We need to re-think what a ‘good’ completion percentage is, and other stats

For as long as NFL stats go back, 77 teams have completed at least 70-percent of their passes (minimum of 45 attempts) over the first three games of a season. This would account for approximately one such team per season dating back to the oldest such example I can find, the 1953 Cleveland Browns.

That team completed 56 of 79 passes (70.9-percent) with four touchdowns, one interception and 11 sacks taken.

The quarterback for the Browns that year was Otto Graham, the Patrick Mahomes of the 40s and 50s, if we can safely assume that like Graham, Mahomes will lead the NFL in passing yards five times, completion percentage four times and passer rating five times, including all three of those categories in 1953 — plus also leading the league with 10.6 yards per attempt, another thing Graham did four times.

Graham completed 64.7-percent of his passes that season and that was in a way like throwing 55 touchdowns. The things Graham did were not rare, they were unprecedented.

The NFL wouldn’t see another team with 70-percent completions through three games for another seven years, when it would again by Cleveland, again under head coach Paul Brown, but now the quarterback was Milt Plum; not quite Otto Graham, conceivably the best player in league history. But Plum did lead the NFL in completion percentage three times in a row, before going to the Detroit Lions in 1962 and seeing his stats disintegrate; in 1963, Plum went 27-of-77 (35-percent) with two touchdowns and 12 interceptions.

Had Brown built a repeatable system? Could he at least repeat it, if not the rest of the NFL?

No.

I’m not an Xs and Os expert, nor can I trace the history of Xs and Os through Brown, Sid Gillman, Don Coryell, Bill Walsh, West Coast Offense, play action passing and all that. But I can trace the numbers and the numbers say that even if the league doesn’t have a “completion percentage problem,” we do have to at least acknowledge that quarterback stats have evolved way past normal stats inflation — if that even exists — and must reassess what things like “he’s completing 70-percent of his passes” and “he has a passer rating of 100” really mean.

It does not necessarily mean that those quarterback are succeeding enough relative to their peers.

No teams would post a completion percentage above 70 over the first three games from 1961 to 1972 and then in ‘73, the Kansas City Chiefs did it with Mike Livingston and Len Dawson. I can’t trace Xs and Os back to the Browns, but I can trace Dawson: he was a backup in Cleveland in 1960 and 1961. Paul Brown had started Plum over Dawson, who led the NFL in completion percentage in every season between 1962 and 1969 with the exception of 1963 and is a Hall of Famer who belongs in the same conversations as Graham.

In none of the years between 1964 and 1969 did Dawson even hit 60-percent completions. But he led the NFL.

Overall, 70-percent completions through three games happened twice in the 70s, five times in the 80s (never twice by the same team or QB) and seven times in the 90s. Again, this wasn’t something done by the same quarterbacks or teams or system and it didn’t even necessarily mean the results were good. In 1993, the Seattle Seahawks posted a completion percentage of 70 with rookie Rick Mirer, but he had thrown one touchdown, five interceptions and was sacked 14 times in three games.

Then beginning in 2000, the St. Louis Rams would first join the party. Coming off of one of the most dominant Super Bowl seasons in league history, the Rams posted a completion percentage of 70 or better through three games in three straight seasons thanks to Kurt Warner and Mike Martz and the all-pros they had. Similar to Mirer, it wasn’t always good: in 2002, Warner had one touchdown and seven interceptions through three games.

But the Rams did heat things up for the league offensively.

In the 2000s, we saw 15 teams join this group, including the regular season undefeated New England Patriots in 2007 and the Drew Brees Saints in 2008, perhaps the two most important influences for the decade to come. As of 2020, Brees has completed at least 70-percent of his passes through three games five times.

More than we had seen from the entire league combined over the 50s, 60s and 70s.

As we know by now, Brees doesn’t need to stop at three games to keep his completion percentage above 70. In 2009, Brees became the fifth QB to complete 70-percent of his passes in a season, joining Joe Montana, Steve Young, Sammy Baugh (Hall of Fame all of them) and Ken Anderson, the 1981 NFL MVP. Anderson completed his 70.6-percent in 1982, when there were only nine games.

You no longer need to be a Hall of Famer or an MVP to complete 70-percent of your passes.

In 2016, Sam Bradford completed 71.6-percent of his attempts for the Minnesota Vikings, the fourth best completion percentage in history. The other five in the top six belong to Brees. Kirk Cousins joined the 70-and-up group in 2018, with Derek Carr and Ryan Tannehill added to the list in 2019.

In the last two seasons, Matt Ryan, Jimmy Garoppolo and Carson Wentz all came within a few completions (or dropped passes) from joining them. Marcus Mariota, now backing up Carr with the Las Vegas Raiders, completed 68.9-percent of his attempts in 2018, one season before being benched for Tannehill.

