The consensus view on John Elway is clear. He was the greatest draft prospect ever, a league MVP, a two-time Super Bowl champion, a Hall of Famer, and one of the most clutch quarterbacks in football history.
But that’s not necessarily what the numbers say. In my quarterback ranking system, which rewards efficiency and longevity and adjusts for era, Elway only ranked as the 26th best regular-season quarterback of all time. If you’re so inclined, it’s not hard to find the numbers to argue that Elway – at least until Mike Shanahan returned to Denver as head coach in 1995 — was overrated. Consider:
- Over the first 10 years of his career, Elway threw 158 touchdowns and 157 interceptions.
- Elway never led the NFL in passer rating, completion percentage, touchdowns, yards per attempt, or Adjusted Net Yards per Attempt. Elway didn’t finish in the top ten in passer rating until his eleventh season in the league. In Net Yards per Attempt, Elway ranked in the top 10 just once from 1983 to 1994 (a first-place finish in ’87); in ANY/A, Elway’s only top ten finishes during his first ten seasons were in ’86 (10th) and ’87 (4th).
- Elway ranks fourth all-time in passing yards, but that’s because he ranks fourth in career pass attempts. While he led the NFL in passing yards in 1993, Elway only finished in the top five in passing yards four times in his career: 1985 (2nd), 1987 (4th), 1990 (5th), and 1995 (5th).
- Elway ranked 2nd in passing touchdowns in 1993, the only time he finished in the top 5 in that metric from 1983 to 1995. Despite throwing the fourth most pass attempts in NFL history, he ranks only 7th in passing touchdowns. In eight of sixteen seasons, including seven of his first ten years, Elway produced a below-average touchdown rate.
Here’s another interesting stat: from 1983 to 1992, the Broncos were slightly better on defense than offense. Over that time period, Denver’s Offensive SRS average was +1.01 while their Defensive SRS was +1.32. On average, the Broncos ranked 12th in points scored and 11th in points allowed. Those Denver teams are remembered as Elway’s teams — and perhaps rightly so — but the defense was just as valuable as the offense.1
Still, consider the following: There are 84 quarterbacks who started at least 80 games and threw 2,000+ passes during their age 23 through age 32 seasons. Elway had a perfectly league average Adjusted Yards per Attempt average during those years, which places him 67th among that group in AY/A+. There are 64 quarterbacks who met those same criteria since 1970: Elway ranks 47th among that group in ANY/A+, tied with Aaron Brooks, Jay Cutler, and Stan Humphries. Statistically, Elway was an average passer in his first ten seasons.
Jason Lisk did his typically excellent job examining Elway’s teammates three years ago. Lisk looked at the Approximate Value of Elway’s offensive supporting cast and found that he had the least support among modern Hall of Fame quarterbacks.2 Well, since Neil came up with the concept of True Receiving Yards and used TRY to measure the quality of each team’s receiving corps, I thought it would be interesting to measure the quality of Elway’s receivers (which includes his tight ends and running backs) using those concepts.
In Breaking Bad style, the open of this post had no explanation. Let me bring you up to speed: that picture represents the (weighted-average version of the) True Receiving Yards of Elway’s receivers during each season of his career. For over a decade, Elway was dealing with a pretty mediocre set of receivers, highlighted by Steve Watson, Vance Johnson, and Mark Jackson (if you’re nostalgic, we can include the third amigo, Ricky Nattiel). Once an older Elway was playing with Shannon Sharpe, Rod Smith, and Ed McCaffrey, his numbers improved significantly.
The graph below shows the quality of the receivers that Elway, Dan Marino, and Jimmy Kelly played with in their age 23 through age 38 seasons.3 As you can see, Elway was playing with a much weaker set of targets than his draft classmates:
Neil mentioned the huge caveat in using TRY to measure the quality of a set of receivers — their numbers are impacted by the quality of the quarterback, so measuring the quarterback’s targets when discussing quarterbacks involves some circular reasoning. But when we’re talking about three Hall of Fame quarterbacks, this concern is muted, and we can see that Marino enjoyed playing with Mark Clayton and Mark Duper, Kelly had Andre Reed and (for three good years) James Lofton, and Elway…. well, his targets (until he turned 35) show up as a distant third.
