<
>
EXCLUSIVE CONTENT
Get ESPN+

Roethlisberger among worst decision makers

Editor's note: This is an expanded version of a passage that appears in Scientific Football 2006.

I track more than 100 metrics in every NFL game, but the most misunderstood and controversial of these has to be the bad decision metric for quarterbacks. The bad decision metric is the method I use to track a quarterback's mistakes. Simply put, if the quarterback makes a decision with the ball that either could have led or did lead to a turnover, he is debited with a bad decision on the play.

The most common types of bad decisions are:

1. Forcing a pass into coverage
2. Staring at a receiver
3. Throwing the ball despite being tackled
4. Misreading a zone defense and not seeing a defender in the passing lane.

On each of these types of plays, the quarterback is noted as having made a bad decision. There are game situations that can force a quarterback to throw the ball into coverage (e.g., on fourth-and-30, Hail Mary plays, etc.) that will not be ruled as bad decisions.

If the quarterback's mistake did not lead to a turnover (e.g., a dropped interception, a recovered fumble, etc.), the mistake is given only one bad decision point. If the mistake led to a turnover, however, it is given two mistake points and also is subject to a graduating scale of points based on how damaging the turnover was (e.g., an additional point for an interception killing a scoring drive, another additional point if the interception led to the opponent's being set up in scoring position, etc.). The scale has an upper limit of five points for any single bad decision.

Bad decision rankings are based on two percentage bases. The first is the standard bad decision percentage. To calculate this, I take the number of bad decisions a quarterback generated and divide it by the total number of attempts. The second percentage rating is the weighted bad decision percentage. This is calculated by taking the number of bad decision points and dividing it by the total number of attempts.

So, which quarterbacks were the best and worst decision makers in the NFL in 2005 according to Scientific Football 2006? At the bottom, I'll list the top five and bottom five in each category. But first, I'll analyze some of the more interesting QBs and where they fell as good or bad decision makers.

Good decision makers:
Tom Brady: The best decision maker in 2005 was Tom Brady, and it wasn't even a close race. Brady made only four bad decisions in 549 attempts, which equates to a ridiculously low 0.7 percent bad decision percentage. No other quarterback came close to falling under the 1 percent mark in that category. Brady also had only nine bad decision points, and his 1.6 percent weighted bad decision percentage was also easily the best.

Drew Bledsoe: The perception of Bledsoe is that he makes a lot of mistakes, but the metrics tell a different story. Bledsoe did rank next to last in the NFL for most interceptions thrown last year (17), but he had just 10 bad decisions. His 1.9 percent bad decision rate was the fourth best in the league, and he was only 0.2 percentage points behind Peyton Manning in that category. Bledsoe did have 21 bad decision points (tied for 28th worst in that category), but his high number of pass attempts meant his weighted bad decision percentage barely missed making the top 10.

Byron Leftwich: Leftwich often is lauded for a number of his positive traits but his good decision making isn't normally one of them. It should be, as Leftwich ranked in the top seven in both bad decision and weighted bad decision percentage. That he did this despite throwing the second-highest percentage of deep passes (nearly 20 percent of his pass attempts were deep) is a testament to his ability to be aggressive while simultaneously protecting the ball.

Bad decision makers:
Aaron Brooks: Brooks had a number of problems last year, but his biggest by far was his decision making. His 4.9 percent bad decision percentage was the fifth worst in the NFL, and his weighted bad decision percentage was almost into double digits (9.8 percent). Those percentages were nearly twice as high as his 2004 totals, so he is certainly capable of performing better.

Matt Hasselbeck: Hasselbeck had a fantastic season last year almost across the board, but his 17 bad decisions and 26 bad decision points were both tied for the seventh-highest total in their respective categories. The West Coast offense operated by the Seahawks is not a high-risk offense, and that makes Hasselbeck's performance in this metric a bit more perplexing.

Brett Favre: It should come as no surprise that Favre had the highest number of bad decisions and bad decision points in 2005. To put his 53 bad decision points into perspective, one could add the totals of nearly any two other quarterbacks in the NFL last year and they still wouldn't surpass Favre's total. Favre's saving grace from a percentage perspective was his extremely high number of pass attempts (he was the only quarterback to top 600 attempts in 2005), but he still ranked in the bottom 10 in both bad decision percentage categories. Green Bay will need a dramatic improvement from Favre in this aspect of his game if the Packers want to be competitive in 2006.

Ben Roethlisberger: Big Ben's 2005 season provides proof that you can win despite having high bad decision percentages. Roethlisberger ranked dead last in the bad decision percentage category and had the fifth-highest weighted bad decision percentage. He was able to get away with those elevated percentages because he had by far the highest yards per pass attempt in 2005 (9.1 yards, the only QB to top the 9-yard mark). If the YPA number drops at all in 2006, Roethlisberger's bad decision percentages will need to show improvement.

Here are the top and bottom five in each percentage category:

KC Joyner, aka The Football Scientist, is a regular contributor to ESPN Insider.
His latest release ("Scientific Football 2007") is available for a special preorder price at his Web site.