By the (Advanced) Numbers: North Carolina

The result of the Techmo Bowl was bittersweet for me. Without question, the win was huge for Virginia Tech and its bowl odds (more on that below). However, after learning of Frank's retirement I wanted the Hokies to play an inspired game; left to believe they might be really dominant down the stretch. Instead, it was a game that could have been lost and the offense especially left me no more confident than before kickoff.

Now Tech returns to Blacksburg for Frank Beamer's last game coaching in Lane Stadium. The Hokies must face 9-1 North Carolina, one of the hottest teams in the country. I expect the atmosphere to be perhaps the most electric I've ever experienced. Casinos have set the opening spread at an unexpectedly narrow 6.5-point UNC victory — are they taking into account the emotions and fan intensity at the game, or do objective computers also agree that this game isn't as lopsided as it looks on paper?

Where the Season Stands

Virginia Tech is currently 5-5. The actual and predicted score difference and odds of winning each game are:

The odds of each possible regular season win total are now...wait for it...:

Needless to say, the team's bowl odds took a bit of a jump after the win against Georgia Tech:

The question has been raised as to whether a 5-7 Virginia Tech team might make a bowl, which of course assumes that any 5-7 teams are allowed to go to a bowl. There is no clear answer at the moment, as the NCAA rule regarding selection of 5-7 teams is not written clearly enough for bowls to understand. The NCAA is expected to clarify the issue this week. If APR is ultimately used, then the team has a reasonable chance. Among potential 5-7 teams Virginia Tech's APR rates fairly well, but not at the top. If APR is not used, then this becomes very subjective. There would be obvious appeal for a bowl to take the Hokies and host Frank Beamer's last game, but another 5-7 team like Nebraska may also be a huge draw for tickets sales. The best scenario is for Tech to win its sixth game and not have to worry about it.

Rankings and Computer Predictions

The computer rankings and predictions for each team:

North Carolina is clearly rated as the better team, bordering on top 10. The odds of a 6.5-point underdog winning are 31.4%.

Next is a look at any overall offensive or defensive advantages:

Clearly the Tar Heels bring a very good offense to Blacksburg, but despite the compliments Larry Fedora may give to DC Gene Chizik they remain a mediocre defense. When each offense is compared to the opposing defense, it becomes clearer how the Hokies find themselves only a 6.5-point underdog to a 9-1 team. Even considering North Carolina destroyed its last two opponents (66-31 over Duke, 59-21 over Miami) who beat the Hokies.

When Virginia Tech Has the Ball

Here is an explanation of S&P+ ratings, and FEI ratings. All statistics are now opponent-adjusted.

Who has the advantage in the passing and rushing game when the Tech offense has the ball?

The Loeffler offense is just so average. Fortunately, it appears the Tar Heels can't stop the run. In fact, they are one of the worst defenses in the entire country at defending the run, so Beamer's intuition to feed McMillian 30 times should produce some offensive movement, kill clock and limit possessions: Tech doesn't want to get into a shootout with UNC.

Offensive and defensive line performance are compared using Football Outsiders' metrics:

The Hokie o-line continues to make running room, while simultaneously exposing the quarterback to far too many sacks. Unfortunately the Heels' defensive line matches the VT strengths, except for power success rate; in third- or -fourth-and short, look for Tech to get the yards they need to convert.

Now let's take a look at the FEI personality traits of the Hokie offense versus the Tar Heel defense:

North Carolina is at least Virginia Tech's equal in these measures, and has a clear advantage in explosive drives. Fortunately scoring fast is likely not part of the Hokie game plan.

The FEI No. 91 Virginia Tech offense is closest in personality to:

1. Louisiana-Lafayette
2. Army
3. Oregon State

The FEI No. 58 North Carolina defense is closest in personality to:

1. UCLA
2. Ole Miss
3. Toledo

When North Carolina Has the Ball

Again, examine pass-run comparisons first:

While Marquise Williams is a big quarterback who is a threat to run, the UNC offense stands out more in the passing game and is likely to test the Hokies' inexperienced secondary. There is no single receiver that can simply be shut down by Brandon Facyson; multiple receivers are a threat.

Offensive and defensive line performance are again compared using Football Outsiders' metrics:

At least some of the success of the Tar Heels' passing game starts with the offensive line, a very good until in all aspects except for power success. Fortunately for the Hokies, that is a strength of their defense, and converting on third- or fourth-and-short may be difficult for UNC.

As for personality traits:

The Tar Heels are good in all phases except for methodical drives, which naturally happens to teams that are explosive (it is difficult to have 10+ plays on many drives when a high percentage of drives average 10+ yards per play).

The FEI No. 65 Virginia Tech defense is closest in personality to:

1. Temple
2. Notre Dame
3. Appalachian State

The FEI No. 10 North Carolina offense is closest in personality to:

1. Western Kentucky
2. Notre Dame
3. Baylor

Special Teams

First we look at the Hokies' kicking units:

Although the unit has slipped from No. 1 since last week, only in the kickoff phase is Virginia Tech an average team.

When the Tar Heels kick:

This is also an excellent special teams unit, except in punting.

Who To Watch Out For

Every offensive skill position player?:

1. RB Elijah Hood is certainly deserving of attention, but when QB Marquise Williams runs he is even better. Williams averages 8.4 YPC. Not that he needs to run: he's completing 66% of passes for 9 YPA.
2. Let's just go with all four WRs: diminutive Ryan Switzer is catching 71% of targets for 10.6 YPT; the three big men are Quinshad Davis (71%/11), Bug Howard (54%/9.9), and Mack Hollins (65%/17.2)
3. CB tandem Des Lawrence and M.J. Stewart have combined for 4 INTs and 24 PBUs on the season

Statistical Key to the Game

I normally hate third down percentage as a statistic, because it is really more of a measure of total offensive movement than anything special about third down (teams with a high conversion percentage have it because they have a low YTG average). But in the case of this game, the stark differences in power success rate between the lines lead me to believe that could be the game changer; whichever team is able to convert on third-and-short more often (assuming UNC has to face third-and-short) has a great shot.

Statistical Prediction

The statistics predict Virginia Tech will finish with 163 yards passing and 199 yards rushing, and North Carolina will finish with 180 yards passing and 208 yards rushing. That's a narrow margin where field position or turnover luck can easily decide the game — I'm looking for UNC to be overly confident and tip those scales in VT's favor.

Virginia Tech 31, North Carolina 27

As always a thanks to Football Outsiders, cfbstats.com, and Minitab Statistical Software.