By the (Advanced) Numbers: The Eagles Leave the Ground

Boston College poses a stern test for a Virginia Tech defense looking to bounce back.

Let's begin with a reflection on the North Carolina game. Getting stomped on defensively is never fun, and the performance on that side of the ball was truly embarrassing. It's easy to say Virginia Tech was down multiple safeties, but the Tar Heels ran the ball at will and if stopping the run is dependent on safety play, there's already a problem.

On the bright side, however, in spite of a slower first half no one would be complaining about a 45-point showing against UNC if they did not know the other side scored 56. The Hokies gained 6.43 YPP, a total surpassed in 2019 only in games against Rhode Island, Wake Forest, and Georgia Tech — none of whom have UNC's talent. The offense scored .577 points per play, well above their 2019 average of .443. Khalil Herbert gained 7.67 yards per rush against a Tar Heels defense that gave up 1.94 and 2.11 against Syracuse and Boston college, respectively. Oh, and he is leading the country in yards per game and 2nd in yards per carry at 10.44 on the season.

It's an unfortunately small sample size, but Hendon Hooker's passer rating — if eligible — would rank 6th nationally, just ahead of Trevor Lawrence.

Try to enjoy what we are seeing on offense this year — without it this team would likely be ending the bowl streak (if there was a wins requirement).

On to the next opponent, the Boston College Eagles...

When I think of Boston College football, three things come to mind:

  1. LB's that are tackle machines
  2. Run-heavy power offense
  3. Performance slightly above mediocrity

Unlike against Michigan, Nebraska, or Ohio State, there is not a single memorable game in the Hokies' series against the Eagles, and certainly not a specific play. None whatsoever. I keep trying to think of one and am drawing a blank here. So no need to even discuss the issue in the comments.

Anyway, the 2020 version of the Eagles features plenty of #1 and #3, but #2 is hardly recognizable. The Eagles rank 8th nationally with a 56.71% passing percentage, and leading RB David Bailey would have fewer carries per game than Braxton Burmeister had the quarterback played the second-half Saturday.

Does that mean the Hokies should fear the potential air attack given a thin secondary?

Where the Season Stands

Virginia Tech is currently 2-1. The actual and predicted score difference and odds of winning each game are:

The odds of each possible regular season win total are now:

The U back?! It is obviously early in the season; there are plenty of question marks on defense and week to week the availability of various starters changes wildly. So take this as even less predictable than usual. Still, a 94% chance of a winning record and 46% chance of finishing 8-3 or better isn't terrible.

Based on FPI, the odds of each ACC team beating the ACC teams on their schedule is:

As expected entering the season, the front runners for the ACC Championship Game are Clemson and Notre Dame. But UNC isn't far behind, and they play Notre Dame, so pencil that in as a huge ACC game. The second tier of teams — still with viable paths but needing a couple of the top teams to falter — consists of Miami, NC State (!), and Virginia Tech. The Wolfpack benefit from not playing either of the top two teams.

Virginia Tech Leads the Nation In...

Field goal percentage, with Brian Johnson at 5/5 on the season. Sure, it's early, but there are only 18 teams that have attempted at least one field goal and made 100% on the season. Interesting, BC is 5th nationally in FG% allowed at 33.33%.

Rankings and Computer Predictions

The computer rankings and predictions for each team:

Keep in mind that many rating systems are excluding teams that have not yet played, so everyone is getting a bit of a rankings boost right now. Still, every prediction has the Hokies beating the Eagles and they center around the current spread of 11. The odds of an 11-point favorite winning is 79%.

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

Neither team is thrilled with middling defensive performances thus far, but Virginia Tech has a clear offensive advantage that translates to a moderate overall advantage. Will starting Hendon Hooker at quarterback extend that gap further? Will the Hokies have enough bodies on defense to bounce back from last week's humiliation?

