The NFL stats that could blind you: Can you pass Barnwell’s test?
The star Cleveland wideouts’ numbers are eerily similar this season, plus numbers that really show the Patriots’ defensive dominance….
Who doesn’t love a good blind statistical comparison? As I look through the numbers and put together my columns for ESPN each week, I inevitably run into simple comparisons between players or teams that surprise me. Some of those end up in columns or tweets. Others get penciled in as future ideas, most of which get written off when the comparisons no longer apply.
Today, though, I’m throwing out an entire column of blind comparisons. I’ll list two players or teams that have similar statistics by several key measures and then reveal who they are. In some cases, we’ll be making comparisons between teams and players from 2019 and the past; in others, I’ll contrast players and teams from this current season to contextualize their ongoing campaigns. In fact, let’s start with one of those:
Jump to an interesting blind comp:
‘Not bad for a running back’
Inside the most dominant defense … ever?
One struggling QB, one improving QB
Can the Browns really win like this?
Serious MVP candidate | Underrated WR
Player A: 61.8% completion percentage, 7.0 yards per attempt, 7 INTs
Player B: 62.4% completion percentage, 7.0 yards per attempt, 7 INTs
Let’s start with a straightforward comparison. Player A is Jared Goff‘s 2019 season so far, with the Rams quarterback hitting these relatively mortal numbers only after getting to face the free parking space that is the Falcons’ defense last Sunday. Goff should continue his return to form against the Bengals in Week 8, but the former first overall pick will have to contend with the likes of the Steelers, Bears and Ravens after Los Angeles’ subsequent bye.
Player B is Bills quarterback Josh Allen, who has unquestionably improved as a passer after a moribund rookie season. If anything, he deserves a bit of a boost given that he has played the Patriots, which is his only truly dreadful performance of the season. The Bills have surrounded Allen with a better infrastructure and receivers than the replacement-level talents he worked with in 2018, and his numbers have rebounded by asking him to throw deep less frequently. Throw in Allen’s distinct advantage as a runner (190 rushing yards and 13 first downs, to 16 yards and three first downs for Goff) and you could very well argue that Allen has been the more productive player this season.
Of course, we’re telling different stories about these players because they look to be on different paths. Goff, who played like a franchise quarterback for most of last season, appears to be taking steps backward. Allen, who looked to be in over his head as a rookie, is on the way up. They’re both roughly in the same spot as below-average quarterbacks, with Goff and Allen ranking 27th and 29th, respectively, in Total QBR. Goff’s track record of NFL success suggests that he’s more likely to improve, but who can be sure?
Player A: 21 tackles, 4 sacks, 20 QB knockdowns, 3 tackles for loss
Player B: 20 tackles, 4 sacks, 14 QB knockdowns, 3 tackles for loss
Player A is Texans star J.J. Watt, who flew into the league lead in quarterback hits after knocking down Jacoby Brissett six times on Sunday. Watt isn’t having a monster sack season by his standards, but this is the closest we’ve seen him get to his 2012-15 peak, when he racked up 190 quarterback knockdowns over a four-year span, 90 more than any other defender over that same time frame. The future Hall of Famer is on pace to finish with 46 hits this season.
You might already have been able to guess Player B’s identity, but it’s proof the apple doesn’t fall far from the tree. Steelers pass-rusher T.J. Watt is tied for second in the league with 14 knockdowns and was tied with his older brother before Pittsburgh’s bye week. The fellow Wisconsin product broke out with a 13-sack, 21-knockdown season in 2018, and he’s tracking to top that knockdown total by Week 10. T.J. has helped keep the Steelers afloat without Ben Roethlisberger, as Pittsburgh’s pass rush ranks third in adjusted sack rate.
Player A: 29 catches, 436 receiving yards, 55.8% catch rate, 11.0 air yards/target, 212 routes run
Player B: 25 catches, 439 receiving yards, 54.3% catch rate, 10.0 air yards/target, 210 routes run
These are two players who find themselves in very similar situations. Player A is new Browns wide receiver Odell Beckham Jr., who hasn’t quite hit the lofty heights fans expected after the stunning trade that sent him to Cleveland in March. Player B is teammate Jarvis Landry, who has virtually identical statistics after six frustrating games.
Beckham has 52 targets to Landry’s 46, but we’re basically dealing with something close to a 50-50 split in terms of usage between the two former LSU stars.
Is this the best way for the Browns to use their two expensive wide receivers? It’s hard to say. I expected that Landry might see targets slightly more frequently, but that those targets would come on shorter passes than the throws Beckham would see. Instead, we’re really seeing the two wideouts playing around the same depth, although Landry has made 19 catches out of the slot, and Beckham has only four.
