Starting Joshua Dobbs: What Can Go Into a Quarterback Decision

The Vikings have been stuck between a rock and a hard place when it came to their decision to figure out who should start at quarterback at the bye. Did analytical decision frameworks make the choice?

Photo by Todd Rosenberg/Getty Images

The Vikings decided to stick with Joshua Dobbs to start this week ahead of backup quarterbacks Nick Mullens and rookie Jaren Hall.

Given Dobbs’ poor performance against the Chicago Bears and Denver Broncos over the prior two games, there was reason to believe they’d make a quarterback change – they said as much after the Bears game.

So why stick with Dobbs? After all, Mullens has a stronger resume, has been with the team longer, knows the offense better and likely wouldn’t have the timing or chemistry issues that plagued Dobbs’ outings against Denver and Chicago.

A data-forward front office and data-friendly head coach would pick the guy with better statistics, right? Even when using more advanced statistics like EPA per play, Mullens dusts Dobbs.

Not necessarily. Analytical thinking is not the same as choosing the option with the best statistics.

Choosing Uncertainty

The Vikings might have been influenced by the concept of variance. The truth of the matter is, against an average opponent, the Vikings aren’t expected to win with their backup quarterback in the game. Given that fact, it might be better to shoot for the moon more often.

Brian Burke, now a senior analytics specialist for ESPN, once ran a blog called Advanced Football Stats, which was then changed to Advanced Football Analytics. The archives are still up and there’s a 2009 piece he wrote about coaches being too timid when playing as underdogs.

In essence, Burke argued, offenses should be more comfortable pursuing high-variance strategies when they’re outclassed. That would mean strategies like going for it more often on fourth down, throwing deep and reducing the total number of possessions in a game.

He found that coaches in underdog situations tend to play conservative, not aggressive. In short:

My theory is coaches are delaying elimination until the latest point in the game—that is, trying to “stay in the game” for as long as possible. Underdog coaches minimize risk all game long hoping for a miracle along the way. They seem to be reducing the chances of being blown out, but this is not consistent with giving their team the best chance to win.

While these can work at cross-purposes – throwing deep can mean punting sooner and increasing the total number of possessions – it’s important to approach in-game decisions with this framework in mind. In order to prove that, he put together some charts that demonstrate the idea.