I'm not sure if any of you remember, but last year I made a post detailing the ACC Coastal Division championship scenarios. It is now a week later than it was when I posted it last year, but nonetheless each of the 7 teams in the Coastal (yes, even 0-4 Duke) still have a shot at going to Orlando in early December.
Right now, there are 15 ACC games remaining featuring at least one Coastal division team, meaning there are 32,768 distinct outcomes for the remainder of the ACC Coastal season. Right about this time every year since around 2008 I've mad a spreadsheet to tally up which outcomes are won by which teams (or how many end up in a tie that isn't broken until the new "computer rankings" tiebreaker step, which is impossible to calculate for obvious reasons), so I have a reference to go by to see what teams have the best shot of winning. Here is a list of how many outcomes are won by each team:
| Team | Outcomes |
|---|---|
| Virginia Tech | 18,916 |
| North Carolina | 10,704 |
| Georgia Tech | 2,018 |
| Pittsburgh | 675 |
| Virginia | 328 |
| Miami | 50 |
| Duke | 2 |
| Rankings | 75 |
Last year, I added functionality to my spreadsheet to calculate relative odds of each outcome happening, which I then used to calculate a more realistic odds for each team to win, because obviously the "every game is an upset" outcome is numerous orders of magnitude less likely than the chalk outcome. I used Sagarin's "Predictor" rankings to set the spreads. Here is what each team's odds look like in that case:
| Team | Odds | Percent |
|---|---|---|
| Virginia Tech | 1.22:1 | 82.0% |
| North Carolina | 5.79:1 | 17.3% |
| Georgia Tech | 152:1 | 0.656% |
| Pittsburgh | 1810:1 | 0.0551% |
| Miami | 12,800:1 | 0.0000783% |
| Virginia | 16,300:1 | 0.0000615% |
| Duke | 17,100,000:1 | 0.0000000585% |
| Rankings | 21,900:1 | 0.0000457% |
Here is a list of the different combinations of tied teams and the number of outcomes of each combination that result in a tie broken by rankings (not all instances of each combination result in such a tie, this is just including the ones that do):
| Teams | # Outcomes |
|---|---|
| UNC, UVA, VT | 51 |
| Pitt, UVA, VT | 8 |
| GT, Miami, UNC | 6 |
| GT, UNC, Pitt | 4 |
| GT, UNC, Pitt, UVA, VT | 4 |
| GT, Miami, UNC, UVA, VT | 2 |
We all knew that at this point we are the favorites to win the division, but I was surprised to see us at 82%. At least until I crunched the numbers and found that we have a 56% chance of winning our last 3 ACC games, which would clinch the division. By comparison, UNC only has a 31% chance of winning their last 3 ACC games, which helps in the unlikely (18%) chance that we drop at least one of these last 3 games.
Comments
In case anyone is wondering, both of Duke's winning scenarios arise from an absurd 6- or 7-way tie at 4-4.
There are 14 games that must go a particular way for this to work out, separated by week:
Duke beats VT
GT beats UNC
Miami beats Pitt
Duke beats UNC
Pitt beats Clemson
GT beats VT
Miami beats UVA
Duke beats Pitt
UVA beats GT
Miami beats NC State
NC State beats UNC
Duke beats Miami
Pitt beats Syracuse
UVA beats VT
If all of the above happens, the UVA @ Wake Forest game this Saturday doesn't matter, which accounts for the two outcomes.
But can Miami beat UVA? That is the question.
Neat little piles of chaos Mr B
Yay numbers! Yay being in the driver's seat!
This is awesome. Thanks for compiling
Thanks for reading! I started doing these in 2008 for my own benefit to explore all the ways we could still make the ACCCG after having lost to Miami to drop to 3-3 in the conference (Miami improved to 4-2, so it looked very bleak), and have made one most years since. Just last year I realized TKP was probably the perfect place to disseminate this info to the internet.
By the way, in case you don't remember, in 2008 we beat Duke and UVA down the stretch to finish 5-3, and Miami dropped their last 2 to GT and NC State to finish 4-4. We beat GT on the head-to-head tiebreaker, and went on to get revenge on BC in the ACCCG for the second year in a row, and beat Cincy in the Orange Bowl to continue the 10-win streak by the skin of our teeth.
What an amazing, but sometimes terrible (at times), season. One of my all time favorite group of guys because it was almost an entirely new defense, held together by Macho Harris, one of the only holdovers from the 2006-2007 glory defense. All new receivers, won games by all facets stepping up at certain times. Definitely not one of our best or most talented, and statistically abysmal on offense, but an easy group of guys to root for after an enormous amount of attrition to the draft and graduation.
This team won so many close games, and is often my go to example of how some groups of personnel just find ways to win and close out games. Not a knock on any of the teams during our slide, but a praise to that 2008 misfit group of one year starters and rookies, and the coaching staff for helping keep them focused and believing they could win.
