As we look back on the 2012 season that was it’s interesting to look back at the 2012 Raiders team that was, celebrate their strengths (few that there were) and discuss the weaknesses. There are many ways to do this – film review, interviews with those involved and, finally, statistical breakdown which will be our focus for today.
As with an actual football team, it all starts with a quarterback. Raiders QB Carson Palmer looked like he was going to have great year at the start of the season, throwing for yards, scoring points and reducing his turnovers but the team’s offense couldn’t get into gear and he slumped in the second half of the season and ended up missing the last game due to injury.
Football Outsiders have a stat which they call DVOA (defense-adjusted value over average) in which the team at Football Outsiders not only take into account the run and pass results (i.e. “run left tackle for 3 yards”) but also whether that run was successful (did it get the yards needed? For example a 3 yard one on 3rd & 1 was successful but a 3 yard run on 3rd & 4 was not). It’s a fairly complicated formula that I haven’t taken the time to properly understand completely but if you wish to, you can read much more about DVOA here.
Carson Palmer’s DVOA for the 2012 season was -2.4% which means that he was less valuable, per play, than an average quarterback per their metrics. Overall, Palmer came in 18th on the list of QBs ranked by DVA (minimum of 100 passes attempted).
Palmer was in good company at 18th (regular season only accounted) as he was right between first time Super Bowl QB Joe Flacco at 17th (-1.4%) and possible rookie of the year Andrew Luck (-5.2%). For what it’s worth the man who took his place in Cincinnati, Andy Dalton, was at 20th place (-5.9%).
Football Outsiders also tracks ESPN’s preferred statistical model for Quarterbacks, QBR, which is also touted to take into account much more information than simple raw numbers. For example, QBR takes out drops, intentional throwaways and splits blame between QBs, WRs and blockers on drops. It also has a “clutch factor” to account for QBs who make passes when the game is on the line or on big 3rd downs instead of accounting garbage time stats as equal with clutch stats.
Palmer’s QBR rating was 44.7, which puts him at 29th out of the 39 QBs that qualified (again, 100 passes thrown). If QBR is to be believed, Palmer wasn’t anywhere near the likes of Luck (11th, 65.0) although Flacco (25th, 46.8) and Dalton (22nd, 50.7) are still within shouting distance.
The biggest difference is likely the weight that QBR puts on the clutch moments as anyone who has watched a Raiders game this year knows that Palmer struggled when the game got to the big moments, often throwing a back-breaking interception while trying to do too much and forcing the ball.
Then there is, of course, the old QB rating metric which has been used for years. A straight statistic, it doesn’t take into account down, distance, anything other than the numbers. By this metric (taking into account QBs with 100 passes or more) Palmer falls at 19th with a season average of 85.3. Directly in front of him at 18th is Kevin Kolb (86.1) and behind him is Ryan Fitzpatrick (83.3), a couple of disappointing fellow QBs in whose company I’d bet Palmer would rather not be mentioned.
But Palmer isn’t really competing with the other quarterbacks that have been mentioned in this piece thus far. There are some teams in which Palmer would the best QB and a number of others in which he wouldn’t but he doesn’t play for any team but the Raiders – at least not currently.
The question becomes then, is he the best choice for the team? Luckily, the minds at Football Outsiders have come up with a stat for this as well. It’s called DYAR and it stands for defense adjusted yards above replacement. It attempts to take all of the information provided by the DVOA, above, but then rate the player with the other possible options on his team.
So, for example, when simply looking at the numbers, the DVOA, SF 49ers QB Colin Kaepernick is almost at the top - #3 overall behind only Tom Brady and Peyton Manning. However, Kaepernick is not the only good QB on his team. He replaced former #1 overall selection Alex Smith mid season this year and if Kaepernick weren’t playing or not well enough to be starting, Smith would be the player to which Jim Harbaugh would turn.
Therefore, DYAR takes into account what benefit Kaepernick brings to the team. In this instance, because Alex Smith also has above average DVOA, Kaepernick doesn’t contribute as much to the yards above his replacement as other QBs do on their team. When the statistics that Kaepernick accumulated in his playing time is weighed against Alex Smith’s, Kaepernick falls to 13th on the DYAR chart.
Taking into account Leinart and Pryor, Palmer stays the same as he was on the DVOA list – right at 18th place. That means that the Leinart/Pryor combo is rated as below average themselves as if they were rated as “average” Palmer, who was already ranked below average on DVOA (-2.4%, remember?) would likely have fallen further.
If you’re curious on where Pryor and Leinart did fall, Football Outsiders does have their data although it’s questionable whether the amount of time they played would be considered statistically significant. Leinart had a terrible DVOA of -66.1% in his 34 passes attempted. Pryor had a much more respectable -6.5% in his 30 qualifying passes.
So which system is best? That’s a very debatable question and one that many have asked for years as we attempt to appropriately quantify what our eyes see on Sundays. What we do know is that Palmer, as he freely admits, has much room for improvement as the Raiders move into the 2013 season with a new offensive coordinator.