Wednesday, December 26, 2007

Has NFL Quarterbacking Gotten Better or Worse Since 1970?

In the November 19 issue of Sports Illustrated, Peter King writes, "Ripping NFL quarterbacking has become a favorite pastime this year." He acknowledges that several teams are struggling to find a decent starter but declares, "there will never be 32 premier passers at one time. And in truth, the overall quality at the position has improved steadily since 1970." King then cites the fact that the average NFL passer rating has increased from 65.6 in 1970 to 73.7 in 1980 to 77.3 in 1990 to 78.1 in 2000 to 81.1 (at the time he wrote his article) in 2007. For some reason, King randomly picks 1980, lists the eight passers who threw for the most yards that season (Dan Fouts, Brian Sipe, Archie Manning, Tommy Kramer, Steve Bartkowski, Lynn Dickey, Ron Jaworski and Doug Williams) and declares that the eight passers who had thrown for the most yards through week 10 of this season (Brett Favre, Tom Brady, Tony Romo, Carson Palmer, Drew Brees, Peyton Manning, Matt Hasselbeck and Donovan McNabb) have a more impressive collective resume; King notes that there is just one Hall of Famer in the first group (Fouts) compared to "at least three" in the second group (presumably King means Favre, Manning and Brady). After that, King mentions that Hasselbeck has posted a better completion percentage and yards per game average than Jim Hart, a Pro Bowl quarterback from the 1970s. King concludes by pointing out that Atlanta and Chicago, two teams that have unsettled quarterback situations now, were no better off in 1972 when Bob Berry and Bobby Douglas/Kent Nix were their respective quarterbacks.

It is hard to imagine a less well reasoned article about this subject than the one that King put together. Let's take it (apart) from the top. There is a difference between each team having a competent starting quarterback and having "32 premier passers in the league"; by definition, there will never be more than five or 10 "premier" passers, even if the overall quality of play is in fact improving: "premier" simply means the best of the best at a given time. A record 61 different quarterbacks have started at least one game this year; some of that is due to injuries, of course, but many teams are having trouble finding one player who can do an even semi-competent job at the most important position in the sport. Whether or not this is a golden age in terms of "premier" quarterbacks, there are certainly a lot of bad starting quarterbacks in the NFL.

King's method for evaluating the evolution of quarterback quality is bizarre, to say the least. Passer ratings compiled prior to 1978 should never be compared to the ones put up in what I call the "Liveball" era (borrowing a concept from baseball); in 1978, the five yard contact rule was enacted and offensive linemen were allowed to extend their arms and open their hands while blocking. Those changes led to dramatic improvements across the board in passing statistics. Another change since that time is the proliferation of the so-called West Coast Offense and its various offshoots that emphasize using the short passing game in place of the running game; quarterbacks in such systems are going to have high completion percentages and low interception rates (because they are not making risky, long throws) and thus their passer ratings will be better than those put up by gunslingers like Terry Bradshaw or Joe Namath. King is oddly infatuated with yardage statistics but after the rules affecting the passing game were liberalized teams sensibly chose to pass the ball more often and thus gain more passing yards; teams gained 181.1 passing yards per game in 1970 (according to King's chart), 214.1 passing ypg in 1980 and 214.6 passing ypg through week 10 in 2007. Obviously, the bulk of the increase in passing yardage happened right after the 1978 rules changes--but King never mentions those changes in his article, leading the reader to believe that there has been some quantum increase in quarterback quality.

It is also not clear why King randomly selected 1980 or why he chose to list the year's yardage leaders as opposed to its passer rating leaders. Yardage is hardly the best way to evaluate a quarterback's greatness. Although comparison of passer ratings between eras can be misleading, there is nothing wrong with comparing the career accomplishments of 1980's passer rating leaders with those of 2007's passer rating leaders (the passer ratings themselves have skewed upwards but it is valid to look at who were the top QBs in different eras and make subjective determinations about which group of players is more distinguished). The top eight 1980 passers by passer rating were Sipe, Jaworski, Vince Ferragamo, Steve Bartkowski, Joe Montana, Dan Fouts, Gary Danielson and Manning. Montana and Fouts are Hall of Famers, while Sipe won the 1980 NFL MVP. By using yardage as his metric, King "conveniently" left Montana out of the discussion of 1980 quarterbacks. In any case, what is so special about 1980? If King had chosen 1981 and looked at the top eight passers by rating then he would have found three Hall of Famers (Montana, Fouts, Terry Bradshaw); if King had chosen 1979 and looked at the top eight passers by rating he also would have found three Hall of Famers (Bradshaw, Fouts, Roger Staubach). It looks like King deliberately chose one year and one category in order to "prove" that there are more great quarterbacks today than there were in previous eras. It should also be mentioned that Hall of Fame status is a somewhat subjective barometer of greatness; Ken Anderson is certainly a worthy HoF candidate and if he were inducted at some point then the 1981 leaders would include four HoFers.

The Hasselbeck-Hart comparison is completely out of left field (pardon the mixed-sports metaphor). What do those two quarterbacks have in common? Even if Hasselbeck is better than Hart, what does that prove? Furthermore, King only compared them in terms of their completion percentages and their passing yards per game, two metrics that clearly favor the post-1978 passer. The references to the 1972 Falcons and Bears are also strange. What light can those situations possibly cast on the overall level of quarterbacking at that time?

King may be right that NFL quarterbacking is better now than it ever has been but he certainly did not prove it in his article, which is a textbook example of misuse of statistical data.

4 comments:

madnice said...

I dont care about the stats. There are so many bad QBs in the NFL its unwatchable. Peter King has no business writing any column. This is the same guy who said Moss to the Patriots is a mistake because of his attitude and hes a step slow. How does he know? Has he played against him? He also said that after years of not voting for Monk for the HOF that after spending time with Gibbs this past year and talking about those old Redskins teams that he will vote for Monk now. What???? Did you not watch him play in the 80s? Gibbs had to change your mind. King is a disgrace.

A lot of these ignorant sportswriters say these ignorant things about these players. They say so and so is uncoachable. This is the NFL. If you are a coach, then coach the players. You have to lay down the law on your team. King has been stealing money for years talking about football. Its no surprise he couldnt support his claim.
The top QBs have greater numbers then years past but that doesnt mean that overall the play is better. There are so many terrible QBs. The main thing that he left out, David, is that the receivers are bigger, stronger and better. These guys are 6-4 with hops and speed. They werent built like that years ago. They present bigger targets plus you cant touch them as it is. All of todays QBs have good to great receivers who are tall and fast, which plays a great part in how the QBs perform.

David Friedman said...

Madnice:

There is no question that Monk deserves to be in the HoF. He was the all-time receptions leader at one time! He is the Artis Gilmore of football, a quiet guy who put up big numbers and has inexplicably been left out of the HoF.

You make an interesting point regarding the size and athleticism of today's receivers but keep in mind that the DBs are bigger and more athletic now, also. I still maintain that the two biggest changes are the rules changes that I mentioned in the post and the development of the West Coast Offense (and its offshoots).

madnice said...

but the dbs are short unlike these 6-4 6-5 receivers.

David Friedman said...

Madnice:

Granted, there are not many (if any) 6-5 DBs but today's DBs are bigger and more athletic, in general, than their predecessors. What I'm saying is that however much WRs are more athletic than they use to be I think that roughly the same thing applies to defensive players. The difference is that the rules changes limit the contact that the defensive players can make. Today's safeties are as big as LBs (or even defensive linemen) used to be--but they cannot touch a receiver five yards past the line of scrimmage.