“He is a better linebacker than that guy. Just look at the difference in tackle totals.”
“This cornerback is clearly elite. I mean look at how many interceptions he’s had over the past couple of years.”
“He’s one of the five best receivers in the league. Just look at his yards over the past three seasons.”
“I want that guy in my locker room any day of the week. He just wins.”
These statements are ones that I have heard repeated time and time again when it comes to evaluating player skill and talent. I have heard it from the experts—and I use that term loosely—on ESPN and the NFL Network. I have read it posted by individuals on various football message boards around the Internet. I have most certainly heard these arguments used on Bleacher Report.
Clearly the premier way to judge players is through objectively viewing every game possible but there are a few problems with this approach.
Although the ability now exists to watch all 256 regular season games with an Internet connection and 20 to 40 dollars it is incredibly time consuming and egregious. Additionally, the average fan would arrive at the same conclusion that a simple statistic would get them instead of thoroughly breaking down the play.
For this reason most fans rely on base statistics and Sunday Night highlight reels to arrive at their conclusion on how well a player plays.
Now it is obvious that statistics are never the be-all and end-all in regards to critiquing play but they have been elaborated further over the years. These additions to statistics—known as metrics—have managed to shed light on stats and have become more commonplace. They’re becoming so present that even some casual fans are referencing them.
Professionals such as Football Outsiders and KC Joyner have all been at the forefront of metrics since 2004 and Stats LLC has been around since 1995. These three outsources are all recognized by the league and its teams as well as various professional media outlets. As these sources are present there is no logical reason to simply look at a single statistic and run away with it.
Statistics needs to be subject to introspection and the following statistics are the biggest culprits.
The tackle is perhaps the most annoying statistic in football. Not because it isn’t useful but because it is so open to interpretation. For this reason, although the tackle is tracked by the NFL and its various teams, it is not considered an official statistic.
Despite the fact that the tackle isn’t official fans often cite the tackle as a premier method to interpret how good 4-3 linebackers or strong safeties are. If a player has a high tackle total he is usually regarded as a “sideline-to-sideline” player.
Unfortunately, despite the forthcoming rambling, the tackle is probably a good indicator of these player’s abilities. The primary issue, however, is that a tackle as a simple number attributes the same value to every single tackle a player makes. Attributing the same value to a tackle that prevents a first down on third and three to a tackle that comes after a 25-yard run isn’t very fair.
However, our friends at Football Outsiders have long concocted an excellent way to objectively view the tackle to determine which ones are “better” than others.
Football Outsiders have given us the metrics of “Stops”, “Defeats”, “Yards per Play” and “Stop Rate”
The “Stop” is defined as a play in which a defensive player prevents the offense from gaining 45 percent of needed yards on first down, 60 percent on second down and 100 percent of the needed yards on third and fourth down. While passes defensed, interceptions and forced fumbles are included in this metric the overall numbers still remain reminiscent of tackles.
The “Defeat” is defined as a play by a defender that prevents the offense from earning a first down on third or fourth down. Additionally a defeat can be defined as a play behind the line of scrimmage or a forced turnover.
The “Yards per Play” metric is defined as the average distance at which a defender makes a tackle, forced fumble, interception or stuff. Football Outsiders has even done the fans the convenience of separating yards per play against the run and yards per play against the pass as well as when the two are combined.
Finally, the “Stop Rate” metric delivers the percentage of tackles, stuffs, sacks, forced fumbles and interceptions that qualified as a stop out of their total plays.
Using the stop metric alongside the defeat metric is a good way to further elaborate on tackles. Together they can tell you how many of the player’s tackles weren’t simply “garbage” or “cleanup” tackles. This does not mean that all tackles that weren’t stops or defeats were meaningless but these metrics do give a better look into when and where tackles occurred when combined with the yards per play metric.
Who should be seen as the superior player?
