Some NFL stats should be ignored
The world of football stats has undergone immeasurable change over the past decade-plus. It wasn’t long ago that you could win an argument about who the NFL’s best linebacker is by saying your guy leads the league in tackles, or who the NFL’s best cornerback is by touting interception totals.
Today, if you bring up either of those stats in an argument, you’re probably not going to win. With deeper metrics at our disposal than ever before, some stats no longer cut it in 2024.
Without further ado, here is the worst stat to use for evaluating each position on the football field. Avoid using these stats in arguments if you want to emerge victorious.
Quarterbacks: Completion percentage
Completion percentage is a wildly misleading stat. It is intended to be a measure of accuracy, but for a number of reasons, it doesn’t tell you anything about how accurate a quarterback is.
Drops. Throwaways. Passes affected by pressure. Bad route running. None of these factors are considered in completion percentage. If it’s incomplete, it’s on the quarterback. No nuance whatsoever.
The biggest flaw with completion percentage is that it heavily correlates with the quarterback’s style of play. Game-managing quarterbacks tend to have the league’s highest completion percentages since they attempt a lot of easy checkdowns. Aggressive quarterbacks will naturally have a lower completion percentage since they attempt tough throws more often.
Jake Browning had the highest completion percentage out of 32 qualified quarterbacks last year. That should tell you everything you need to know.
Running backs: Rushing touchdowns
I could not care less about a running back’s touchdown total. More than anything, it’s just a product of how many goal-line carries the player received, which is dependent on a number of factors that had nothing to do with the RB’s individual play.
How often did the team get into the red zone? When there, how often did they happen to find themselves downed at the 1- or 2-yard line? In those instances, what was the play-caller’s preferred run-pass split? The answers to those questions are the primary factors in a running back’s touchdown total.
In 2023, 32% of all rushing touchdowns were scored from one yard out, 45% were scored from within two yards, and 55% were scored from within three yards. Touchdown totals are solely based on the number of carries you get in this range and how well the offensive line blocks for you on that small sample of plays. So, when you have a guy like Breece Hall, who only got one carry within three yards of the end zone over the entire 2023 season, how is he supposed to rack up touchdowns? Considering that, it’s actually impressive he got five rushing touchdowns.
It would be somewhat worthwhile to isolate a running back’s success rate in goal-line situations, but even then, you’re leaving out the context of how well the offensive line blocked for him.
Touchdown scoring is such a small part of a running back’s job. I am much more interested in how he contributes to moving the ball throughout the other 97% of the field.
Wide receivers/Tight ends: Drops
Okay, hear me out. I’m not saying dropped passes don’t matter. What I’m saying is that, on the whole, it’s not a catastrophic weakness for a wide receiver or tight end to have.
These are the wide receivers who led the NFL in drops this past season, per Pro Football Focus:
- Puka Nacua, 13
- Tyreek Hill, 10
- Davante Adams, 9
- Jaxon Smith-Njigba, 8
- Mike Evans, 8
- Stefon Diggs, 8
- K.J. Osborn, 7
- Courtland Sutton, 7
- Jaylen Waddle, 7
- Zay Flowers, 7
- Calvin Ridley, 7
That’s a star-studded list. At tight end, it’s the same story. Travis Kelce was second (8), while George Kittle and Sam LaPorta tied for third (7).
Drops are frustrating in the moment. However, unless you are egregiously bad in that area (see: Lazard, Allen), drops don’t happen frequently enough to severely drag down a player’s overall impact.
Look at Davante Adams, for example. He had the third-most drops with nine. Yet, if he dropped just two fewer passes, his drop rate would fall to 6.4%, which is right at the league average for the position (6.5%). Take out another drop, and now he’s out of the top 10 in total drops despite having the fifth-most targets in the league. That’s all it takes to drastically alter your positioning in the drop department. But do two or three drops over a 17-game season really matter that much?
Don’t get me wrong, drops are still a key part of the wide receiver position. It’s just a smaller piece of the pie than many think.
Offensive linemen: ESPN’s win rates
ESPN has a “win rate” system for offensive linemen, calculating each player’s pass-block and run-block win rates. The database isn’t publicly accessible, but they will occasionally share top-10 lists or reveal the ratings of certain players.