Who To Watch Out For

Naturally BC has playmakers at the LB position, but this team also features solid performance in the passing game. This could be interesting to watch as literally no Boston College quarterback has ever done anything memorable against Virginia Tech and we don't really need any further discussion on the topic:

  1. QB Phil Jurkovec leads the Eagles' pass-happy offense with 38 attempts per game (8th nationally) resulting in a rating of 145.14 (29th nationally)
  2. Jurkovec's top target is sophomore Zay Flowers, who averages 19.29 YPC (29th nationally) for 101 YPG (15th nationally)
  3. The LB duo of Isaiah McDuffie and Max Richardson are 2nd and 3rd nationally in tackles at 41 and 38 respectively; the pair has also combined for 9 TFL

Statistical Key to the Game

Improved offensive play hasn't been a mirage, and I haven't seen a playmaker at running back like Herbert since Ryan Williams or David Wilson. That offensive FPI rating is with the QB1 only playing about one half of a football game thus far — with Hooker's ability to stretch the field, look for the run game to get even more explosive.

The defensive struggles, on the other hand, aren't just due to COVID absences. While I tend to believe UNC will prove to be the worst defensive performance of the year, I'm not optimistic that the unit will end up anywhere better than an average defense. Hamilton requires time to get the right athletes into his scheme and that doesn't happen mid-season.

The Eagles will try to attack the secondary by air, and if successful the Hokies will have to choose whether to continue getting burned by the pass or drop more players into coverage and get burned on the ground. Virginia Tech will need to limit their opponent's YPA to less than 7.5 — not an excellent mark by any means, but enough to outscore them and walk out 3-1.

Statistical Prediction

I'm anticipating more defensive frustration, but nothing like the level seen against UNC. Assuming COVID doesn't wreck the starting team, I think Hooker and company come out firing on all cylinders and don't care that the defense can hardly get off the field because points will be easy to come by.

Virginia Tech 41, Boston College 31

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

Comments

there is not a single memorable game in the Hokies' series against the Eagles, and certainly not a specific play. None whatsoever. I keep trying to think of one and am drawing a blank here.

I concur.

I did sneak into Lane once during a bye weekend. I was surprised how many people showed up to just take in the majesty of our stadium for a few hours on such a cold, rainy evening.

Click here to destroy wall.

I remember a very specific BC game at Lane. It was the first time I'd been back at Lane in a some I was a student a decade earlier, and I ended up going home cold, soggy, and angry. See if you can guess which BC game this was.

FWIW the only other time I've made it back to Lane was the opener with Ohio State where we lost and Brewer got hurt. It's been two decades since I've seen the Hokies won in Lane.

BTW I have BC 38 VT 31. That defense isn't getting fixed this year.

BT(A)N is back!!

EDIT: (fixed) Also, I think you have your labels backwards here. Shouldn't all the predictions be on the "Hokies Win" side of things?

Warning: this post occasionally contains strong language (which may be unsuitable for children), unusual humor (which may be unsuitable for adults), and advanced mathematics (which may be unsuitable for liberal-arts majors)..

Yep you had the guys guts to point out what I didn't. I was super confused reading this chart

EDIT: damn swype inputs...

"The Big Ten is always using excuses to cancel games with us. First Wisconsin. Then Wisconsin. After that, Wisconsin. The subsequent cancellation with Wisconsin comes to mind too. Now Penn State. What's next? Wisconsin?" -HorseOnATreadmill

Edit: I have no idea what I was talking about and I'm confused as well. Another couple hours of sleep should help. 🙄

You're correct...will update soon. I have no idea why it is backwards as I re-opened the file and it is showing correctly but whatever.

To confirm: all computers show the Hokies winning this game!

I believe it's good to go now. And you're supposed to blame this stuff on me.

Oh it's so good to have this back

"The Big Ten is always using excuses to cancel games with us. First Wisconsin. Then Wisconsin. After that, Wisconsin. The subsequent cancellation with Wisconsin comes to mind too. Now Penn State. What's next? Wisconsin?" -HorseOnATreadmill

Agreed! Have been waiting for this!

Chick Patty w/ Cheese

Math says we win them all except Clemson, here's to hoping you did not miss any decimals anywhere...

I can imagine no more rewarding a career. And any man who may be asked in this century what he did to make his life worthwhile, I think can respond with a good deal of pride and satisfaction:
“I served in the United States Navy"

KCCO

Well, not exactly...math says we would be favored in all but Clemson right now, but the chances of each number of wins shows there is likely at least one other loss somewhere in there. A 75% chance of beating a team is 0.75 expected wins, not 1.

