I’m sure there’s someplace to find this, but my connection is slow, and my will is weak…
How is the QB Rating calcuated?
One of the reasons I’m asking is this quote from the Washington Post regarding temporary Redskin’s starter Todd Husak’s rather ugly performance against the Chiefs last week:
“I don’t pretend to understand the formula used to determine passing ratings, but Husak’s 3-for-10, 27-yard performance earned him, statistically, a zero, according to the box score.”
I mean, that’s bad, but a Zero? Or are passing rating and QB rating different?
The passing rating formula looks tough, but it is actually very simple. It is explained well at the site above, and in easy-reading detail in the book The Hidden Game of Football by Bob Carroll, et al.
Yeah, the QB rating is about as complicated as sports stats get, makes the OPS look like a peice of cake, excepting the BCS of course. I won’t bother researching and explaining it since the links above cover it adequately enough for the casual observer.
To get to what I assume is the crux of your question, you didn’t mention it, but I bet the player had an interception. Also being lazy and of a weak will, I’m not going to scrape up the box score or anything, so lets just assume. The completion percentage and positive yardage are going to always provide a positive number, but when you factor in negative plays like an interception you can dip into the negative. I’m not sure what the league policy is on records but perhaps all negative numbers are recorded as zero, or maybe the sports writer in question simply choose to round it to zero. Those numbers with one INT gives him a rating of -4.7.
But hey…it could be a typo too. Or maybe the stats guys didn’t record the 4th string QBs numbers before the story went out. Could be several explainations.
For the players’ sake, I kinda hope this is true.
Otherwise, I think Ryan Leaf had a QB rating of -207.8 against the Dolphins last night. (2 attempts, 0 completions and an interception.)
Yeah, you don’t see too many games where the home team fumbles 70 times, loses 20 of them, gets 907 yards in penalties, is sacked zero times for a 100 yard loss and STILL wins by 20 points!
I’m starting to rethink playing Fantasy Football on Yahoo this year…
They’re most accurate when taken over a season or a career. A one-game passer rating tends toward one extreme or the other. If a QB passes for 3 TDs and 0 INTs in a game, his rating is going to be through the roof. If he passes for 0 TDs and 2 INTs, well, his rating is going to look more pathetic than his numbers would suggest. It’s even worse with preseason games, when QBs split time and play for only a few drives. Saints QB Aaron Brooks went 10-for-10 for 120-something yards, 1 TD, 0 INTs in 2 drives, good for a ~152 rating. Jeff Blake, in 2 or 3 drives, was 4-9 with 1 INT, good for a rating of 18.1. Hmmm.
The passer rating takes into account touchdown percentage, interception percentage, completion percentage, and yards per pass attempt. It does NOT take into account sacks or other factors like dropped passes (which, of course, are the receiver’s fault and not the QB’s). Case in point: Rob Johnson had a better passer rating than Doug Flutie last year in Buffalo, but he got sacked one every 7 or so snaps, by FAR the worst in the NFL. Flutie tended toward the other extreme, plus he had a better record as a starter. So who is the better QB? Anyway, someone at NFLtalk.com did what he called his “adjusted” passer ratings, taking into account these things and also QB scrambles. The results were interesting; I wish NFLtalk were still around so I could link to it.
In each category, as the first link explains, the range is from 0 to 2.375. Anything above or below that range is just snapped to the min/max. IIRC, college football doesn’t have such a range, b/c Shaun King (highest passer rating in college history) had a rating around 180. The maximum in the NFL is 158.3. I’m assuming that, since the college game is easier, many good QBs can attain a rating > 158.3. So why have a maximum or a minimum in the NFL? Beats me.
Of course, by the same token the QB gets credit for a completion when the receiver makes a great catch of a poorly thrown ball, and the QB gets credit for a long TD pass when a RB takes a screen pass and runs 80 years for the score, so these things tend to even out over the course of the season.
Too my mind the better QB rating system is from statistian Allan Barra, which is adjusted yards divided by pass attempts, with adjusted yards equal to passing yards less sack yards less 40 times interceptions, figuring that each interception is like losing 40 yards or the equivalent of a punt.
