If I’d told you a couple of weeks ago that North Carolina, Oklahoma and Villanova would be packing their bags for a Final Four date in Houston, you probably wouldn’t have batted an eyelid. (Assuming you weren’t a Kansas or Oregon fan.) According to FiveThirtyEight’s NCAA Tournament prediction model, the Sooners and Tar Heels were the most likely men’s teams to emerge from their regions, and the Wildcats were a strong second choice in theirs. That trio is as chalky as a group featuring a couple of No. 2 seeds can get.
Syracuse, however, would have stopped some traffic. The Midwest’s Final Four entrant is a major stunner — its region of the bracket featured a No. 1 seed (Virginia) that ranked as the third-best team in the country in our pre-tournament team ratings and a No. 2 seed (Michigan State) that many analysts believed deserved a No. 1, plus a strong second tier of teams including Purdue, Utah, Iowa State and even better-than-its-11-seed Gonzaga. Meanwhile, the 10th-seeded Orange were uncertain bets to merely make it past Dayton in the round of 64, much less forge a path all the way to Houston.
Before the tourney began, both our model and Ken Pomeroy’s odds assigned Syracuse somewhere between a 1 percent and 2 percent chance of making the Final Four. Comparing Pomeroy’s log5-based probabilities this year to archives from every year going back to 2006,1 that makes this year’s Orange the fourth-most-unlikely Final Four participant of the past 11 NCAA tournaments.
|YEAR||TEAM||FINAL 4 PROB.|
Because Syracuse had stretches of staggeringly bad regular-season play for a Final Four team, our NCAA Elo ratings — which estimate a team’s strength at a given moment — rated the Orange as the 41st-best team in the 68-team field going into the tournament. (Among Final Four teams since the tourney expanded to 64 teams in 1985, only No. 52 VCU in 2011 and No. 44 George Mason in 2006 ranked lower.) By contrast, UNC was the third-best team in the field by pre-tournament Elo, Villanova was fifth, and Oklahoma was ninth. So in terms of the disparity between Syracuse’s pre-tournament Elo and the average of the other Final Four teams, the team is the sixth-most-out-of-place Final Four entrant since 1950.2
|YEAR||TEAM||ELO||OTHER FINAL FOUR TEAMS||AVG. ELO||DIFF.|
|1952||Santa Clara||1416||Kansas, St. John’s, Illinois||1834||-418|
|1950||Baylor||1459||CCNY, Bradley, NC State||1782||-322|
|1999||Ohio St.||1893||UConn, Duke, Michigan State||2182||-289|
|1963||Oregon St.||1704||Loyola-IL, Cincinnati, Duke||1966||-262|
|1968||Ohio St.||1730||UCLA, UNC, Houston||1991||-261|
|2016||Syracuse||1772||Villanova, Oklahoma, UNC||2031||-259|
|1972||Florida St.||1751||UCLA, UNC, Louisville||1997||-247|
|1986||LSU||1804||Louisville, Duke, Kansas||2049||-244|
|1960||NYU||1766||Ohio St., California, Cincinnati||2008||-241|
|1965||Princeton||1687||UCLA, Michigan, Wichita St.||1923||-236|
Ratings aside, the team earned its way with a strong performance in the tourney, particularly when it outlasted Gonzaga and Virginia this past weekend. A lot has changed over the past two weeks; now Syracuse has nearly caught up to its Final Four peers, making it the sixth-most-improved Final Four team by Elo since 1950.
Teams like these don’t have a terrific track record once they get to the Final Four: Of the nine teams alongside Syracuse above, only three — Bradley in 1954, Villanova in 1971 and Kansas in 1991 — advanced to the championship game, and none won the title. The most improved Final Four team to win a title was Villanova in 1985. (For what it’s worth, Butler — which was the 61st-most-improved team and an even less likely Final Four entrant than Syracuse — came within a few inches of bank from snatching the title from Duke in 2010.) Syracuse will get a chance to improve that success rate against North Carolina on Saturday, and based on its performance these past few weeks, it would be foolish to count the team out again.