If things don’t work out for Mariota or Carr with the Raiders, at least they could try going to the Vikings to backup Cousins. But the question I’ll be asking myself over the next 14 weeks is: how many more names will join the 70-percent club?

And how many of them will be on the precipice of being benched, if not having been benched during the season they were completing 70-percent of their passes?

Through three games for all but two teams (the Baltimore Ravens are completing 78-percent, the Kansas City Chiefs are only at 64.6-percent but that’s destined to go up), there are 10 quarterbacks completing at least 70-percent of their passes.

That’s more than the 50s, 60s, 70s AND 80s combined.

This includes the two quarterbacks getting the most attention for MVP through three weeks (Russell Wilson and Josh Allen) but also Teddy Bridgewater, Gardner Minshew, Ryan Fitzpatrick, Carr and Philip Rivers, who leads the league at 78.3-percent; Rivers may go to the Hall of Fame one day but at 39 and with three touchdowns against three interceptions, he doesn’t quite give the feelings for the Colts of a Peyton Manning type.

Rivers’ mark is the third-best in history through three games after Brees in 2018 (80.6-percent) and Tom Brady in 2007 (79.5-percent). Wilson’s 76.7-percent is the fifth best, right behind Alex Smith in 2017, during a season in which everybody knew he’d be traded after it was over if he didn’t win the Super Bowl.

Derek Carr is a regular on this list. Gardner Minshew has already appeared on it twice during his two years in the league, having completed 73.86-percent of his passes through three games in 2019 and 73.83-percent of them in 2020. Eli Manning in 2018 is on it, his last season in the league. So is Jacoby Brissett, the quarterback benched for Rivers in Indianapolis.

And now Jared Goff has completed at least 70-percent of his passes through three games in three of his four seasons under head coach Sean McVay.

What we have is a combination of talent and opportunity (scheme evolution, rule changes, better coaching and preparation for the league, potentially even less talent on defense, a waning desire to run the football) that has led to an NFL where completing 70-percent of your passes and posting a rating above 100 or even a Y/A above 8 is not necessarily indicative of quality play or at least is not indicative of quality play relative to the other 31 starting quarterbacks in the NFL.

Two games into his career, LA Chargers rookie Justin Herbert has completed 69.5-percent of his passes with 7.8 Y/A and a passer rating of 90.5. He has been deemed “not ready” by head coach Anthony Lynn and there’s a desire to get veteran Tyrod Taylor back as soon as possible from a puncture lung. A passer rating of 90.5 ranks Herbert 23rd in the league today but would have stood as the ninth-best in the NFL in 2008.

And Herbert’s completion percentage of 69.5 would be more than 2 points better than NFL leader Chad Pennington in 2008.

In 2010, within the previous decade, Herbert’s numbers, if held for the majority of a season, would rank first in completion percentage, sixth in Y/A and his rating would have put him directly in the company of Brees, Manning and Ryan that season, all 2010 Pro Bowl quarterbacks.

And yet nobody would have watched Herbert’s last two games and thought, “This is a top-10 quarterback right now.” Optimism for that in a few years? Absolutely. But if Herbert had posted those numbers as a 22-year-old “not ready” rookie in 2010, it would not be too dissimilar from seeing Mahomes dominate in 2018 when the tide had raised all the boats.

The NFL completion percentage so far in 2020 is 66.4-percent. The average passer rating is 96. That’s the average.

Aaron Rodgers posted a passer rating of 95.4 ... last season.

Take a look at this player: 65.8-percent of his passes are completed, he averages 7.6 yards per attempt, he posts a passer rating of 95.5. Today, those numbers would essentially be average at best. They’re remarkably close to the 2018 campaign of Mitchell Trubisky.

In 2002, it won Rich Gannon the MVP award and helped send the Raiders to the Super Bowl with the most potent offense in the league following the Warner era in St. Louis. And almost 20 years later, Jon Gruden returned to the league to sort of do something similar with Carr, except the bar has been raised exponentially.

I realize that it is early in the season and looking at a three-game sample size is dangerous if you’re trying to pace it out for a full year, but even looking at all the other small three-game sample sizes in league history, the last two or three years have stood out in ways that no other era can compete with. And I think having this in mind will be valuable when discussing quarterbacks and statistics moving forward.

On Sunday, the LA Rams lost to the Buffalo Bills 35-32 in large part due to Josh Allen completing 24 of 33 passes for 311 yards. His 72.7-percent completions was the highest mark he’s had through three games and he’s sitting on 71.1-percent for the year, less than 10 months from the 2019 season, in which he completed 58.8-percent of his throws. Allen was meant to be one of the least accurate first round quarterbacks in history and before he’s made his 32nd start, he’s shamed the numbers of Otto Graham.

This is not to be unexpected with the context of 70 years of NFL evolution. But what about the last five years of evolution? What passing offenses are capable of seems to be advancing faster than we can keep up with anymore.