Superbowl XLVII (2013) – Baltimore Ravens’ Offense Predictability

In preparation for an upcoming talk at the Microsoft NERD center on R and Tableau integration, I created a dashboard in Tableau to measure the predictability of the Baltimore Ravens offense. I’ll use this tool during the Superbowl to attempt to predict the Ravens’ plays based on their 2012 regular season data – unfortunately, playoff data isn’t available yet. I chose the Ravens offense because the 49ers offense changed dramatically once Colin Kaepernick took over Alex Smith’s starting role as quarterback. John Harbaugh made a dramatic change by firing offensive coordinator Cam Cameron in week 15, so that’s another variable to keep in mind when using the Play Predictor.

How to Use the Play Predictor
1. Set the Quarter (5 = overtime)
2. Set the Yards to Endzone (< 10 = within 10 yards of the endzone)
3. Set the Down
4. Set the Score Difference (negative = Ravens are trailing)
5. Set the Yards to 1st Down

The Play Predictor will then display the different types of plays that the Ravens ran in that particular situation. Of course there are many other variables that can't be accounted for such as defensive and offensive formations, injuries, crowd noise, etc. I'll use this dashboard during the Superbowl and record the predictions and play outcomes (run or pass) to determine the predictability of the Ravens offense. I'll post the results here afterward. UPDATE: View the results here.


Comments (0)

Join the discussion, leave a reply!