What if we compare Elway’s targets to those of the two Hall of Fame quarterbacks he met in the Super Bowl?
Montana, even before he got Jerry Rice (at age 28), was throwing to players like Dwight Clark, Freddie Solomon, and Roger Craig. Favre had a rotating group of players, but his targets rate as much better than Elway’s, too. Some would say that Favre “made” less-than-star receivers into star receivers; if that’s the case, you would have to conclude that Elway was playing with receivers a couple of tiers below “less-than-star” or that Favre was so much better than Elway that he was able to do so much more with the same.
Next, let’s compare Elway’s receiving corps to two other star quarterbacks of the ’90s: Troy Aikman and Steve Young. As before, we’re comparing the quality of the receivers based on the quarterback’s age, not any specific year:
Much has been made about how Aikman struggled his first two seasons: perhaps that’s because his number one receiver was Kelvin Martin. Similarly, Young struggled in Tampa Bay — perhaps traced to that dot in the bottom left of the picture — and then was blessed with outstanding receivers the rest of his career.
We see more of the same — when Esiason was playing at MVP-levels, he had great receivers. Moon was blessed with strong receivers for almost his entire career. Elway, during his prime, was a one-man show when it came to his team’s passing game.
The proper response to those who bash Elway isn’t to argue that Elway’s numbers from 1983 to 1994 were actually very good; the appropriate answer is that he was plagued with mediocre teammates. I’m not sure how bad his receivers need to have been to “make up” for Elway’s league-average numbers, but consider the final four years of his career. From 1995 to 1998, Elway played with the best teammates of his career. And he put up outstanding numbers despite playing at an age where most quarterbacks are on a steady decline. Elway was one of the best 35-to-38-year-old quarterbacks in NFL history. That’s an irrefutable statement, and considering his two rings during this time period, one could argue that he was the best ever over those ages (and with those rings he even has a leg up on where Peyton Manning and Tom Brady will finish).
In some ways, it feels like people have tried to invent reasons to explain why Elway was so good despite his merely solid numbers.4 I think the more intellectually honest explanation is to understand that the NFL is a team game, and the quarterback is only responsible for a portion of his team’s offensive production. Compared to the other star quarterbacks of his era, the evidence indicates that Elway had the least support from his teammates. In the ’80s, the only skill-position teammate of his to make the Pro Bowl was running back Sammy Winder.
Once Elway started playing with star teammates, his production was outstanding, and even better once you consider his age. His raw numbers pale in comparison to players like Montana, Young, Favre, and Marino. I’m not sure if he was a better player than any of those great quarterbacks, but I do think much of the gap is best explained by each player’s supporting cast, and not by each player’s ability.
- On the other hand, it’s worth pointing out that the ’83-’92 Broncos won more games than their Pythagorean record would have predicted, so perhaps Elway was responsible for more wins than his passing numbers would indicate. [↩]
- Actually, Lisk found that Elway had the second worst supporting cast, behind Brett Favre, who Lisk included in the list on principle. However, as Lisk noted, that was going to change over time as Favre’s teammates accumulated more AV. [↩]
- Note that I included all seasons where the quarterback led his team in pass attempts. I then used a weighted-average on the season level of all receiving yards, so it’s not a game-by-game average of those games when these quarterbacks started. This is most clearly seen in the dips for Marino at age 26 and Elway at age 27, which was the 1987 season. The strike makes it appear that those teams’ supporting casts were weaker, since 20% of the games were played with replacement players. [↩]
- Scott Kacsmar has done a great job dispelling the myth that John Elway holds the all-time record for 4th quarter comebacks. Elway did not record 47 4th quarter comebacks; in reality, he had 34 4QCs and 46 game-winning drives, marks which place him behind Peyton Manning and Dan Marino on both counts. More importantly, when trailing in the 4th quarter by one score with possession, Elway’s teams went 34-46-1 in those games, nearly identical (but slightly worse than) the records of Marino and Manning. [↩]