Did Larry Fedora make the right call using the onside kick?

After scoring the go ahead touchdown in yesterday’s game, North Carolina head coach Larry Fedora decided to go for the surprise onside kick up just one point with 4:03 left in the fourth quarter. If you need a replay, here it is from Inside Carolina on Twitter…

The move ended up working out really well for the Tar Heels, as UNC was able to run out the clock and prevent UVA from ever getting the ball back to score. But ignoring the results of the play, did Coach Fedora make the right call?

Here is the changes in Win Probability for different outcomes in the game.

[table id=6 /]

A note about the Win Probability calculator used — it is for NFL teams only, so the numbers aren’t specific to UNC and UVA or to college football. But I did add in the -7 line for the game, which shades the win probability data accordingly in UVA’s favor.

It is interesting how the difference between the touchback and the onside recovery is only 6% points in Win Probability. Assuming that the calcualor was 100% accurate, Coach Fedora made an excellent decision, because an onside kick in UNC’s favor adds 25 points to UNC’s win probability, while a failed onside kick only drops six points.

There are a couple of other things to consider when examining whether or not Fedora made the right choice. For one, many of the Carolina players on the sideline were saying that the Virginia players were turning around too quickly and that UNC could sneak an onside kick in there. Here is an excerpt from Andrew Carter’s story about the onside kick, which reveals more about that exchange…

"It happened again and again, and kept happening, and so various North Carolina players took the information back to the sideline and alerted the coaching staff that the onside kick was there if the coaches wanted to try it.“Just from what they were doing the whole time during the game, in order to block us they had to drop back so fast,” Mack Hollins, the UNC receiver, said after the Tar Heels’ 28-27 victory against Virginia on Saturday."

What Hollins is saying is that all game long, Virginia’s front line of kickoff blockers were turning around quickly and going to block, potentially allowing for a quick onside kick to counteract the offensive blockers leaving early.

If that is the case, and it appears that was true looking at the video again, Fedora made the right call listening to his players and trusting their opinions.

Also, in order to analyze the decision, you also have to take into account UNC’s porous defense. UNC’s defense allowed 443 total yards to Virginia over the course of the game. With 4:03 left in the game, Virginia would have been inclined to run some clock and hand the ball off to Kevin Parks. But UVA only averaged 3.7 yards per rush, versus 6.9 per pass, illustrating how UNC’s run defense was better than the pass defense. Maybe UNC could have gotten a stop if the team had failed to get an onside kick, but they would have been much more likely to hold Virginia from the 25 yard line on a touchback than on UNC’s 46 after a failed onside attempt.

Also, we should look at where Ian Frye, UVA’s kicker, could make the potential game winning kick from. On the season, Frye’s long kick is 47 yards, and he is 4-5 from the 40-49 yard range. Frye is a very solid kicker and would be in range anywhere inside the 30 yard line.

Assuming Frye’s range starts with a 47 yard field goal, Virginia would have needed to travel 16 yards in 4:03 minutes of game time to get in field goal range, a very doable task, after recovering the onside kick. Then again, travelling 55 yards, as Virginia would have needed to do after a UNC touchback on the kickoff in question, doesn’t bring the issue of the clock into play, only the issue of how good the defense is. I think UVA would have marched down the field and scored at least a field goal whether UNC tried the onside kick or just a normal kick due to the time left in the game and the relative weakness of the Carolina defense. Looking at the defense and field goal effects, Fedora made the right call.

According to a Harvard Sports Analytics article from October 6th, surprise onside kicks were successful 60% of the time in the NFL from 2001 to 2010. In this game, the 60% chance of recovery was probably boosted up a bit because of what Mack Hollins noticed on kickoffs. Fedora rolled the dice, knowing very well that “[everyone] would be crucifying me right now” if UNC didn’t recover (quote from Andrew Carter’s article again). But I think regardless of the outcome, the decision making process was solid and the UNC coaching staff is just happy the move ended up being a winning one.