This system should be rated
X
Paul Zimmerman has covered the NFL for Sports Illustrated
since 1979. His exclusive online column appears each
Friday.
Posted: Fri December 12,
1997
I was watching a sports show over the weekend and they got
on the subject of the rating system the NFL uses for its
quarterbacks. Naturally this drew laughs right away. So
complicated no one can understand it, you have to be a
computer genius to figure
it out, all the
blah blah
blah that ignoramuses have used from Day One to cover basic
stupidity and
laziness.
Just to reinforce the idea, they interviewed a few big-name
QBs, and naturally these guys went along with the idea.
"Beats me how the thing works," "I don't
have the faintest idea," etc. Of course, they were
playing it for laughs, but quite a few
quarterbacks I've talked to know
exactly how the system works—and they know all about its
failings. Then Brett Favre, who had professed ignorance,
said an interesting
thing.
He said that at the end of the first half of one game, the
Packers were backed up, with time for one play, and Mike
Holmgren told him to just let it fly downfield. Favre said
to himself,
The hell with that, I could throw an interception and screw
up my
rating. So instead he dumped the ball off to a back, who gained
about 20 meaningless yards. The cash register jingled and
Favre's rating went up.
He might not have understood the actual mechanics of the
system, but he knew what helped and what hurt. But why
shouldn't a quarterback make sure he understands it, since hundreds
of thousands of dollars in incentive bonuses in his
contract could be—and frequently are—keyed to the
ratings? Every decathlete in track and field has a detailed
knowledge of the intricate
points system and that one involves 10 categories instead of the
four in football. Maybe football players are just dumber.
Or maybe it isn't cool to admit that this might be
something to be learned, rather than sneered
at.
To figure out the system you need a table of instructions,
which you can get by writing to the NFL. Then all you do is
look at a quarterback's statistics in four categories,
record the total points allotted in each category, add 'em
up, and then—OK,
I'll admit, here's where it gets a little loopy—divide by 60
to establish his rating on a classroom-style 100-point
scale. I've asked the league stats people a million times:
Why not just use an absolute scale and eliminate the need
to divide by 60? Then
you'd just add the points up. And I've always gotten the same
answer, which makes as much sense to me as it does to you:
"That's the way it's set up." I quoted Ralph
Waldo Emerson, "A foolish consistency is the hobgoblin of
little minds," but by then I'd
lost my
audience.
These are the four
categories:
Percentage of
completions. Too liberal. When the system was set up in 1973, the NFL
was coming off a season in which the league average was
51.7 percent. Right now it's 56.2. That completion
percentage, which brands a QB
average, earns him rating points equivalent to 87.3, which would
place him ninth among the 31 rated quarterbacks. In other
words, in this system,
average equals
superior.
Percentage of
interceptions. OK, no quarrel with
that.
The final two categories,
yards per
attempt and
TDs per
attempt, are both keyed to a high completion rate. The whole thing
is too heavily loaded toward the play-it-safe guys, the
dinkers. Which was why a dinker like Milt Plum of the old
Cleveland Browns could run up the highest single-season
rating in history,
before Joe Montana and Steve Young. Yes, he was way ahead of
down-the-field throwers like Joe Namath, Sonny Jurgensen
and Johnny Unitas, all Hall of
Famers.
So I suggested to the NFL stats people, why not key two
categories, average yards and average TDs, to completions,
not attempts? That way you don't make the thing a pure
dinker's system—you don't make passing percentage the
be-all and end-all. The
answer I got was that the system had been changed enough—eight
times since stats were first kept in 1932. Well, why not
change it one more time if it would make it
better?
The problem is that these people are lords of their little
kingdom and if they get challenged by an occasional
lunatic, such as me, all they have to do is sit tight and
it'll blow over. I've been haranguing for years, for
instance, that the ugliest play
in football, the quarterback kneel at the end of a game or
a half, should not be counted in the rushing stats. They're
not rushes, they're uglies, zilches, give-ups. Stick an
asterisk at the bottom of the stat sheet—"Three
kneels for minus-four yards,
not included in total statistics"—and that way you
don't foul up a more accurate presentation of what's
happened.
Say a team rushes for a 140 yards on 35 carries, a
respectable 4.0 average. Then at the end of the game the QB
kneels three times for minus-four. The team's averge has
been knocked down to 3.58, which seems kind of crummy.
Project that to a whole
season and you get my point. The better teams are penalized
because they win more games and kneel
more.
I've brought this up at countless league meetings. I used
to badger the Competition Committee about it endlessly.
"KNEELS ARE NOT RUSHES!" Tex Schramm used to
throw up his hands and say, "There he goes again. Mr.
Kneel. Don't you realize we have
important things to work on?" Yeah, like the TV
contract.
OK, I'm getting myself all worked up and my wife had to go
get me a cup of hot cocoa, but honestly, did Columbus have
it this rough? No, he didn't and he probably would have had
a better week handicapping than I did last time. So here
are my get-even
picks:
Tampa Bay to knock off the
Jets in the Meadows. Bill Parcells wins and he takes the
credit. He loses and it's the players who quit on him.
Never mind the fact that the offensive line is a shambles,
with both tackles down and the rookie left guard, Dirt Bag
Burns, refusing to talk to
the media after his one and only start. Oh,
brother.
San Francisco over
Denver in a game that looks meaningful for the Broncos but really
isn't (if K.C. keeps winning). They'd be better off resting
their defensive ends, both of whom are playing with torn
triceps muscles. Actually it's not all that urgent for the
Niners either,
since they can clinch home field for the entire playoffs with
either a win here or next week against Seattle. But hey,
it's a Monday-night happening. Jerry Rice's possible
return, the ceremony retiring Joe Montana's number (a
couple of weeks after an
FBI investigation retired Eddie DeBartolo's number)—all are
reasons to favor the home
team.
Giants to take the NFC East by beating
Washington. One of the Skins' few potential big-play people, wideout
Michael Westbrook, looks more and more like the classic
coach-breaker. He did zilch against the Cardinals last week
until Aeneas Williams got hurt. Then he caught a couple of
dinkers. The
Giants' corners will eat him up.
New England over the
Steelers. I'm still not convinced Pittsburgh can beat a quality
team on the road.
Cincinnati to lay more flowers on the Cowboys'
tombstone. I can't remember when I've been more wrong about a game
than Carolina-Dallas last weekend. What was I looking at
when I predicted a Cowboys' blowout? I was looking at the
Panthers' loss to New Orleans, that's what. But the way the
Cowboys are dragging
themselves around now, any team with zip and emotion will
beat them.
Detroit, strictly on a hunch, to upset the
Vikings. Five turnovers in Miami, two screwed-up extra points and
a partially blocked punt, and the Lions
still nearly
won.
Here's an intriguing game: Arizona-New
Orleans. The innovative emerging superstar, Jake Plummer, working
behind perhaps the worst offensive line in the business,
going against an active, nasty defense. Logic says the
defense forces three or four turnovers and the Saints slop
their way through
another one and kick a field goal at the end.
But
emotion
says the Cards will rally behind their boy
wonder. Arizona is the
pick.
Previous
editions of Dr. Z
| 
