Georgia Tech - Wofford Advanced Box Score

Each Week FTRS will take apart the traditional box score and present our readers with the key insights on how the game was played. You can view all the Advanced Box Scores in the Football Analytics Story Stream. For any questions about the metrics listed below please first refer to the Advanced Stats Glossary for definitions.

Long time readers may remember this column from last year but I have some new goodies as well. For those new readers I will explain each graphic throughout the column. As always any comments about the presentation our welcomed in the comments.

Advanced Box Score

• 0 sacks, 0 turnovers, 55% success rate on 3rd downs. Was this game flashy? No. Did we play our best? No. But we took care of the small things necessary to win.
• Also a 50% success rate on 1st downs is awesome. This shows the offense was able to put themselves in good position and not have problems on 2nd or 3rd and longs.
• A lot has already been said about the success of our passing game, but these numbers are awesome; 17.6 yards per dropback and a 63% success rate. And again, NO PICKS! We may not be able to produce the same output against higher competition but we will worry about that later.

Field Position

I am attempting to visually represent field position throughout the game. I already listed the average starting position of each team in the Advanced Box Score but I want to look at where each team had the ball on the field the most, not just where they started. This graphic shows a smoothed distribution of how many plays occurred at each spot on the field for each team (GT is in gold, Wofford is grey). The higher the bump the more plays a team had at and near that spot on the field. The top number are just for looks, if you notice the x-axis is distance from endzone, so far right is bad field position and far left is about to score.

• Georgia Tech really did a nice job of moving the ball down field and running plays in the opponent's territory.
• Wofford was unable to move the ball too much. The have a very high spike at ~60 yards out and only gained 47% of their possible yardage.
• This is very much a work in progress visual so any suggestions or criticisms would be appreciated.

Win Probability

This is something I worked on this off-season. The y-axis is Georgia Tech's chances of winning the game, and the x-axis is time elapsed. Each dot represents an individual play.

• Because Georgia Tech was at home vs an FCS team the model gives them an approximately 80% chance of winning right off the bat. From there the game was basically even until GT pulled away in the 3rd quarter. The TD drive to start the 3rd quarter basically put the game away as Georgia Tech had a 95% chance of winning at that point.
• The biggest play in the game was Wofford's 92 yard touchdown run just before halftime. It increased their predicted probability of winning the game by 12 percentage points, going from only a 7% chance to nearly 20%.
• Georgia Tech's biggest play on offense was the 4th and 5 pass from Justin Thomas to Tony Zenon. Before the play Georgia Tech had a predicted win probability of 87% and increased that to 92% by converting the 4th down. Here were the 5 biggest plays of the game according to my model:

(*) WP = Win Probability, WPA = Win Probability Added

At the end of the day Georgia Tech got a win against a team it should have beaten. There were some good things like our success in the passing game, ability to win the field position battle, and not turning the ball over. But also some things we need to work like our run defense.

Thoughts or comments of your own? Critiques or suggestions for the visualizations? Put them in the comments.