Sharp Analytics uses an advanced statistic called Beta_Rank to rank teams both before and during the season.
Not to get too into the weeds, but here's a simplified breakdown of what it measures:
Beta_Rank = Offense + Special Teams - Defense
Offensive = Drive Efficiency + Play Efficiency + Explosiveness + Limiting Negative Drives
Negative Drive = <3.3333 yards per play on a drive
Defense = Opponent Drive Efficiency + Opponent Play Efficiency + Opponent Explosiveness + Negative Drives
Special Teams = Punting + Punt Return + Punts inside the 20 + Kick Return + Kickoffs + Field Goals
Beta_Rank puts a lot of weight into strength of schedule. It doesn't consider big games against poor teams, let's say Georgia Tech, but it does consider bad games against poor teams, say ODU or Furman. The goal is to measure how teams will perform against high quality opponents.
For those, like myself, who are interested in getting nerdy with it, here's the full explanation from the creator of Beta_Rank, Rob Bowron.
https://www.sharpcollegefootball.com/post/what-is-beta_rank
Here are the latest 2020 projections for ACC teams:
#2 - Clemson
#12 - Notre Dame
#18 - UNC
#35 - Virginia Tech
#36 - Florida State
#41 - Louisville
#42 - Miami
#47 - Boston College
#50 - Virginia
#55 - Pittsburgh
#59 - NC State
#73 - Duke
#78 - Wake Forest
#91 - Syracuse
The projection takes into account a recruiting measurement as well as returning production, a measurement that Tech ranks 5th in the country.
Here's the simplified breakdown of Tech's measurements and rankings:
Predicted Beta_Rank - 0.56
Offense '19 Rank - #58
Offense '19 - 0.09
Defense '19 Rank - #42
Defense '19 - (0.26)
ST '19 Rank - #42
ST '19 - 0.04
Team Returning Production Rank - #5
Offense Returning Production Rank - #36
Defense Returning Production Rank - #6
Recruit Rank - #44

Comments
Yeah, not sure why an Independent team is here in the ACC rankings.
Thanks for sharing and this looks like an interesting way to rank teams, but IMHO Notre Dame should not be included in the ACC.
They play multiple ACC teams every year, so the measurements are greatly impacted by other ACC teams.
I can't argue that they play multiple ACC teams, but ND fought really hard to not be included in the ACC, so I don't think we should give them the privilege of being included.
I think it would be a useful (for bench marking purposes) to compare against any team that plays a third of the conference in a given year.