I think you misspelled "snowball in hell"
No, no, he got it right.
Even Han Solo's Nope-ing out of that one.
We all know this is the best Han Solo gif:
I like to keep it simple.
The result of any calculation I can make just tells me that it's better if we go 1-0 this week.
You've really bought into the whole 1-0 thing haven't you
It's my new caveman religion.
It worked in 1999.
Y'all math experts over here doing your crazy mathamatical calculamatations with spreadsheets and stuff and I'm just over here like
via GIPHY
What is "Total number of fans at a LOLUVA spring game"?
I'm no math whiz, but I can count to 10 thank you very much! (fingers only needed occasionally)
Who draws a 5 from the bottom down? That kid is not going far
Well given he doesn't have an opposable thumb, are you surprised?
Okay, am I missing the joke again? The clip only shows the last stroke.
I think the only number relevant to us going to the ACC Championship game is going 1-0 this weekend......
Is this from the future when the Rose Bowl is a playoff game and Fuente leads us there? Because if so, count me in.
*subconsciously tried to read this in Morgan Freeman's voice*
There's your problem. You need to read it consciously, out loud in a Morgan Freeman voice.
Bulls eye every time.
No way I can copy that voice man
Easiest things to know for this weekend:
Atlantic
* BC is already eliminated from the Atlantic.
* The loser of NC State/Florida State is eliminated
** The winner is eliminated with a Clemson win over Syracuse or a Louisville win over BC
* Syracuse is eliminated with a loss to Clemson
* Wake Forest is eliminated with a loss to LOLUVA and either a Clemson or Louisville win
Coastal
* Duke is eliminated with a loss to the Hokies or a UNCheat win over Georgia Tech
* LOLUVA is eliminated with a loss to Wake Forest and either a Hokies or UNCheat win
* Miami is eliminated with a loss to Pittsburgh and either a Hokies or UNCheat win
* Georgia Tech is eliminated with a loss to UNCheat
* (fairly certain) Pittsburgh is eliminated with a loss to Miami and wins by both the Hokies and UNCheat
FSU and NC State are both already eliminated. The winner of that game would have to a force a tie at 5-3, and Louisville and Clemson would have to be included in that tie. Even if WF and/or Syracuse gets roped into a 4 or 5 way tie, neither FSU nor NC State would have enough tiebreakers to win out.
(I spent way too much time last weekend doing the projections.)
I know it would be very detrimental to the perception of the ACC as a whole, but I kinda want Syracuse to win out (starting with Clemson this Saturday). That would set us up a rematch for the ACC crown in a stadium that is anything but that cursed place up north.
Might as well dream big because no matter who they play, if the Hokies win the ACCCG it's going to be detrimental to the league.
If by "detrimental" you mean ACC playoff chances.
I was messing with FiveThirtyEight's playoff predictor and if Alabama wins out and Virginia Tech wins out there was around a 50% chance of VT making the playoffs. That seemed better than I had expected.
Don't care. Burn it all down.
Go 1-0 a few more times. Boom see you on New Year's Day.
Beat Duke
I've been waiting for this information since the day after the BC game.
Well, we do all know the #goACC motto:
In case anyone was wondering, we got the exact results we needed in the 4 games this weekend to mathematically turn the Coastal into a 2-horse race. I wonder if there's ever been a time when 5 of 7 teams were mathematically eliminated in one week.
Anyway, with 3 weeks to go, there are 11 games involving Coastal teams remaining, meaning 2048 possible outcomes, with percentages pending updated computer rankings:
I like our chances.
The way I see it, VT has a ranking, has been looking shaky, and will therefore have a big, old target on their back.
The teams VT has left on their schedule have nothing to lose. They'll be playing as if the VT game is their whole season, and they'll be up for that game. VT better play that way as well, particularly vs UVa.
I'll be looking to go 1-0 this week.
For LOLUVA it literally is their whole season. From the hype of BM turning around the program, to the reality of what everyone outside the program knew would happen this year, a win vs VT would completely justify their entire season.
For ND, VT and USC are the last chances to make up for a grossly disappointing season. So in a way, those 2 games are their entire season.
GT needs a W to go bowling (likely get it vs LOLUVA unless JT has a relapse of his allergy to the football that provided da U with 2 easy scores by the D). The reality is that UGA is their whole season, though. Go bowling or not, beating UGA makes for a successful season.
To add to your ND point, they must win out to be bowl eligible. Right? Or does ND have a special clause where they can go bowling whatever their record is?
If they're high enough on the list, they could get selected at 5-7 if there are still vacancies.
But I think ND has too much pride to accept a bowl invite at 5-7.
Back in 2009 they declined a bowl bid at 6-6.