If Kevin Kolb throws 100 or more passes for the Buffalo Bills this year, and first round pick E.J. Manuel also throws 100 passes, it will mark the fourth straight year that Kolb, through injury or benching (or both) has been part of a quarterback committee with multiple passers. The last quarterback to throw 100+ passes and have a teammate do the same four years in a row? Well, you will have to go to the bottom to see the answer.
My interest in Kolb, though, comes from an appreciation for what his career allows: a look at how changing quarterbacks affects teams. Last year, Kolb took a beating, taking 27 sacks in just five starts. When he went out, the Cardinals quarterback situation went from bad to worse, and very little stayed consistent about the performance. Ryan Lindley and John Skelton combined to take the same number of sacks in more than twice as many games, reducing the rate in half. In exchange, the completion percentage, yards per attempt, and touchdowns plummeted while the interceptions shot up from 1.6% to 4.3%.
The year before, John Skelton proved he was a winner, and the switch from Kolb produced more interceptions and fewer sacks. Three years ago in Philadelphia, the Kolb/Vick dynamic actually saw both sacked at the same rate (hardly a sign that Kolb is good at getting rid of the ball), but the yards per attempt and touchdowns were much better with Vick. There’s a pretty good chance that we will see a drastic shift in numbers again this year if both Kolb and Manuel play.
Four years ago, I wrote about what happened when teams changed quarterbacks, looking back at history and how the rate stats of the duos compared. With four more years, I thought I would revisit it using just the recent years. I shortened up the pass attempts limits to 100 passes to get more pairs, and just looked at the last five seasons.
To compare the teammates, I again use the league-adjusted ratings in completion percentage, yards per attempt, touchdown rate, interception rate, and sack rate. What this means is that every quarterback is compared to the league average for the same season. A score of 100 represents a quarterback who is league average in a category. Every increase or decrease of 15 points from that average score equals one standard deviation change.
The phrase “standard deviation” causes some eyes to gloss over, so I’ll give some examples to illustrate what a one standard deviation move looks like, or what it might mean for a pair of quarterbacks to be at least two standard deviations apart in a category. For example, a two standard deviation move in completion percentage would be like going from Matt Ryan to Ryan Tannehill for the 2012 season, a drop off of over 10%. For Sack rate, it would be like going from Drew Brees in New Orleans to Jay Cutler in Chicago.
Here is a summary of how many of the 47 pairs of quarterback teammates were within one standard deviation, at least one standard deviation or two standard deviations different in a particular category.
So what do we see? Largely the same things I saw looking at the other set of pairs. Interceptions and Sacks are the two most volatile statistics when teams change a quarterback. In the other three categories, less than 10% of the pairs saw a huge swing when a new quarterback took over. When a team had a wide fluctuation in completion percentage, it required a pretty significant stylistic change upon the quarterback switch–think Tebow for Orton in Denver.
Keep in mind that when a quarterback switches teams, like Kolb is doing moving to Buffalo this offseason, the most consistent categories are sack rate and completion percentage. That’s not good news for Buffalo’s sack numbers, because Kolb has been dreadful pretty much his whole career (and obviously had it even worse last year behind the Arizona line). Kolb has taken a sack 10.3% of the time he has dropped back the last three years, while the other quarterbacks were sacked 7.5% of the time with the same teammates. It’s a fairly big factor in why he cannot stay healthy and gives us more info to study on this issue.
What this all tends to suggest is that quarterbacks are way more responsible for sack numbers than the general public believes, that interceptions are way more random than people accept (they are noticeable and costly and people tend to ascribe moral implications to them), and that system/teammates matter a fair amount for categories like yards per attempt and completion percentage.
I have looked at quarterbacks changing teams, and teams changing quarterbacks, but it struck me that there is something else that I can do to look at the issue–what happens when quarterbacks and teams don’t change over a season. Splits happen, and we can check the stickiness of these categories when one quarterback does all the throwing, by dividing up the season. That is next.
Oh, and the trivia question. The last quarterback to have a four year stretch where they and a teammate both threw 100+ passes was Kurt Warner, from 2004 to 2007, with Eli Manning, Josh McCown, and Matt Leinart (twice).
[photo via USA Today Sports Images]