Alternate title for this post: Mega Jinx. But, I just can't help myself. As you know, I'm a big fan of Moneyball. You might also know that I'm a huge UNC fan - and they happen to field a basketball squad. So I'm hearing all these so-called basketball prognosticators on every possible sports media outlet prognosticate about how Washington State vs. UNC on Thursday night is going to be this great game - that Washington State "slows games down", and that could give UNC's fast-paced offense problems - especially since UNC plays "spotty defense". HELLO - has anyone actually looked at the STATS!?
So, with the keen appreciation that anything can happen in any one game (*me hoping to avoid the jinx*), allow me to do some of my own prognosticating - and, in the spirit of Moneyball, I will back-up my claims with empirical evidence. Here goes: Washington State has scored more than 76 points in exactly ONE game this season, and that was back in early November. UNC has scored 76 points or under only SIX times this season (half of those in November), and they won ALL SIX of those games. So, Washington State will have to play its best offensive game of the season (and hope UNC plays "spotty defense" - more on that later) AND most likely hold UNC to its worst offensive output of the season in order to win. To be fair, Washington State actually did hand their 2nd and 3rd round opponents (Notre Dame and Winthrop) to their lowest point totals of the season, so it can be done (if lightning strikes 3 times in a row).
But total points isn't the most interesting statistic (neither is avg. number of possessions per game, where UNC has 28% more than Washington State). Both points and possessions are naturally going to be higher even for UNC's opponents, because UNC sets a fast pace. More comparative statistics are: what is the likelihood that each team will score on each possession, and what is the avg. points-per-possession? In both categories, UNC leads Washington State. So, even if Washington State "slows down" the game by limiting possessions, UNC is the likely winner because they score more often (meaning they are more balanced on offense) and, on avg., more points-per-possession.
Now, let's address the "UNC plays spotty defense" claim. Again, like straight batting average in baseball, total points per game is a low value statistic. A much better defensive statistic is defensive points-per-possession allowed. In that category, Washington State bests UNC by (drum-roll please): .014 points-per-possession. For you liberal arts majors (myself included), that's about 1 point per 72 possessions - almost meaningless. So, effectively, UNC's defense is just as effective as Washington State's, statistically. Not to mention that UNC creates more steals-per-game, more blocks-per-game and has the best rebounding margin over its opponents in the nation.
Just to test my hypothesis, I looked at Notre Dame and Winthrop. Both of those teams have lower offensive points-per-possession and Notre Dame has a higher defensive points-per-possession (Winthrop on avg. plays a little better defense than Washington State, but in a much weaker conference). So, no matter the number of possessions, Washington State is the likely winner, statistically, of both those games.
So, based on this astute analysis, what say you? I say even the oddsmakers are sand-bagging at an 8-point spread: UNC 85, Washington State 68, in a not-close game.
CREDIT: For those of you who are as offended as I am by the fact-free prognosticators, I got all of these stats from an awesome website called Basketball State. You should check it out, and tell them I sent you!