What happens if you pick an arbitrary statistic, like say effective points field goal % (eFG%), and use it as a benchmark for who should win each game of the big show? Well it gives some interesting predictions.
Lets start with the biggest headline this produces. It says that Florida Gulf Coast could plausibly win its play in 16 seed and beats UNC. Yes I just typed that.
Using sports reference’s eFG% UNC and Florida Gulf Coast are in a tie at 52.1%. Now if you look at some other stats like their simple rating system (SRS) it’s not even close, but assuming both teams shoot near their average attempts from beyond the arc and inside the arc and make close to their season averages, this might be a game to watch.
Not any less strange is #2 Xavier vs. #15 Weber State. This one is actually heavily in Weber State’s favor, with them shooting 56.3% and Xavier at 51.5%. Then looking at opponents points per game Weber State once again actually holds an advantage with both teams allowing exactly 2,273 points this season but Weber State having 2 extra games on their schedule, (66.9 vs 71.0). Keep in mind two things here though, the first obviously being that Xavier has played a much tougher schedule than WS has this season. More interestingly though it means that Xavier typically allows more points than WS is even expected to score, which should suggest a higher than average scoring game for them and vice versa for Xavier.
Stephen F. Austin vs West Virginia: SFA comes in at 55.0% eFG% while WV comes in at a middle of the pack 50.2%. Of all the conferences with some upset potential the East certainly takes the cake. WV really falls surprisingly low as far as FG% goes, they make up some serious ground as far as rebounds and steals however.
Michigan vs. Notre Dame: Assuming Michigan wins its play in game, they sit at a damn good 55.1% where Notre Dame falls behind at 53.5%. Keeping in mind that their SRS rating is nearly identical (14.36 for Michigan and 14.25 for ND), their projected PPG is nearly identical (65.9 vs 65.4), strength of schedule, and rebound percentage among others, this game could easily come down to a matter of who shoots the ball at a higher percentage.
If by some miracle all of this actually were to come true it would result in an amazing set of games down the line including Maryland vs. Kansas, Miami vs. Arizona, Oklahoma vs. UNI and Oklahoma vs. Duke, Michigan vs. Indiana, and Indiana vs. Iowa State.
My championship ends up being Oklahoma vs Indiana, 68-62 Sooners. Boomer Sooner!
EDIT: Just to add some wight to the concept, if you look at eFG% it would’ve predicted a close game between Hampton and Iowa State in 2001, George Mason over UConn in 2006, and VCU over purdue in 2011 (but has Kansas thrashing them so shit...).