🤣🤣🤣🤣

I can imagine no more rewarding a career. And any man who may be asked in this century what he did to make his life worthwhile, I think can respond with a good deal of pride and satisfaction:
“I served in the United States Navy"

KCCO

Careful, we don't like WTF losses around here

"If you don't have time to do it right, when will you have time to do it over?"

around here

Who does?

How I Learned to Stop Worrying and Love the Jet Sweep

Not so surprised about us being favored against Miami. Their wins have looked less impressive with each passing week, and their offense's profile shows a team that has been very good on third and long (until last Saturday), BUT they put themselves into third and long far too frequently. Their 1st and 2nd down success rates are not great, and therefore have become reliant on playmaking on third down. Clemson rarely has any difficulty with teams that make it harder on themselves than it has to be. It's very much still a game we could lose, but I do not think they were anywhere close to deserving such acclaim heading into that Clemson game.

Well, the betting public likes them. They opened as 10 point favorites over pit and that's grown to 13.5

Warning: this post occasionally contains strong language (which may be unsuitable for children), unusual humor (which may be unsuitable for adults), and advanced mathematics (which may be unsuitable for liberal-arts majors)..

I don't think I would ever bet on a Pitt game. They are close to Auburn tier of "you cannot trust this team to win or lose ever by any amount of points."

That said, I don't think Pitt is good (because of their garbage offense) so that doesn't mean much to me. I think Miami is a better team than Pitt because they have dangerous athletes on offense.

I also want to point out that I don't think Miami is a bad team. I am not super confident that we will beat them, but I was moreso just commenting on why the numbers don't like them as much.

If Pitt wasn't missing their QB, I would be tempted at that line.

Unlike against Michigan, Nebraska, or Ohio State, there is not a single memorable game in the Hokies' series against the Eagles, and certainly not a specific play.

I may not be following you, but I do think there have been a few memorable games against BC. The 2007 Championship game were VT bested BC to avenge a last minute regular season rally by Matt Ryan. Then there was also the 2016 game where Jerod Evans tied a school record with five touchdown passes and the Hokies limited Boston College to 124 yards and six first downs in a 49-0 victory

Ut Prosim Ad Dei Gloriam

...

I don't know what you are talking about - in 2007 Virginia Tech only played Boston College one time in the ACCCG and beat them 30-16 and the name Matt Ryan certainly doesn't sound familiar. Please check your facts and do not report back on what you find or ever suggest again there was another game against Boston College in 2007.

Ok, I'll even give you a leg for that. ;>)

Ut Prosim Ad Dei Gloriam

Chick Patty w/ Cheese

I will read this over the next few minutes, just wanted to first say how happy I am too see BTAN again!

Whatever. It was one bad year.

Seasonal Brew means High ABV for football season and standard the rest of the year.

there is not a single memorable game in the Hokies' series against the Eagles, and certainly not a specific play. None whatsoever. I keep trying to think of one and am drawing a blank here.

2010 - The Hokies walked into the Eagles home field and shut them out 19-0 at their house. The first time in over 20 years for the Eagles to have been shut out. The battle cry of Let's Go - Hokies resounded across their stadium as we enjoyed crushing them at home.

I was there! half the crowd left at halftime to go to the Bruins' preseason game.

We were in the corner of the end zone, nine rows up from the field.

Great analysis. I have missed that "Simulated Win Totals" bar chart oh so much this season, and while I'm sure I wouldn't be "happy" with a 7-4 season, it's nice to see that the data backs up my thoughts from earlier that it's a reasonable floor for this group, considering there's over a 75% chance of finishing there or better.

One question though (and I'm sure I'm nitpicking here). I may be missing something but unless you're calculating the expected spread differently than the odds of winning, I don't understand how a jump from 63.1% to 63.2% corresponds with a shift in the spread from 3 to 6. Seems it should be almost flat between Miami and Pitt, much like the line between Wake and Louisville.