A guy who had 100 tackles with an average yards per play of 3.5 with 66 stops and 20 defeats or a guy who had 130 tackles but an average yards per play of 6.0 with 62 stops and 8 defeats?
The tackle totals tell you that the latter player is superior but are his 30 extra tackles really worth the 2.5 extra yards he gives up per play?
Now obviously other factors must be placed into context such as the system that a guy plays in or where on the field the tackles were made despite distance. If you want to further do some research on a player and see if they are truly a “sideline-to-sideline” tackle you can also check out their player splits page on Sports Illustrated.
For all of the aforementioned metrics you can search Football Outsiders with the player’s name to find their respective player page.
The sack is probably my favorite statistic out of the ones that I have deemed “overrated”. Perhaps it’s because I’m a Steelers fan and they specialize in that. Or perhaps it is because I am a fan of defense and love seeing pretty boy quarterbacks get their comeuppance.
Regardless of my care for the sack it also remains one of the most overused—and thus overrated—statistics by fans.
Usually fans equate pass rush ability with the total number of sacks that a player gets. Like the tackle it’s not that bad of an idea, in theory. The culmination of a pass rush is the sack and at a basal level it then makes sense to say whoever gets the most sacks is the best pass rusher. Unfortunately this is not true as it fails to factor in rush rate, pressures and quarterback hits.
As much as I can be found defaming Pro Football Focus for their attempts to become a big player in the metrics game they do manage to do one thing well.
They have a section of their site dedicated to simple player participation. In this section they tell you how many snaps a player took pass rushing, covering and playing the run. This is the one thing that Pro Football Focus offers to the metrics game that others do not. I wish that most other metrics places would include snap details so that I could completely abolish my use of Pro Football Focus.
What is impressive about this section of the site is the ability to determine the “rush rate” of a player or how often they barrel towards the quarterback. If a player rushes the quarterback 999 times out of 1,000 snaps are they truly the best pass rusher in the league simply because they notched the most sacks?
Additional stats that are to be considered when considering how good of a pass rusher somebody is are quarterback hits and quarterback hurries.
Hits can be found in any GameBook over on ESPN as soon as the game has finished telecast and gone final. However, our friends at Football Outsiders keep track of them for us as well. Additionally they keep track of quarterback hurries as well. While other places manage to track hurries their standards of a hurry have universally been considered to “open”.
Just because a pass rusher does not get to the quarterback for a sack does not mean they did not do their job. A hurry or hit can result in an interception which is arguably better than a sack. The league leader in hurries in 2010 finished tied for 24th in sacks. The league leader in hurries in 2009 finished second in sacks but was widely considered to garner most of those sacks in two games. In 2008 the league leader in hurries finished 11th in sacks.
Clearly the sack is not the only way to determine pass rushing ability and the hurry and hit should be considered as well.
In the earlier half of the past decade the interception was the bane of my existence when it came to debating cornerbacks.
Way too many times did I have to hear about “Deltha O’Neal” or “Lito Sheppard” being lockdown cornerbacks because they had high interception totals. Perhaps what frustrated me the most was the simple fact that looking at guys like that for just one game would tell you they were the furthest thing from lockdown. Rather they were often in the right place at the right time.
Then K.C. Joyner turned the football world on it’s head and offered metrics to the public in 2004.
Stats LLC had long recorded coverage metrics since 1995 but never were they offered to the mere public.
Now there were metrics corroborating what was obvious to the people watching the games. Furthermore, they were allowing the true lockdown corners like Sheldon Brown to be recognized when they weren’t snagging interceptions.
Since then many people have thrown their hat into the ring including Football Outsiders, Tony Pisano and Pro Football Focus and have offered their metrics to the public. Since then metrics have almost become commonplace as accepted in judging cornerbacks.
Prior to coverage metrics being common, the interception was often considered the best way to judge a cornerback by casual fans. In fact, some people still use the interception to grade out cornerbacks as well as safeties.
