Give Me Bad Numbers: Pac-10 Quarterback Analysis, Part II

Posted by: Avinash on Tuesday, August 5th, 2008

The college quarterback rating formula is absurdly reductionist. It overvalues completions rather than completion percentage, it overvalues touchdowns rather than drives, it overvalues interceptions by placing them all at the same value, and it really really really overvalues yardage without analyzing yards after catch. Almost all of the things plugged into the formula represent outcomes rather than processes, and you probably know how I feel about that.

Basically, if you throw the ball a lot and pick up tons of yardage, you can expect to get a +100 passer rating as long as you don’t toss four picks along with it. So you can see that there are too many inadequacies with this system–they don’t really provide an accurate measure of the quarterback’s performance, just a general overbrushing stat to please the fan.

And just by looking at each of the Pac-10 quarterbacks who started this year, you can say the same thing about their relative performances. Note that the Pac-10 defensive ratings are ranked from best to worst.

Here’s the actual graph.

Oooh. Colors. Probably can’t make good sense of it though, so I’ve separated it up, player by player.

Rudy Carpenter (ASU): Mr. Fantastic

“Them QB ratings love Mr. Carpentar. I tell ya sir, he is a good ol-fashioned quarterback. Definitely gonna be somethin’ in the NFL one day, I say. He got what it takes if them stats are a good reflection. Yessir you gotta fine young man….”

I mean look at those stats. All 9 games played in conference, all above the record. Yet you can stare at those stats and find something wrong if you think back to the results of the 2007 season.

(think about it…)

(think about it…)

According to the brilliance of the quarterback rating system, two of Rudy Carpenter’s three best efforts came against USC and Oregon. That sounds about right, because Carpenter and Arizona State lost BOTH of those games by double digits. Hm, wait a minute, that doesn’t sound right at all.

Meanwhile, he played about at par to the other six Pac-10 passing defenses he faced. He won all of these games fairly handily. So when he played at his worst his team won. Makes plenty of sense huh?

Nate Longshore (Cal) and John David Booty: I AM THE MEAN!

Did a whole post on Longshore (re-read it here). The one thing to notice is his numbers are at par against the best opponents, and only bipolar during Cal’s last two games against Washington (great) and Stanford (yugh). Not too much to retread on here.

Interestingly enough, the also-injured John David Booty did not deviate much from the mean. However his best performance came against the best; although he did not play against Oregon or Arizona (two of the top 5 passing Ds in the league), he thoroughly obliterated the Sun Devils on Thanksgiving night. Otherwise, he had so-so games against UCLA, Oregon State, and Cal, and was just bad versus Stanford (loss) and Washington (nearly lost). So what was the difference between USC and Cal if not quarterbacks? Defense, of course.

Dennis Dixon: Amazing upside potential

Dixon’s stats are just incredible. Two out of the park performances against ASU and USC (albeit he was playing in Autzen). His Arizona game hardly counts because he was playing for only a quarter. Two obliterations of Stanford and Washington. His worst performance came against Washington–and the Ducks scored 55 points in that game. And a total fluke loss against Cal, when everything that could go wrong did go wrong. I really am high on Dixon. You can tell, right?

Anyway, these numbers suggest that Dixon would have cruised to the Heisman if…well…you know…

(Let’s move on.)

Willie Tuitama (Arizona): His Coach Sucks

It’s hard to gauge the value of a quarterback on a team that refuses to run the damned ball. Ultimately that arm’s going to tire out, and you’re going to see crappy performances. That Oregon upset took almost four hours and I spent every second of it thinking of the ways I despise Mike Stoops. Total hack who stole Leach’s offense and could do absolutely nothing with it but manage an upset every now and then. You’re not going to convince me otherwise, and I will be popping brandy the moment he’s canned.

Okay, enough about Stoops. Back to Tuitama. He had great performances against the terrible Washington defenses, yet struggled horridly through the entire Stanford game. He destroyed a lame-duck UCLA team, yet was in constant catchup mode throughout the Cal/Oregon State games. His huge Oregon victory was a fairly strong effort, yet it was barely better in rating than the USC game because of the overvaluing of completions and interceptions. All in all, a much bigger question mark for a returning QB than Longshore is for Cal, since I still don’t know how strong his value is to his team (Arizona’s secondary play IMO has been far more important, and that’s never a good thing for a good team).

Jake Locker (Washington): Schrodinger’s Cat

Remains a puzzle entering his sophomore year, although a somewhat intriguing one. Oregon, UCLA and Arizona were all fairly strong efforts, but he also threw up stink bombs against the two best and two worst defenses in the Pac-10. Considering his pertinent running abilities, he also got completely roughed up last year. Makes you wonder if Ty made the right decision to go with the young phenom rather than stick with the senior Bonnell, especially considering the brutality of last year’s schedule. The schedule is equally brutal this year. Locker really is in a no-win situation; he will probably lose many games next year, and if he wins he’s likely to meet turf on numerous occasions.

Tavita Pritchard (Stanford) and Sean Canfield (Oregon State): Benchwarmers

Their performances are inconclusive since they were always being shuffled around with either senior starters (Ostrander) or future prospects (Moevao), but they were at best adequate last year.

Pritchard might have upset USC, but he was 10-29 for 144 yards and a pick before the memorable score, and his passer rating was well below the mean. It was, as usual, a case of the Trojans beating themselves. The other two games he started he finished 1-1, with a decent game against the Wildcats in victory, and a dreadful effort against the other QB in this segment. He saw dual action in the Big Game and was barely visible in the Washington defeat (3 completions). Ostrander was still leading this team most of last year, and Pritchard did little to show he deserved the job outright for this year (a much lower profile QB battle is raging in the Farm).

Amazingly, Canfield was worse (surprising for a team that finished 8-4), making dopey decisions in almost every game he played for the Beavers and only had one game above average (Stanford, and only barely so). The Cal upset rated quite low because Canfield did not throw a pick or a score, and that’s what really boosts ratings. However, Oregon State’s success was predicated on an awesome front seven and the powerhorse Yvenson Bernard, with Moevao coming in late to right the ship.


Was whatever against good teams (at or around the par) and really struggled against middle of the road passing Ds. But man, he owned the Apple Cup didn’t he? He should just play Washington the rest of his life. At the moment though, he remains undrafted.

Anyway, you can tell I’m a little unsold on this passer rating stuff. How can you have a rating system that has a quarterback’s best efforts coming in LOSING performances? I know football is a 50-man game, but come on. This is a ridiculous system in place for determining a quarterback’s value. I’ll try to look at better ways to assess which quarterbacks will mean the most to their teams in 2008.

To see the entire spreadsheet of passer rating data, click here.

Any further analysis of the quarterback rating or these particular quarterbacks?

Related Articles
    None Found

Comments are closed.