Good spot - I've been trying to figure it out and think the background reason is a difference in home field advantage this season without fans. I didn't think it was worth stressing over because "what kind of person would notice such a trivial detail?" but here we are and hats off to you, macraw83.

I'll try to use a different adjustment next time to ensure these make more sense.

Why do we have less chance of beating Clemson than UNC? I mean Clemson is good...

(And I love these analysis)

We put the K in Kwality

UNC game already happened, so 0% chance of pulling that one out now. Clemson game has yet to happen, and while the score difference is predicted to be worse, our chance of winning is still greater than zero.

Yep, this...for games that already happened I update to the actual score difference and chance of winning (0% or 100%).

Those are stats I can bet my money on

"If you don't have time to do it right, when will you have time to do it over?"

I would love to find the bookie that will let me bet on games that has happened in the past.

🦃 🦃 🦃

If you ever find yourself with some spare time, it would be neat to have the game-week FPI probabilities included with the completed game results so we can see how accurate (or inaccurate) the formula's prediction was.

I don't have that for FPI yet, but if you're curious about the Vegas Spread here are some select values (to avoid a ridiculously long response):

Closing Spread Game Count Win Record Win %
-30 44 43-1-0 97.70%
-29.5 33 31-2-0 93.90%
-29 44 42-2-0 95.50%
-28.5 45 43-2-0 95.60%
-28 81 79-2-0 97.50%
-27.5 60 58-2-0 96.70%
-27 59 57-2-0 96.60%
-26.5 45 44-1-0 97.80%
-26 46 46-0-0 100.00%
-25.5 39 36-3-0 92.30%
-25 50 47-3-0 94.00%
-24.5 51 47-4-0 92.20%
-24 113 111-2-0 98.20%
-23.5 68 63-5-0 92.60%
-23 70 64-6-0 91.40%
-22.5 47 46-1-0 97.90%
-22 38 36-2-0 94.70%
-21.5 69 67-2-0 97.10%
-21 140 133-7-0 95.00%
-20.5 102 93-9-0 91.20%
-20 93 89-4-0 95.70%
-19.5 78 71-7-0 91.00%
-19 54 51-3-0 94.40%
-18.5 77 69-8-0 89.60%
-18 73 69-4-0 94.50%
-17.5 118 108-10-0 91.50%
-17 166 146-20-0 88.00%
-16.5 145 133-12-0 91.70%
-16 75 68-7-0 90.70%
-15.5 84 72-12-0 85.70%
-15 72 57-15-0 79.20%
-14.5 172 144-28-0 83.70%
-14 263 228-35-0 86.70%
-13.5 194 160-34-0 82.50%
-13 127 105-22-0 82.70%
-12.5 101 70-31-0 69.30%
-12 101 79-22-0 78.20%
-11.5 115 93-22-0 80.90%
-11 121 91-30-0 75.20%
-10.5 184 139-45-0 75.50%
-10 241 174-67-0 72.20%
-9.5 166 118-48-0 71.10%
-9 100 83-17-0 83.00%
-8.5 163 128-35-0 78.50%
-8 119 80-39-0 67.20%
-7.5 293 199-94-0 67.90%
-7 433 304-129-0 70.20%
-6.5 329 227-102-0 69.00%
-6 216 149-67-0 69.00%
-5.5 181 120-61-0 66.30%
-5 107 62-45-0 57.90%
-4.5 188 120-68-0 63.80%
-4 243 142-101-0 58.40%
-3.5 414 252-162-0 60.90%
-3 594 339-255-0 57.10%
-2.5 361 198-163-0 54.80%
-2 179 98-81-0 54.70%
-1.5 195 94-101-0 48.20%
-1 237 117-120-0 49.40%
0 100 50-50-0 50.00%

I was just pointing out that the graph showed 0.0% for UNC but 17% for Clemson, despite the Clemson data point being lower for CU. I get now that it's because the UNC game had a higher win probability (they don't reset after a win or loss). Thought it was a funny analysis that we had less than zero chance to beat Clemson.

We put the K in Kwality

I think you're still looking at it wrong. The y-axis isn't % chance of winning, rather the predicted (or actual for completed games) score difference. Positive for wins and negative for losses.