By: Anthony Jones
Contrary to what many people think, advanced analytics are for more than just 4th down decisions.
This week we’ll look at one of the many alternate use-cases: an analysis of play execution on 1st, 2nd, and 3rd downs, through the first half of the season. On 1st and 2nd downs the traditional baseline metric for team comparisons is yards/attempt and on 3rd downs the metric is based on conversion percentage. These metrics are then overlayed with win probability gains/losses. The result is a more complete picture of actual team efficiencies.
The charts below are grouped by Offense and Defense and then by down. The horizontal axis indicates team. The light gray bars and left vertical axis, together illustrate the appropriate baseline metric. And the light blue line with shaded areas, the percentage labels, and the right vertical axis, together indicate win probability gains/losses. And the chart values are sorted by the baseline metric.
Offense
Using the Tampa Bay offense as an example of how to read these charts, a quick glance shows us that traditional metrics (i.e., yards/attempt and conversion pct) can be misleading. For example, on 1st downs despite having the 5th highest overall yards/attempt number (5.9 yards/attempt, including all runs and passes), TB is tied for the league’s lowest win probability number. More specifically, on 1st downs the Bucs’ play execution has, on average, cost them -3% in win probability per game. Luckily for them, they’ve been able to offset this loss with good win probability gains on 2nd downs (+6%) and on 3rd downs (+6%).
For TB, the next logical step is a deeper analysis of their 1st downs (i.e., by play type, field position, quarter, score differential, etc.) but that’s for another column.
Defense
On defense, on 1st downs the Vikings play execution has cost them, on average, -9% in win probability per game. However, like the Bucs offense, the Vikings defense has offset that loss with good win probability gains on 2nd downs (+9%) and on 3rd downs (+10%). Again, the next logical step being a deeper team-specific dive.
Bottom Line
The point here is that the first step to capitalizing on (or fixing) any execution issue(s) is having a process in place to correctly identify them (i.e., with advanced analytics). Every team has a unique set of strengths and weaknesses, which stands to reason because all teams are different. Effectively leveraging analytics is about focusing on what matters for your team. To do that you need to understand decision-making AND play execution in the context of their impacts on win probability.
For a more in-depth explanation of these analyses please contact me at ajones@edjvarsity.com. And for more information about how coaches in the NFL, NCAA and High School levels are using our analytics to coach with confidence, visit www.edjvarsity.com.