I spent a good chunk of my afternoon doing something I didn’t expect to be doing 24 hours earlier, parsing the profiles of the ACC top seed (don’t say “regular season champion,” since no such title exists) North Carolina Tar Heels and ACC Tournament champion Duke Blue Devils. Yes, I could have thrown the Gonzaga Bulldogs in for good measure, but I don’t think the Selection Committee is going to kick a one-loss team that went 6-0 in games against the Top 50, with just one of those wins coming at home, off the top line. And that’s the case even if the West Coast Conference isn’t anywhere near the level of this season’s ACC.
Selection Sunday Bracketology 2017: UNC/Duke No. 1 seed debate could go either way
Seriously, it’s that close.


The two teams are remarkably even, sitting right next to each other in the RPI table (UNC ranks fifth, Duke sixth). Both did well in games played against that metric’s Top 100 — Carolina went 17-6; Duke 16-7. And while the Blue Devils would be the first eight-loss team to earn a top seed, the Tar Heels have dropped seven games themselves.
Digging deeper, the Blue Devils’ Top 25 (9-4) and Top 50 (12-6) records are superior to the Tar Heels’ (6-4 against the Top 50, with all the wins coming at home, and 11-5 against the Top 50). Plus, Duke won two head-to-head matchups, including one on a neutral floor in Brooklyn.
North Carolina went above .500, 6-5, in true road games, while Duke went just 3-6. But the Heels’ overall road/neutral record of 10-7 isn’t quite as good as the Blue Devils’ 12-7.
And while it’s true Duke finished three games behind UNC in the regular season, the Blue Devils’ league slate was a bit tougher — they played 13 games against league opponents in this projection, while Carolina played just 11. That discrepancy also helped close a gap in schedule strength. Duke ranked just 88th in non-conference strength of schedule, with North Carolina 49th. But after the ACC slate is factored in, the Blue Devils grabbed the advantage, ranking third while the Heels are 11th. Super-sized conferences and the imbalanced league slates that result just make regular-season conference championships, even unofficial ones, not mean as much as they did 15 or 20 years ago.
Therefore, I’ve given Duke the nod as the final top seed at the top of the South regional. North Carolina ranks fifth overall, which means they’re destined for the Midwest. As the top No. 2 seed, the Tar Heels cannot be sent to the closest available region, the East, since the Villanova Wildcats, the No. 1 overall seed, are already there. Bracketing rules also require that they’re separated from Duke. So it’s a possible Elite Eight matchup with the Kansas Jayhawks for Roy Williams’ squad.
With that determination made and my bubble choices made earlier this afternoon, here’s my final projected bracket of the 2016-17 season.
New York (Fri/Sun) | Kansas City (Thu/Sat) |
|---|
Memphis (Fri/Sun) | San José (Thu/Sat) |
|---|
EAST | MIDWEST | SOUTH | WEST |
|---|
LAST FOUR IN | FIRST FOUR OUT | ARRIVALS | DEPARTURES |
|---|
By Conference: ACC (10); Big East and Big Ten (7); Big 12 and SEC (5); Pac-12 (4); Atlantic 10 (3); American and West Coast (2); 23 one-bid leagues
After scrubbing the field again, I made a few changes, usually of a seed line or so. I attempted to try to be as consistent with my reasoning for placing teams as possible, and hopefully the Selection Committee did the same. We will know soon enough and I’ll be back later Sunday with a review of my performance and the real bracket.