I am skeptical of this system, as it tends to spit out some bonkers conclusions. For example, according to ESPN’s Brady Henderson, Laken Tomlinson “finished 34th among all guards (left or right) in pass block win rate in 2023.” I’m assuming that’s out of at least 65-70 qualifiers since it includes both guard spots, which means the system thinks Tomlinson was an above-average pass blocker. Any Jets fan can tell you how ludicrous that conclusion is.
It’s unsurprising that this system is inaccurate when you consider its methodology. ESPN’s explanation is quite vague: “Our pass rush win rate metric tells us how often a pass-rusher is able to beat his block within 2.5 seconds.” This is done with GPS tracking. Basically, ESPN uses tracking data to determine whether the defender or blocker “won” the rep. If the defender gets past the blocker within 2.5 seconds, it’s a win for the defender and a loss for the blocker; vice versa if he fails to get there.
I understand what they’re going for, but you can’t determine the winner of a rep by looking at dots on a screen. It’s also silly to use a firm 2.5-second barometer. What if the ball is out before that? What if the defender got chipped and double-teamed from a wide-nine alignment and still won in 2.51 seconds? Plus, what if he only wins because the blocker makes a mistake and leaves him unblocked? Is that a win? So much context is lost without having a human eye decipher the play.
For all the faults of PFF, this is an advantage they have over metrics like ESPN’s win rate. Since humans are charting the data, you don’t get the wonky results that can come from letting a computer decide how a player is performing. Whereas ESPN’s system thought Tomlinson played well in 2023, PFF charged Tomlinson with 51 pressures allowed, the most among interior offensive linemen.
There might be other players for whom ESPN’s win rate produces a more accurate result than PFF’s tracking. Neither system is perfect. Both have their strengths and flaws. In general, though, I’m not a fan of using GPS tracking to evaluate play in the trenches.
Defensive linemen: Sacks
Sacks are awesome. The more a defense gets, the better. However, they’re not a good tool for evaluating individual players.
The main issue is their erratic nature. So many things need to break right for someone to get a sack opportunity on a given play. The coverage needs to be good. The quarterback needs to hold the ball long enough and stay in a spot where the defender has a lane to reach him. On top of that, you’re competing against your teammates to get home first.
If you watch a player rep-by-rep, you realize how rare it is for all these things to happen simultaneously. It’s why even the most dominant defenders in football can only sack the quarterback around once every 30 plays. For instance, T.J. Watt had 19 sacks on 551 pass-rush snaps last year, which is one sack every 29 pass-rush plays.
Given how rare sacks are, it’s common for a handful of sack opportunities to go the wrong way over the course of 17 games, causing a player to lose a few sacks compared to what they probably should have had. This makes them look substantially worse in the sack column just because of a few fluky things that didn’t bounce their way. It’s why evaluating defensive linemen based on their entire body of work is wiser than solely using sacks.
Quinnen Williams had 12 sacks in 2022. This past season, he had 5.5. So, was he less than half as good? Absolutely not. When you watched him on a play-to-play basis, he was just as dominant, if not more so. The difference between 12 sacks and 5.5 sacks is almost unnoticeable when you’re evaluating a player on every snap throughout a season. Yet, the difference looks massive in the box score, sending a misleading message to casual fans who base their opinions solely on this one number.
Williams boosted his pass-rush win rate from 13.7% in 2022 (7th among IDL) to 19.5% in 2023 (3rd). He improved from 52 pressures in 2022 (8th) to 70 in 2023 (3rd). Williams was winning more often, yet got fewer sacks. Why? Well, sometimes, it just happens, because that’s how sacks work. It’s not always linked to a decline in individual performance.
Despite his personal sack dip, Williams had a greater overall impact on generating sacks for the Jets’ defense. His teammates benefited from the havoc he caused. After ranking 10th in sack rate (7.5%) in 2022, the Jets’ defense jumped to fifth (8.4%) in 2023. That rise can be traced directly back to Williams.