I wonder if rookie head coach Freddie Kitchens plans to make changes to his offense coming out of the bye, although Cleveland’s 10-game close to the season starts with the juggernaut Patriots defense on Sunday.
Player A: 127 carries, 618 rushing yards, 7 rushing TDs
Player B: 135 carries, 602 rushing yards, 6 rushing TDs
Player C: 30 catches, 306 receiving yards, 5 receiving TDs
Player D: 35 catches, 305 receiving yards, 2 receiving TDs
This is another one you might be able to see through. Player B is Cowboys running back Ezekiel Elliott, and Player C is Colts receiver T.Y. Hilton, but Players A and D are the same guy: Christian McCaffrey, who is engendering serious MVP consideration as he has carried the Panthers to a four-game winning streak.
The 2017 first-round pick has fielded more than 56% of Carolina’s offensive touches so far while playing 96% of the offensive snaps. Though Kyle Allen has impressed since taking over for an ailing Cam Newton, Allen has thrown only six passes for 26 yards without his star halfback on the field.
Would the combination of Elliott and Hilton be worth an MVP? I’ll get into a longer discussion about McCaffrey’s MVP chances when I file my midseason awards ballot in a couple of weeks, but McCaffrey is on pace to approach nearly 2,500 yards from scrimmage over a full season. I don’t think I would say either Zeke or Hilton are having particularly great seasons by their own standards, so while the combination of the two is pretty impressive, McCaffrey probably has to be the best rusher in the league before he attracts significant MVP attention.
You could make a strong case for Dalvin Cook as the league’s top runner this season, leaving receiving out of the equation.
Player A: 57.1% completion percentage, 6.8 yards per attempt, 3.4% INT rate, 77.1 passer rating
Player B: 60.1% completion percentage, 6.7 yards per attempt, 2.8% INT rate, 77.9 passer rating
This one’s just here to contextualize how much the NFL has changed over the past 20 years, and how important it is to put stats in context. Player A is the collected passing performance of the NFL during the 1999 season; though that includes work by backups and the occasional trick play, this is essentially what average quarterback play looked like 20 years ago.
Player B is Andy Dalton‘s 2019 season so far. I understand if you haven’t really been paying close attention to an 0-7 Bengals team, but Dalton ranks 28th in passer rating and Total QBR. An average quarterback from 1999 — a passer like Steve McNair or Jeff Garcia, by the numbers — would be barely playable by the numbers in 2019.
Kurt Warner’s debut season as a starter came in 1999 when the future Hall of Famer took the league by storm and won MVP. His passer rating that year was 109.2. Three quarterbacks have a better passer rating this season than Warner had then, and one of them is Kirk Cousins. The No. 2 quarterback that year by passer rating was Steve Beuerlein, who was nearly 15 points behind Warner at 94.6. Fourteen quarterbacks have a better passer rating this season than the former Panthers starter. The game is very different.
Player A: 243 catches, 379 targets, 14 drops, 3,681 receiving yards, 15 TDs
Player B: 243 catches, 362 targets, 14 drops, 3,153 receiving yards, 15 TDs
Let’s contrast a superstar everyone takes for granted and one who doesn’t always get the same sort of recognition. Player A is Falcons receiver Julio Jones‘ cumulative statistical line over the past three seasons. While there was a stretch of time in which the Falcons star seemed bizarrely incapable of crossing the goal line, a thankfully healthy Jones has been as dominant as anybody in football over the past three seasons, averaging 94.3 receiving yards per game. About the only hole you poke in Jones’ game — and it might just be Atlanta’s philosophy in keeping him on the field — is that the star wideout typically plays about only 75% to 80% of the Falcons’ offensive snaps when most No. 1 receivers play closer to 95% of their team’s snaps.
Player B is a guy who gets fantasy recognition without necessarily getting best-wideout-in-football hype. It’s the 2017-19 combined line for Chargers receiver Keenan Allen, whose numbers put him in a similar ballpark to the Falcons’ star. Jones does average right around 1 more yard per target than his Chargers counterpart, which is valuable, but Allen’s catch rate is about 3 percentage points higher. Jones is the better receiver, but there’s less of a difference between the two than you might think.
