Luke Winn's Breakout Sophomore formula predicts strong seasons for Marquette's Deonte Burton, UNC's Kennedy Meeks, Richmond's Shawndre Jones, Brown's Leland King and USF's Chris Perry.
Now in its sixth season, The Breakout Sophomore Formula™ is my attempt to identify future stars who didn't stand out in traditional box scores as freshmen, but had tempo-free stats that indicate high-scoring potential.
A certain degree of obscurity is required for a player to qualify: He cannot have averaged much more than 20 minutes per game as a freshman. But while he was on the floor, he had to use a go-to-guy's share of offensive possessions (around 24 percent or higher) with a respectable level of efficiency (an offensive rating of at least 100.0, or one point per possession). The underlying theory, as first proposed by Basketball Prospectus, is that go-to-guys tend to act like it from the start of their careers, even in limited playing time. The formula also takes opportunity into account: Has the team's roster changed enough to make room for a breakout?
While last year's edition didn't yield any major breakouts -- the best two picks, Gonzaga's Przemek Karnowski and Wisconsin's Sam Dekker, played big roles on ranked teams but weren't all-conference selections -- the formula has historically done quite well. It has IDed on-the-verge scorers such as VCU's Treveon Graham, Ohio State's Deshaun Thomas, Cincinnati's Sean Kilpatrick and Texas' Jordan Hamilton before they were stars.
Before the 2014-15 picks are unveiled, these are the prominent sophomores I deemed too obvious, since they played major, high-usage minutes as freshmen: Anthony Barber, North Carolina State; Marcus Foster, Kansas State; Billy Garrett, DePaul; Damian Jones, Vanderbilt; Rysheed Jordan, St. John's; Isaiah Taylor, Texas; and Nigel Williams-Goss, Washington. Other popular-among-pundits breakout picks such as Louisville's Terry Rozier, Arizona's Rondae Hollis-Jefferson or Kentucky's Harrison Twins and Dakari Johnson, didn't fit the usage formula. These are the five who best fit within the parameters: