Our five-part primer on how to beat the March Madness basketball odds wraps up with a closer look at the cause, and the solution, of all our problems: the numbers.
Jason's 2015-16 record as of Mar. 10: 15-11 ATS, 1-2 Totals
Have you read Moneyball yet? Good, then you don't need me to convince you about the value of using statistics, and the danger that comes with using them the wrong way. If you haven't read Michael Lewis's ground-breaking 2003 book, I cannot recommend it highly enough. Even if it's about baseball.
Meanwhile, let's talk March Madness. The NCAA men's basketball Tournament is just days away, which means there will be millions of recreational bettors entering the marketplace. They're here to have some fun, and maybe win a little cash. They're not going to use statistics to help them with their choices; Duke fans will bet on Duke, and Duke haters will bet against Duke. Our “job” is to take their money by making smarter, well-informed decisions when it comes to the teams, the players, and the NCAA Tournament Odds. So where do we get our information?
Daniel-San, This Not Tournament
That's where most analysts fall short. Growing up as basketball fans, we get used to the basic, traditional statistics that everyone's exposed to, especially the Big Five: points, rebounds, assists, blocks and steals per game. These numbers do tell a story, but they don't tell the whole story. No set of numbers can. In the words of 20th-Century scholar Alfred Korzybski, the map is not the territory.
Sometimes, the numbers outright lie to us. This year's top scorer (per game) in Division I hoops is junior swingman James Daniel from the Howard Bison, representing the MEAC. Daniel scored 27.1 points per game, and won the MEAC Player of the Year award for his efforts. Well done, but Daniel didn't do a very good job at the other end of the court. Here's another junior guard/forward who did: Malcolm Bernard of the Florida A&M Rattlers. If only we had some stats for defense to prove it.
Oh wait, we do: Sports-Reference has a college basketball website filled with all kinds of stats, both traditional and “advanced,” and freely available for us to use. According to Box Plus/Minus (BPM), which estimates how many points above average a player contributes to his team, Daniel's presence added 4.2 points per 100 possessions to the Bison cause – on offense. On defense, he gave the opposition 3.5 points, for a total BPM of +0.7.
As for Bernard, he led the Rattlers with 14.4 points per game, but he also shot the ball a lot and wasn't terribly accurate (42.1 percent on field goals); his offensive BPM worked out to –1.1. But on defense, Bernard was a giant, giving FAMU 3.2 points of value for a total BPM of +2.0. That was the best BPM in the entire conference.
You can use these efficiency-based numbers at the team level, and even the conference level as well. Ken Pomeroy's rankings have become industry-standard over the past decade, and the Simple Rating System at Sports-Reference is a handy way to estimate a team's value in terms of points above or below average. Other ratings systems (Jeff Sagarin's, for example) are also available. If you're not using stats like these to help inform your NCAA Tournament Picks, well, go back and read Moneyball before you start placing any more bets, and may the sphere be with you.