broken into two categories—real passing yards and extra passing yards.
Real passing yards would be the yards for which a quarterback is directly responsible, while the extra passing yards would be classified as the yards after the catch, for which the backs and receivers are mainly responsible.
This would offer a more accurate analysis of a quarterback's stats, especially for a quarterback who makes a living out of hitting running backs for short screen passes.
Passer rating plus (PR+)
I can't take credit for this. In his last article, Top-10 Quarterback Seasons in NFL History, Bryn Swartz coined this statistic as a way of comparing a quarterback's passer rating to the league average, similar to the way major league baseball does with earned run average.
What this does is provide a better comparison of the quarterback to the rest of the league. For example, a 90.0 passer rating in 2008 is far less impressive than a 90.0 passer rating in 1940.
If the league passer rating in 2008 was 82.3, a quarterback's 90.0 passer rating would generate a passer rating plus of 109.3. If the league passer rating in 1940 was 52.6, a quarterback's 90.0 passer rating would generate a passer rating plus of 171.1.
In this, the 1940 season would be far more impressive than the 2008 season.
It's easy to look at the end of the year and see that a kicker was 27-for-30 on field goals for a 90 percent success rate. I often wonder though how the particular kicker performed in clutch situations.
How do you define clutch? I think it would be any kick in the fourth quarter within two scores either way, along with the final two minutes of the first half.
Obviously if a kicker lines up for a 35-yard field goal with seven seconds to play in a tie game, it means a lot more than a 21-yarder early in the second quarter.
Kickers don't get a whole lot of clutch kicking opportunities, but I think this statistic would help to effectively separate the good from the great.