Blazing through the offseason like...
ESPN released their initial 2017 FPI rankings after National Signing Day.
Preseason FPI is designed to take the guesswork out of preseason ratings. It is an automated ranking intended to measure team strength going forward. It is not a ranking of who will have the highest win total (which is dependent on schedule) or who is most likely to make the College Football Playoff.
The model comprises four major components: the last four seasons of performance on offense, defense and special teams, with the most recent season counting most; information on offensive and defensive returning starters, with special consideration given to a team returning its starting quarterback or gaining a transfer quarterback with experience; a four-year average recruiting ranking of four systems (ESPN, Scouts, Rivals and Phil Steele); and head coaching tenure. These four components interact and are assigned different weights depending on the team to produce preseason FPI.
Combining all of the factors above produces a predicted value on offense, defense and special teams, which represents the number of points that each unit would be expected to contribute to the team's scoring margin if it were to face an average FBS team on a neutral field.
Five ACC teams comprise the top 25: #4 Florida State, #7 Clemson, #14 Louisville, #16 Miami and #22 North Carolina State. The Hokies are 31st and West Virginia is #53.
Comments
It might just be me, but I really hate ESPN's self-branded ratings/rankings of everything. QBR is stupid too.
It's like the Weather Channel trying to name winter storms to generate their own hype and legitimacy.
Winter Storm Bronco gets a lot of hype then just leaves 1 to 2 inches.
Not the first time Bronco has disappointed with 1 to 2 inches
Does this mean that they weight the different factors differently on a team by team basis? That doesn't seem very statistical.
Well, they mean more if you are a blue blood or someone we have to hype up for kickoff. Wouldn't do if our ratings were down.
There are so many things they could have meant by that. They could just mean that the weights change depending on how they interact with each other.
https://en.wikipedia.org/wiki/Interaction_(statistics)
Sounds like they use machine learning, but wrote in a "dumbed-down" way that didn't really translate well.
Seems legit. We lost a lot of firepower on Offense and Beef on Defense so it will be up to F & F to prove themselves again that last year was not a fluke.
I have no problem with us at 31. #16 Miami and #22 NC State is what throws me off. And #14 Louisville for that matter.
*checks schedule*
*sees that ncsu and usce are playing in charlotte opening weekend, probably on espn*
Believe it or not, there is a growing shift towards an ACC circlejerk, particularly from ESPN. NC State all year got propped up as a better team than they actually were. They totally should have beat Clemson, but they followed that up by allowing BC to break their almost two season long ACC losing streak... so you tell me. They have a good DL, and almost beat Clemson, so #22 somehow.
Well we can almost beat Clemson with the best of em
Losing to Clemson is the bane of my existence and has been since 2011.
This...especially since we most definitely did NOT lose to Clemson before 2011
I remember when ESPN had 3-5 Ole Miss in the top 15 of FPI.
What your not impressed they lost to good teams?
*you're
ESPN has been trying to tell you, what part of quality losses don't you undestand? /s
So did just about every other advanced metric. They were playing decent football (though 15 probably was too high)
Those losses outweighed the wins in their favor.
Miami and NCST in top 25!
Not sure how I feel about FPI yet, but ESPN's SOR (strength of record) ranking is a joke.
FTFY
http://www.thepredictiontracker.com/ncaaresults.php
FPI did pretty good. It has been since it launched.
It's marginally better than the opening line.
I don't mind us at 31
I do mind us at 31 when Miami is 16
If your system has VT at 31 and NCSU at 22, your system is worthless
Computers actually liked state last year and loved Miami. FWIW S&P+ preseason projections have us at 25, NCstate at 27 and Miami at 18.
Computers actually liked state last year and loved Miami.
That there is the best reason to take these preseason rankings with a grain of salt. Miami always gets hyped because they're "da U", and their recruiting classes look good on paper, so I can kinda see why they get propped up in the rankings, but I don't understand the NCSU ranking at all. I mean, their recruiting hasn't been stellar by any means, and they lost some of their top players, too.
So, they're saying Da U is back?! #16
)
backbakBuried by Hokies.....that's appropriate for a QB playing with a ball signed by a midget robot.
Everyone knows that the best measure of how well a team will do is the vegas O/U on their season wins...seems like they're pretty accurate most of the time...and this is 100% serious
Yessir, Vegas has too much incentive (enormous amounts of money) not to have the best of best analytics.