So, I had noticed that there wasn't a thread for this season (based on this thread from last year) earlier today while talking to a friend of mine. With that in mind, and all credit to piercepj, I went ahead and tried to gather as much information as possible to re-create these graphs. The sucky part is that most of this data is not archived (as far as I can tell), so I had to search for the best ESPN FPI data that I could, and I recorded the most recent data from the sources that piercej used last time and will update those graphs accordingly going forward. With that in mind, this week it's only going to be ESPN FPI's data.
EDIT: I just got an S&P+ graph up, but it's not looking pretty. I got it from this thread.
EDIT (9/29/18): I changed the S&P+ back to a percentage, I'm using a normal distribution formula in Excel with a mean of 0, and a STDev of 17. This is from the same reddit post above, I just found some time to create the formula.
I'll try to edit the posts and repost images and update on a weekly basis, but if anyone feels that a new post every week would be better, let me know.
Without further ado here's the graphs:
Week 8: Everything still looks sorta ok from the 4-letter, but I don't think they really watched Thursday's game...
Week 8: Not much change here, which is kinda the point of these ratings I think. We did get a little Miami bump.
Week 8: Welp, Massey has lost all confidence in VT, which is sad since we're his Grad School alma mater...
Week 8: Not much change here surprisingly.
Comments
Pink for LOLUVA, good choice
Hey its 2018! UVA is allowed identify pink as an inferior or a superior color!
/s
Pitt is way too high. They were in that game fir a half with PSU then the wheels fell off. I would rank it the third most likely loss currently after ND and Miami
They were in it until their coach that has anger issues blew a gasket and they sucked from that point on
I turned to that game right as he got flagged. Then they talked about how he mentioned earlier in the week that Penn State receivers push off. Fuck that man.
Is it possible that it's just a misunderstanding? Like when he says "the receivers push off" do you think he means "the receivers rekt my corners which I put on an island"?
Excellent work. It's so beautiful.
Given Duke's recent injuries and the few weeks we have to prep vs lesser opponents, I gotta think we'll be a 10pt favorite or so going into that game. This assumes we handle ECU & ODU, continue to progress and mature, and avoid injuries. May not be the case but that game is setting up nicely. Let's start with a big win over ECU this weekend.
Yeah, but Duke with those injuries, traveled to Baylor and beat them.
The same Baylor whose best win so far this year is against UTSA, and who already has twice as many wins this year as they had all of last year. Baylor is nota good football team.
I just updated the charts, they were a bit too hard to read IMO, and I also added the S&P+ chart as well.
The S&P+ rating leans on last year's results as well as returning starters and recruiting fairly heavily until 4 or 5 weeks in to the season. I think the low number of returning starts on defense is weighing down the Street&P+ rating right now.
They include their preseason projections the entire season, but the proportion of preseason-to-in-season weight starts shifting dramatically after only a few games.
In the words of Bill Connelly, "programs tend to gravitate towards their historical mean", hence the inclusion of preseason projections even in their end-of-year rankings.
You're doing the lord's work Miner (with all gratitude to piercepj for taking charge of these the last several seasons of course)
Seeing our odds against GT drop after this weekend is puzzling.
Great work here, but I have a request for the sake of the mobile version of this site. Can you go back and edit the title? The lack of space in "PROJECTIONS/PREDICTIONS" is causing the title field to stretch too far on the tracker.
Done
Graphs have been updated as of this afternoon, but I think some of the data will need to be re-accessed later this week. I'll keep an eye on it.
I like the Kevin Flag image to end the ECU line but should those win probability lines just stop instead of bottoming out? There might be a misconception that ECU won with the way the graphs display now.
I've thought about it, but at the moment it would crowd the graphs in my opinion, and with all of the (unfounded) social media hate we got for Whit's response to ECU canceling, I'd say they won lol.
Why the hell do we only have around 50% chance of winning against Duke lol
CAUSE WE AINT PLAYED NOBODY PAWLL
True
Road game and Duke looks decent
Doesn't take into account it's a road game in location only. And Duke is getting hammered with injuries.
I'm not sure how to feel about being predicted to win every game from here on out. On one hand, awesome... On the other, that's a lot of pressure to put on a team this young.
I'm not impressed with the Fighting Irish. Their defense is suspect, and their offense doesn't scare me to compared to our defense. I don't like to predict big games. I don't want to be let down, but I will feel very confident if we can get through ODU and Duke.
Since this will update throughout the year, my expectations I can check back on:
VT 64, odu 13
VT 20, Duke 17
VT 24, ND 16
VT 38, unc 20
VT 27, GT 17
VT 31, bC 27
Pitt 26, VT 10
Miami 28, VT 24
VT 44, uva 27
Bud allowing the hoos to get 27?
Their offense looks pretty good.
And maybe it's garbage time points.
They scored 16 against Indiana so I'd go with a "to be determined" on that one.
It was raining sideways, so "UNC Rules" apply.
He also gave UNC 20 points against us. They haven't scored 20 points against us the last 2 years combined, and they aren't a better team this year.
VT 64, odu 13Uh, no.VT 20, Duke 17Rev: Duke 27-17VT 24, ND 16Rev: VT 34-30VT 38, unc 20
VT 27, GT 17
VT 31, bC 27
Pitt 26, VT 10
Miami 28, VT 24
VT 44, uva 27Rev: VT 35-31Game I am most worried about is BC at this point.
BC and Miami, I think they will be better than they looked against LSU. ND is a complete toss up to me. I know they have loads of talent, but they just don't put a consistently good performance on the field. Hopefully the crowd can stay on them when they come to Lane.
I have zero faith in Rozier. If we can limit turnovers and get pressure on the QB, which doesn't seem to be a problem against Miami this year, that should be a very winnable game for us.
That sounds like another certain team we've beaten this year...
I'm really behind this week with updating the graphs, it's almost as if I've been in a depressive and anti-social state since the beginning of this football week.... no idea why though, surely the team didn't lose as a ~98% favorite, right? RIGHT?!
Updating these graphs after a heartbreaker was always the worst part of my week. God bless you.
Take heart, they're sure to look better next week!
Course correction incoming
Charts are updated for Week 5. Man it feels good seeing projection lines going upwards with a large slope!
I think that ACC took a bigger swing than any
I've been a bad TKPer....charts are finally updated lol.
I was a good TKP'er this week, projections are updated a few days ahead of the game...as depressing as they are.