Longtime readers will know that I’m a big believer in Bill Connelly’s SP+ metric, as it evaluates teams on efficiency metrics and controls very well for strength of schedule.
Nate Manzo posted a tweet that got my attention this morning, as it laid out the teams projected to move most from the end of the 2019 season rankings to the 2020 preseason set, posted last week.
Can you do one from preseason 2019 to current projections?— Chadwick (@Chad_Floyd) February 20, 2020
With Carolina’s season-ending rank of 34th, the jump up to 17th in the projections is a massive leap— among top-40 teams, only Florida State has a greater projected jump (because under-performance under Willie Taggart while maintaining good recruiting, but I digress.) As Connolly’s numbers lean heavily on returning production and recruiting rankings, this checks out. The teams listed as gainers all underachieved based on their 5-year norms...or bring back a lot of talent.
Ever the curious sort, I wanted to see how these numbers compared to the preseason (again, not postseason, which Manzo did excellently) projections from 2019. I figured he could do the work for me...then I went down a rabbit hole and did it myself.
In a rare journalistic win, what I found was that my theory checked out.
North Carolina football is on a better trajectory (by far) than anyone in the Power Five, and is only outpaced by Navy and Central Michigan by any ranking.
Based on raw ranking, Carolina was 66th heading into the 2019 season with an SP+ (the points better than the “average” FBS team) of 0.7. For 2020, the ranking is 17th— a jump of 49 spots into the lower tier of the elite class of college football. Navy jumped 59 spots after a bounce-back 2019 season, Carolina jumped 49, and the only other Power Five teams in the top 10 were Louisville (+41) and Illinois (+31).
This is all a convoluted way of answering the question, “Which team improved the most based on their 2018 results in 2019, and also is set to continue its ascent in 2020?” Easy enough, right?
The SP+ metric boosted the Heels from 0.7 to 16 points above average, a jump of 15.3 “points” per game— third in the nation behind Navy’s 17.2 and Central Michigan’s 17.1. And check out that gap between UNC and number four on the list, Western Kentucky.
In plain English, what does it mean? What the Tar Heel Blog staff, ‘way too early’ preseason rankings, and pundits around the nation have been telling you: North Carolina football is on the come-up in a BIG way.
When compared with its peers in the ACC, the Carolina football program is set to make, by far, the biggest jump in the Coastal. This makes sense when you consider the depths of hell they went through in 2017 an 2018, but also serves as a reminder that a team with a recruiting uptick and a massive amount of returning production should be on the rise:
And, as it relates to the “average” college football team, the numbers are even more striking in UNC’s favor:
Now, a cynic might warn that this is an over-correction based on a one-year sample— and they wouldn’t necessarily be correct!
For posterity, some numbers in the opposite direction: Syracuse was not as good as its top-25 AP ranking going into last year after a 10-win season, and Connelly had them 53rd going into 2019. After a lackluster 4-8 performance, Syracuse is projected to be 11.2 points per game worse than their last projection because they were substantially worse in 2019— and the factors weighed in this formula saw a regression, just perhaps not one that stark.
SP+ hedged against the Orange’s 2018 performance because the underlying numbers didn’t tell the whole story about their season— and to a degree was proven right, but maybe was not right enough.
As it relates to the subject at hand, SP+ sees a huge rise for the Tar Heels, as their on-field performance far exceeded what the numbers expected. They rose from 66th in the preseason to 34th at the end of the year.
Even so, the #17 ranking for 2020 seems pretty optimistic, until you consider that the Heels rank 18th in the country in returning production, behind a lot of teams that rank in the top 40 in year-over-year jumps (among the 17 ahead of UNC, only Houston projects to be worse in 2020 at a whopping 0.1 points down from 2020.) Factor in stabilized recruiting rankings from their #18 finish with the 2020 class, staff continuity (not measured by this system), and their late-season performance in blowout wins against N.C. State and Temple, and the formula spits out a team going places.
Places that, in fact, have broken the numbers for a year-over-year change in projection.
Places that the Tar Heels have not been in a long time.
Multivariate analysis is not for everyone, and the tl;dr version of this post is this: Buckle up. Carolina football is on the rise.