The issue here though is that the interception still remains, more often than not, the result of an errant throw as opposed to an excellent play on the ball by the defensive back. I am not saying that all interceptions are errant throws but the majority of them in fact are.
Fortunately for us the good folks at Stats LLC have worked alongside Sports Illustrated and each team’s official statistician. Each statistician when scoring an interception scores it based on type of throw while eliminating as much bias as possible.
On the bottom of each defensive back’s splits page on SI.com you can find how the statistician scored their interceptions. For example the statistician for Baltimore scored two interceptions to league leader Ed Reed on dropped/tipped passes, one on an overthrow, one on a wide throw and three more from jumping routes. The remaining interception wasn’t scored because the QB was hit while throwing the pass.
I am in no way saying that the interception is a bad thing. Nor is it a bad thing to get lots of them. After all as Rams (Ronald Bartell), Steelers (Ike Taylor) and Redskins (Carlos Rodgers) fans can tell you it still takes skills to actually be able to catch the ball when it comes to you.
What I am trying to say is that just because a defensive back totals a lot of interceptions from time to time doesn’t mean they’re stout in coverage. If Ed Reed can have four interceptions quantified as errant imagine what other lesser defensive backs have.
My antagonism for receiver production is one that has arisen recently. More specifically it is due to the fact that receivers like Brandon Marshall have been graded out as better players than they actually are due to their production.
Now I am not saying Marshall is not a good player but his production mirrors how frequently he is used rather than his overall skills. Nothing emphasizes this more than target numbers which can be found anywhere hosting Stats LLC statistics.
When a wide receiver regularly needs around 160 targets to get their production than they are largely inefficient and tend not to be as good as advertised, although there are a couple of exceptions to this rule.
Last offseason ESPN regularly toted Brandon Marshall as a top five receiver because he regularly had 100 receptions over a three year period. However, what they ignored was that he was top three in targets every one of those seasons eclipsing the 155 mark every year and twice surpassing 170 targets.
Sure the 100 receptions were nice but the fact is he had to be thrown at more than everyone else in order to catch them. In comparison Wes Welker had nearly identical production while eclipsing the 150 target mark only once over that three year span. Had other receivers been targeted that often their production would have exceeded those totals easily.
Usually when a receiver is being thrown at 160-plus times they produced not because of quality in that given season but due to quantity. Larry Fitzgerald’s 2010 season is indicative of this. The only receiver recently that has produced well with 170 targets is Andre Johnson in 2008 and 2009 which is a small outlier in a large pool.
I hate this statistic for the reason that it is way too oversaturated by the media and the casual fan. This “statistic” has made mindless drones of my fellow Steelers fans who simply bank on this statistic to rate Ben Roethlisberger as opposed to the plenty of other pertinent ones that are just as viable to rank him as a top quarterback.
The primary problem with this particular statistic is that it ignores that football is a team game. In fact, football is arguably the premier team sport in the entire world. Attributing a win or loss to one particular player in most instances is an unfair and fallacious thing to do,
Granted a quarterback has more effect on a football game than any other player but the reality is he is one of 22 players on the field at a given time. A quarterback does not effect special teams scoring, nor does he personally prevent the opposing offense from scoring.
Essentially a quarterback can be incredibly inefficient and do everything in his power to lose a game but if his defense and special teams are playing at their best he could still receive a win.
Additionally a quarterback can be knocked out a game that their team was losing only for the team to eventually win and be credited with a win on their record. Sure these rare occurrences but it is indicative of why it is not fair or logical to attribute a win solely to a quarterback.
Unfortunately there does not exist a viable calculation tool or metric in order to find out how to properly consider how important to a team’s success a quarterback was in a given year.
So there you have it...
In case you’re normal and don’t want to view 256 regular season game a multitude of times there are several extra steps you can take to ensure that you’re not just rehashing the statistics from the player pages on NFL.com. Hopefully if you were unaware of them you read through the lengthy article and are now willing to utilize these metrics in evaluating players.