Player A: 14 games, 238 carries, 1,341 rushing yards, 5.5 yards per carry, 61 rushing first downs, 12 TDs
Player B: 14 starts, 230 carries, 1,271 rushing yards, 5.5 yards per carry, 75 rushing first downs, 8 TDs
This one might not be obvious at first glance. Player A is Adrian Peterson‘s rookie season. Heading back to 2007, the seventh overall pick made an immediate impact as a runner, racking up four 100-plus yard games in his first five outings. In November of that season, Peterson set the single-game NFL rushing record by going off for 296 yards and three touchdowns in a victory over the Chargers. The only other running backs who carried the ball at least 10 times in 2007 and are still in the league besides Peterson are Frank Gore and Darren Sproles.
Player B is Lamar Jackson‘s production as a runner since entering the league. Jackson was used in sub packages while backing up Joe Flacco last season, but after 14 starts, his numbers are eerily similar to Peterson’s debut campaign. Peterson has more touchdowns, and Jackson will never get the running volume to produce a 296-yard game, but the Ravens’ starter has moved the chains more frequently and been a more consistently impactful runner.
Oh, there’s something else he can do …
Player A: 61.0% completion percentage, 7.4 yards per attempt, 6.39 adjusted net yards per attempt
Player B: 65.0% completion percentage, 7.2 yards per attempt, 6.35 adjusted net yards per attempt
Player A is Lamar Jackson’s performance as a passer over the past two years. Player B is Matthew Stafford‘s performance over that same time frame, and Stafford has the largest cap hit of any player in football.
Adjusted net yards per attempt weighs touchdowns and interceptions in a more realistic way than passer rating while also incorporating sack yardage; after you do all that, Jackson has been just as effective of a passer as his far more expensive counterpart. As Jackson said earlier this season, “not bad for a running back.”
Player A: 228 completions, 373 attempts, 61.1% completion percentage, 2,764 passing yards, 7.4 yards per pass attempt, 16 TDs, 8 INTs, 89.3 passer rating
Player B: 198 completions, 360 attempts, 55.0% completion percentage, 2,673 passing yards, 7.4 yards per pass attempt, 17 TDs, 9 INTs, 84.2 passer rating
I know what I said a minute ago about comparing quarterbacks across eras. Indulge me for a second. Player A is Lamar Jackson‘s passing numbers across his first 14 starts. Player B is Steve McNair‘s passing numbers across his first 14 starts.
The Oilers started McNair only six times in his first two seasons before moving to Tennessee and handing him the full-time job, so these numbers actually stretch into McNair’s third season. By the time McNair started his 14th game in the league, Jackson could have as many as 26 starts under his belt. They’re certainly not the same sort of quarterback, but it’s fun to see the similarities between a former Ravens standout and Baltimore’s current star quarterback.
Team A: 66 possessions, 0.86 points per drive, 18.2% of drives ending in turnovers, 50% three-and-out rate
Team B: 63 possessions, 0.83 points per drive, 20.6% of drives ending in turnovers, 49.2% three-and-out rate
Let’s make a defensive comparison here. Team A is the season-to-date output for the San Francisco 49ers, who have allowed 10 points over the past three weeks while comfortably handling the Browns, Rams and Washington. The Steelers are the only team to score more than 17 points against the 49ers this season, and that came in a game in which Kyle Shanahan’s offense turned over the ball five times. Richard Sherman & Co. have absolutely stifled opposing passing attacks.
That’s why it’s so interesting that these numbers compare so closely with the last five weeks of the 2013 regular season for the Seattle Seahawks, who are Team B. That was a defense which will likely send three players to the Hall of Fame in Sherman, Earl Thomas and Bobby Wagner, let alone a second tier with guys such as Bruce Irvin, Cliff Avril, Michael Bennett and Kam Chancellor.
After this five-game stretch, the Seahawks rode that legendary defense to a blowout victory over the Broncos in Super Bowl XLVIII. We’re still a long way away from that for the 49ers, but for a team that intercepted two passes across all of 2018, it’s incredible to think that we could make viable comparisons to the Legion of Boom after six games this season.
Player A: 561 pass attempts, 11 drops
Player B: 109 pass attempts, 10 drops
This is a reminder of just how important it is to end up with the right people around quarterbacks. Player A in this grouping is Jared Goff‘s 2018 season, which the Rams quarterback spent under the tutelage of Sean McVay and with a bevy of talented receivers. Player B is Josh Rosen‘s 2019 season with the Dolphins, in which his receivers have dropped a staggering 9.2% of his passes.
To put that in context, no other quarterback with 100 pass attempts or more this season has been subject to a drop rate of more than 5.5%. The last time a quarterback struggled with a drop rate this high over a 100-plus attempt season was Blaine Gabbert in 2012.
