Lee Jenkins' article in the New York Times today starts out well. Here's the clever opener:
On one side of the field Sunday will be a quarterback who completed more passes this season than Joe Namath in 1968, for more yards than Roger Staubach in 1977, at a higher percentage than Johnny Unitas in 1970.
He threw more touchdowns passes than Joe Montana in 1981, with fewer interceptions than Terry Bradshaw in 1979, and with a higher passer rating than Len Dawson in 1969.
On the other side of the field will be Peyton Manning.
For most of this week, Manning will be the one prompting comparisons to all those Hall of Fame quarterbacks in their championship seasons. As the quarterback of the Indianapolis Colts, Manning has stretched defenses and scorched record books.
But the real statistical marvel of this season is Rex Grossman, quarterback of the Chicago Bears, who has managed to put up numbers on par with some Super Bowl legends and yet is still regarded by some as the reason that his team cannot possibly win it all.
Jenkins is doing all kinds of cherry picking here to make a point: Unitas was at the end of his career in 1970, Staubach played in fewer games and at a much higher level of efficiency and Namath's greatness wasn't in his completion percentage, but in his extraordinary ability as a downfield passer in an era and league where everyone sacrificed short completions for deep passes. Passing-wise, pro football was simply a different sport than it is now. But Jenkins' larger point is well-taken: you don't have to get great quarterback play to win a super bowl, though it sure helps.
But Jenkins is really fudging when he says the following:
Take away the sound bites and the bloopers, and Grossman is a fairly typical Super Bowl quarterback. He ranked near the top of the league this season for interceptions (20), but also for touchdown passes (23) and yards (3,193). His completion percentage, 54.6, was respectable. His passer rating, 73.9, was not.
Grossman really was one of the worst in the league at throwing interceptions - only two quarterbacks threw more. He ranks seventh in touchdown passes thrown, and 12th in yards passing, hardly the equivalent good of his bad interception figure. And, in the modern NFL, 54.6 is not respectable. Of 32 qualifying quarterbacks this year, Grossman's figure puts him 28th. Jenkins makes it sound like the passer rating is somehow separate and apart from its components. But, add all these bad results up: the low completion percentage, the terrible interception rate and the fact that he didn't do anything exceptionally well and there's no mystery to the low passer rating.
Whether the league's passer efficiency statistic is the best way to evaluate QB performance (the Football Outsiders and Wages of Wins authors say, emphatically, "no!"), the Jenkins piece highlights an on-going problem in sports journalism: a failure to put numbers in any kind of context. One reason why many sports journalists are dismissive about the importance of stats (not that Jenkins is) stems from the fact that many of these same folks simply don't take the time to evaluate the meaning and limits of the data. Of course, there's more to life than a QB's yards per attempt. But, properly understood, statistics can give us a picture about the type of player we're talking about, how he stacks up to the competition and, by extension, what sorts of contributions he makes to winning games. It's fashionable these days, as I've complained before, to look at a QB's won-loss record, though that's a frankly inane and hopelessly circular way of determining whether a QB helped or hurt his team's chances. And it just doesn't really tell us much that we don't already know - we can look in the standings to see the Bears' record. But, reducing discussion of QB's performance to whether their team wins also relegates discussion of a player's actual stats to context-less and often inaccurate reporting.
If you don't believe the data matter, you can ignore them altogether. But, it shouldn't be too much to ask, if you're going to use the statistics, to make some sense of them. Of course, sports reporters aren't alone in this regard. See, for example, the economic "boom" we're now supposedly experiencing for a classic case of context-less use of data.