|Connelly’s Top Ten: Holt Magic, Brady is Awesome, Exorcist Wicked Scary||Sox Take Two From SF Giants||Retirement Looms Large For Big Papi||Hey David ! FIGURE IT OUT.|
After watching Tom Brady pick apart the Tennessee Titans like Jeffrey Ross picking apart celebrities at a roast, a question came to mind: did Brady hit the quizzical “perfect” passer rating of 158.3?
Brady so thoroughly dissected the ironically named Titan defense at snowy Gillette Stadium on Sunday, he played just five minutes into the second half, making his stats even more eye-popping: 6 TDs, 0 INTs, 380 passing yards — a line most QBs would be happy to post in two games, let alone 35 minutes.
When Brady’s passer rating of 152.8 was announced in the third quarter, even Phil Simms was stumped. The stats are worth repeating for emphasis: Six touchdowns, zero interceptions, 380 yards, and a mere five incomplete passes on 39 attempts. Oh, and Brady did it in the middle of a snowstorm.
Tom Brady played a Gisele Bundchen of a football game. But not only did he not have a perfect passer rating, he wasn’t even the highest rated passer on Sunday. That honor went to Drew Brees, who posted a 156.8 on this line: 23 of 30, 4 TD/0 INT, 369 yards. Brees (playing inside the climate controlled Superdome) threw more incompletions on fewer attempts, two fewer TDs, and for fewer total passing yards, yet he earned a higher passer rating than Brady.
In the words of Kyle Broflovski’s mom: What, what, whaaat?
The discrepancy requires some investigating. It’s time to don the old deerstalker cap and crack the perfect passer rating code once and for all.
According to the NFL, retired Pro Football Hall of Fame Vice President Don Smith led a committee that cooked up the current passer rating formula in 1973.
Smith and company settled on four categories to compile the measure:
Interested in computing a QB’s passer rating yourself? You’ll need a calculator. Each component is computed as a number between 0 and 2.375. The four components are then added, divided by 6, and multiplied by 100: (C + Y + T + I / 6) X 100. Told you you’d need a calculator. And yes, there will be a quiz at the end of this column.
So how does a QB achieve a “perfect” rating of 158.3? His completion percentage must be at least 77.5% (Brady on Sunday: check, 85.2%); ratio of TDs to passing attempts at least 11.9% (Brady: 15.4%, check); he must throw zero interceptions (check); and average 12.5 yards per passing attempt (Brady: 11).
And there’s the rub. Despite crushing completion percentage, ratio of TDs to passing attempts, and throwing zero picks, Brady did not throw for 12.5 yards per attempt. In other words, the formula heavily rewards those that sling the ball downfield — QBs like Drew Brees, who averaged 12.3 against the Giants on Sunday.
Thirty-four different quarterbacks have hit the mark of 158.3 since 1973, the most recent being your vacant-eyed friend and mine, Eli Manning. Perhaps no “perfect” game more demonstrates the flaws of passer rating than Manning’s October 11 performance against Oakland.
Manning’s line: 8 of 10 passing for 173 yards, 2 TDs, 0 picks. Passer rating: 158.3. Ahh, nothing says “perfect game” like 8 of 10 passing.
Clearly, in using passer rating to evaluate a QB’s performance in a single game, the measure lacks a critical component that takes into account the number of passes a QB actually attempts. Imagine lauding a baseball pitcher with a perfect game for tossing four innings — he wouldn’t even qualify for the win.
Ironic, isn’t it? Quarterbacks are far more likely to have a higher passer rating, and a “perfect game,” if they throw fewer passes. In fact, of the 34 QBs to hit the perfect mark since 1973, none have thrown more than 32 passes, 27 have attempted 20 or fewer, and only two have topped 30 attempts.
Which brings us back to Don Smith, the man credited with creating the current passer rating formula back when Jim Plunkett was under center for the Pats in ‘73. According to the NFL, Smith envisioned passer rating as “a measure to compare passing performance from one season to the next.”
In a 2004 interview with the New York Times, Smith said his committee “sought to use the four categories to create a 10-year standard of achievement against which quarterbacks could be measured.
Turns out passer rating was never intended to be used with small sample sizes, let alone a single game.
And that’s how the statistic was properly applied, more or less, for 31 years. Then Peyton Manning came along in 2004 and sparked the single-game, perfect passer rating frenzy.
By any statistical measure, Peyton Manning was a man among boys in the first two games he played in the 2003-2004 playoffs. In the AFC Wild Card game against Denver, Manning was 22 of 26 for 377 yards, 5 TDs and 0 picks.
In gushing about Manning’s performance, CBS announcers Greg Gumbel and Phil Simms quickly ran out of adjectives. Producers saved the day by providing Manning’s passer rating for the game, 158.3, the highest possible score according to the measure — no, not the highest possible score, the perfect score. So what if passer rating wasn’t intended to measure single game performance? So what if Manning threw incomplete passes? He was perfect, and here was the stat to prove it!
Thus the so-called perfect passer rating was born. And announcers, reporters, and fans have been using it ever since. Hey, who cares about facts when you’ve got hyperbole?
Tangent: Brady’s 2007 passer rating of 117.2 ranks second all-time to Peyton Manning’s 121.1 in 2004. A comparison:
|Brady’s 2007||Manning’s 2004|
|50 TD, 8 INT||49 TD, 10 INT|
|4,806 yards||4,267 yards|
|68.9% completion on 578 attempts||67.6% completion on 566 attempts|
|8.3 yards per attempt||9.2 yards per attempt|
Seeing the stats side by side, it’s once again clear the passer rating measure heavily rewards the yards per attempt component.
One more tidbit to add the comparison: Manning played 10 games indoors in ’04; in ’07, Brady played two regular season games under a roof.
We’ve already documented three flaws to the passer rating formula: (1) QBs that throw fewer passes can easily accumulate a higher score, (2) the yards per attempt category trumps the other three categories, and (3) the single-game, “perfect” passer rating is largely a media-created misnomer. But, it’s the fourth flaw that really makes a “perfect” passer rating look silly.
According to the formula, the most points a QB can be awarded in each of the four categories is the seemingly arbitrary 2.375 points. For example, if a QB hits the targeted completion percentage component of 77.5%, he is awarded the full 2.375 points. If he scores above the targeted percentage, as Brady did on Sunday with his 85.2%? Doesn’t matter, he is still only awarded 2.375 points.
Even if Brady was truly perfect and completed 100% of his passes, threw 12 TDs, and did it all in the middle of a snowstorm (oh, wait, he really did do that), he still would have received just 2.375 points in the completion and TD categories. And given his 11 yards per passing attempt, he still would have scored just 152.8.
Let’s hear it for the NFL passer rating everyone: a deeply flawed, inappropriately used statistic that’s all the rage!
A passer rating of 152.8 does not accurately reflect Brady’s dismantling of the Titans on Sunday, but we will admit this: passer rating nailed Kerry Collins’ dismal outing, awarding him a grand total of 4.9 points.