DraftAce Quarterback Ratings Through Week 9

The NFL Quarterback Ratings are the single most confusing statistic in all of sports. None of baseball's SABRmetric stats can combare to insanity that is the QB Rating. We know that it is intended to rank quarterbacks, but no one really knows what goes into the formula. And for that matter, no one really cares. It has long been considered a fradulent stat.

I have toyed around with an alternative to the QB Rating in the past, and I think I've finally settled on a formula that I like.

One of the biggest flaws in the current QB Rating is that it does not adequately compare the quarterbacks - which is the basic purpose of the system. The current design of the formula is better suited to compare quarterbacks of all eras on an even playing field. However, that isn't what we ultimately want it to do.

The most common use of the QB Rating is to compare quarterbacks in a given season. Is Peyton Manning having a better year than Drew Brees? The current QB Rating attempts to answer that question, but at no point does it ever actually compare Brees to Manning. Instead, it compares each quarterback to a predetermined set of numbers and gives each a score out of 158.3 (seriously, why 158.3?)

My new system was developed based on the premise that each quarterback should be compared to every other quarterback in the league. Instead of using a complex formula to determine a "perfect" quarterback, players are judged only based on how they compare to others in a given year (technically, it could also be extrapolated out over any number of years if desired).

The formula utilizes four basic statistics which I will outline below:

Completion Percentage - there's no real explanation needed for this one. Better quarterbacks have a higher completition percentage.

TD per Attempt - I'm not crazy about most touchdown statistics because they unfairly favor quarterbacks from pass-happy offenses or from teams with poor defenses that are forced to air it out. TD per attempt levels the playing field for all quarterbacks.

Interception Percentage - not a very common statistic, but I chose this rather than TD-Int ratio or total interceptions to even the playing field for all quarterbacks. For exmaple, Drew Brees is likely to throw more interceptions than David Garrard because he will have more attempts. But his interceptions percentage may be significantly lower.

Attempts per Sack - Sacks are often attributed to a weak offensive line but this is not always the case. Sacks can often be avoided and quarterbacks deserve to shoulder some of the blame. I anticipate some disagreeing with the inclusion of this stat, but I would argue that sacks are as much a stat for offensive lineman as touchdown passes are a stat for receivers. Both parties play a role in the outcome. Again, attempts per sacks rather than the total number evens the playing field. Quarterbacks that attempt more passes are likely to be sacked more, but its the percentage that really matters.

Yards per Attempt - I use yards per attempt rather than yards per completition because it is a better gauge of the effectiveness of the quarterback. A quarterback with a low completition percentage that frequently throws the ball downfield will still have a high Yds/Comp, but his Yds/Att will be significantly lower.

So those are the stats, now for how the formula is calculated:

For each statistic, quarterbacks are compared to the best quarterback in the league in that given category. For example, if Peyton Manning leads the league with a 72.1 competion percentage, each qualifying quarterback's completition percentage score will be his completition percentage divided by 72.1. (e.g. Drew Brees has a 65.0 comp pct, his score in this category will be 65.0/72.1). This allows for a maximum score of 1.00 in each category.

This is done for each category. Once those numbers are calculated they are added together and divded by four (since there are four categories). This will give us a number between .000 and 1.00 (I use three digits for the sake of showing as much difference as possible beween the quarterbacks).

Here are the ratings for this season through Week 9 (all quarterbacks with a minimum 100 pass attempts were included).

I prefer this formula for three reasons. One, I like how it actually compares players to each other. It gives you a better idea of the difference between quarterbacks on a level playing field. And two, it is infinitely easier to understand. A perfect score in this formula is 1.00 - doesn't get any more simple than that. And third, it allows you to easily see the components of the formula. As you can see in the table above, each quarterback's rating in each category is readily visible. This is a drastic change from the current QB Rating where you really have no idea what has led a quarterback to his rating.