The more you think about it, the less it makes sense to base your opinions of a defensive lineman on his sack total. It’s such a tiny part of the pie.
If somebody sacks the quarterback once per game, they’re a superstar. That puts you on pace for 17 sacks, which is a guaranteed Pro Bowl appearance. But the league’s top defensive linemen play at least 700 defensive snaps each year – most from 800-900 and some over 1,000. In a given game, they’ll play 40-60 snaps. Should one play out of 40-60 plays tell the whole story about their impact on the game? Probably not, so 17 plays in a sample of 700-plus snaps shouldn’t tell the whole story of their season.
Sacking the quarterback is not a defensive lineman’s primary job as a pass rusher. Getting a sack is the ultimate prize, but his main job is to win as often as possible. Beat the man in front of you, cause some havoc, and good things will happen. Doing this at a high rate is the goal. If it results in some sacks for you, great. If not, you’re still affecting the game positively in some way.
Another thing we haven’t touched on yet is the fact that not all sacks are identical. Some aren’t worthy of much praise, if any. Players can get “cleanup sacks” on plays where the sack was actually created by someone else’s initial pressure. You can also get phony sacks by chasing the quarterback out of bounds on a scramble one yard behind the line of scrimmage, or by touching him after he stumbles. A sack isn’t necessarily an elite play by the player who registered it.
Wins/Pressures > Sacks.
Cornerbacks, linebackers, and safeties: Tackles
This might be the worst one on the entire list.
Let’s start with cornerbacks. Tackles mean absolutely nothing for a cornerback. Nothing. At. All. If I see another person cite a cornerback’s tackle total while making a comparison, I’ll… well, I’m not going to do anything, but I would be mildly annoyed.
What is a cornerback’s job? To avoid letting his man get the football. So, if he’s doing his job well, he should be getting fewer tackles, if anything. Sauce Gardner’s measly total of 57 tackles in 2023 is a testament to how elite he was at preventing targets in his direction.
If a cornerback gets toasted for a 50-yard bomb and brings the receiver down, he gets a tackle. By using tackles as a method to evaluate a cornerback, you’re basically saying that you think this was a good play.
It would make more sense to evaluate cornerbacks’ tackling activity by isolating a certain type of tackle, such as tackles for loss or run stops, but even then, not every player is being evaluated on the same plane. The number of tackle opportunities a cornerback gets is largely determined by his role. Does he play in the slot or on the outside? Slot corners are much more involved in tackling than outside corners. Does he play in a man or zone scheme? Does his team like to blitz the corners? These questions mean more in determining a cornerback’s tackle total than how good they are at tackling.
On the surface, it might seem like using tackles makes more sense for linebackers and safeties than it does for cornerbacks, but I argue that it’s useless for those positions, too. Everything we just said applies to linebackers and safeties. You can still get tackles for getting beat in coverage, and your role is still the primary factor in determining how many tackle opportunities you get.
If you’re a free safety in a scheme that asks you to play single-high most of the time, you’re not going to get a lot of tackles. If you’re a strong safety who usually plays in the box, you’ll get plenty of tackles. It’s not a fair comparison.
The same goes for linebackers. Fred Warner serves as a perfect example. Warner is unanimously considered the top linebacker in the NFL right now, yet he’s never ranked higher than 10th at the position in tackles in a single season. This past year, he was 17th.
There are a couple of reasons for this: 1) Warner is outstanding in coverage, so he deters a lot of potential targets in his area, and 2) his role in the 49ers’ defense asks him to blitz at a low rate and play with more depth than most linebackers, taking him away from the box and therefore reducing his frequency of tackle opportunities.
Getting a lot of tackles doesn’t mean you’re helping the team. Denver’s Alex Singleton was third among linebackers with 177 tackles last season. Yet, he had the most missed tackles (29), the third-most yards allowed in coverage (672), and tied for the most touchdowns allowed (5). Being a sieve in coverage is probably what netted him so many tackles.
This stat makes my blood boil. Please think of this humble stat nerd the next time you consider looking up a player’s tackle total. There are better ways to evaluate cornerbacks, linebackers, and safeties in the big 2024. Let’s evolve as a society.