Rosen’s numbers through two seasons are awful. There’s no denying that. It’s also difficult to imagine a quarterback who has been dealt a worse hand in terms of supporting casts than the former Cardinals first-round pick, who has played behind the league’s worst offensive line in each of his two pro seasons. After this season, Rosen will likely be replaced by the first overall pick in the 2020 draft and either consigned to the bench or traded as a distressed asset to another team, likely as a backup. Goff, who spent a year flailing under the final year of the Jeff Fisher administration in Los Angeles, can probably sympathize.
Player A: 52.3% completion percentage, 4.2 yards per attempt, 9.2% INT rate
Player B: 50.8% completion percentage, 5.0 yards per attempt, 7.8% INT rate
Player A is quarterback Nathan Peterman, who was badly overmatched in his four starts with the Bills from 2017 to ’18. The fifth-round pick posted a passer rating of 32.5 on 130 pass attempts; when you adjust for era using Pro Football Reference‘s index statistics, Peterman is the second-worst quarterback in the history of football with 100 or more pass attempts.
1:13
Ryan Clark, Jack Del Rio and Dianna Russini break down why Bill Belichick has been so successful as the Patriots’ coach.
Player B isn’t really one player; it’s the collected line of all the quarterbacks who have faced the New England Patriots‘ defense this season. Those quarterbacks have combined to post a passer rating of 35.6, just 3.1 points better than Peterman’s mark. I get that the Patriots haven’t played a tough slate of opposing quarterbacks, but they’ve still gone up against veterans such as Ben Roethlisberger and Colt McCoy and highly touted young players such as Sam Darnold and Daniel Jones. Bill Belichick’s defense has reduced them to collectively performing about as poorly as the worst quarterback of the past 80 years.
Team A: 3 total TDs, 18 INTs, one passing TD
Team B: 4 total TDs, 18 INTs, one passing TD
You can probably sense a theme here. Team A is the 2019 Patriots defense and what it has allowed in the first seven games. The Pats have allowed as many touchdowns on offense (three) as they have on defense, which would be crazy if it weren’t for the fact that they’ve scored as many touchdowns on defense (also three) as they’ve allowed.
Team B is a seven-game run from Weeks 5-11 from the 1969 Minnesota Vikings‘ season. It’s the last time a defense posted a similar-or-better touchdown-to-interception ratio to this Patriots team over any seven-game stint in any season. For whatever you want to say about their quality of opposition, there have been plenty of great defenses that went up against terrible quarterbacks over the past 50 years. None has been as dominant as the Patriots.
Of course, the Vikings also pulled this off in an era in which it was much easier to post these sorts of numbers. In 1969, opposing offenses averaged 20.9 points per game and threw 1.4 interceptions per contest. In 2019, NFL teams average 22.5 points per game while throwing interceptions nearly half as frequently, averaging 0.8 picks per game. There’s a strong argument that this is the best defensive run we’ve ever seen from any team in post-barnstorming football history.
Team A: 2.94 standard deviations above the mean, 36.4 points per game
Team B: 3.05 standard deviations below the mean, 37.1 points per game
Let’s finish with more of an indirect comparison of greatness. Team A is the 2007 Patriots offense and its points scored (excluding defensive and special-teams scores) through the first seven games of the season, when they laid waste to the NFL. Teams eventually caught up in the second half and began to slow down the Pats, but after a 49-28 dismantling of the Dolphins in Week 7, the Pats had scored 255 points, 39 more than any other team and nearly 120 more than the league average.
By standardizing the Pats’ performance, we can compare them across eras. Three standard deviations above the mean is the simple definition of an outlier, so the 2007 Patriots come very close to that standard.
As you can probably guess, Team B is the 2019 Patriots. By the same measure, they’ve allowed a total of 30 points across touchdowns scored by the opposing offense, field goals and extra points. The second-placed 49ers have allowed 58 points in six games; prorating that to seven games gets them to 67.7 points. I’ve gone through the other teams who have had a bye and similarly adjusted their totals.
The Patriots’ defense is 3.05 standard deviations below the mean in terms of points allowed. The cool thing about standardizing scores is that you can easily translate them across different measures, so given that the Pats’ defense is 3.05 standard deviations below the mean in 2019, we can figure out what an offense that is 3.05 standard deviations above the mean would have looked like in 2007. As it turns out, if an offense were 3.05 standard deviations above the mean over the first seven games of 2007, it would have scored 37.1 points per game, a little under one point per contest more than the Patriots actually did during their incredible start to the season.
If all of that’s too math-y for you, let me put it as simply as I can: The 2019 Patriots defense has been more dominant than the 2007 Patriots offense through seven weeks.
About The Author
