CBT Exam Week Essays: How useful are advanced statistics?
For college students and college basketball fans, Exam Week is the worst week on the schedule. For students, this week is the culmination of three months worth of procrastination, cliff notes and Wikipedia. For college basketball fans, it’s the lightest week of hoops action we will see all season.
With so very little going on this week in terms of action, the staff at College Basketball Talk is going back to school. Over the next five days, the CBT Staff will be responsible for answering an essay question in one of five different subjects.
Today we have the dreaded math exam.
The use of tempo-free and advanced statistics has become more commonplace in recent years thanks to the work of guys like Ken Pomeroy and Dan Hanner. What is your opinion on the use of advanced statistics as analysis aids. What value do these metrics add to post-game evaluation in your opinion. Please use at least one example of accurate or misleading statistics in your response.
By Rob Dauster
The advent of advanced statistics in college basketball is incredibly useful.
Between Kenpom’s efficiency profiles, the in-depth player and team breakdowns available on Synergy and all of the other sites doing yeoman’s work to try to enlighten us about and dispel myths about our favorite teams, there’s no shortage of data available for a college hoops junky to get their fix.
The key, however, is understanding how to use that information in concert with what actually happens on a basketball court.
Advanced statistics have really taken off over the last decade in all sports, not just basketball. The most famous case is that of Billy Beane, the Oakland Athletics GM that was the subject of the ‘Moneyball’. He used sabermetric principles to find value in players that others had cast aside, and as a result was able to build the A’s into one of the best teams in baseball despite having one of the lowest payrolls in the league.
But focusing on stats works much better in baseball than it does in hoops because baseball is a game made up of a series of events involving individuals. A pitcher throws a pitch, the hitter swings and puts the ball in play. Lather, rinse, repeat.
Basketball is a much more free-flowing, one that requires five players to work together both offensively and defensively. So much of what happens on the court -- Who can execute offense effectively? How well does a team communicate defensively? Is a player a good rebounder because he’s simply bigger and more athletic, or does he understand how to box out? Who is a team’s leader? -- doesn’t show up on a stat sheet or a Kenpom page.
The perfect example is the argument involving Marcus Smart.
Smart, a freshman, is Oklahoma State’s starting point guard, a top ten recruit and, thus far, the difference maker for the Cowboys. He’s averaging 13.0 points, 7.4 boards and 5.0 assists for the Pokes, but what he’s done setting a tone defensively and in the locker room is why Travis Ford’s club currently owns wins over Tennessee and NC State. He’s played like an all-american, and was called the best player in the country through the first month of the season by Mike DeCourcy.
DeCourcy doesn’t believe in advanced analytics, however, which is why he’s probably unaware that Smart has an effective field goal percentage of 39.7% -- he’s shooting 20.6% from three and 43.9% from two -- and an offensive rating of 99.3 (he averages 0.993 points per possession used, which isn’t very good) despite using 28.0% of Oklahoma State’s possessions when he’s on the floor, a very high number. Those ugly numbers are why John Gasaway, one of college basketball’s leading ‘stat nerds’, has Smart ranked as the 14th-best freshman thus far this season.
And, as I wrote yesterday, neither of these gentlemen are right, because you simply cannot ignore a) the effect that Smart’s presence on the floor has had for Oklahoma State this year, or b) just how inefficient Smart has been with the ball in his hands. He may be a great leader, a great defender and a better-than-we-thought creator, but that doesn’t change the fact that he hasn’t shot the ball well at all and he’s turned the ball over far too often.
Advanced statistics are most effective when they are used to help quantify trends that your eyes tell you exist or as a way to determine what to watch for when a certain team takes the floor.
It’s inarguable that understanding efficiency breakdowns and possession-based logs, like Synergy’s database, make you smarter about basketball and what a specific team or player is doing on the court.
But it’s also impossible to use those number correctly without being able to watch a game and understand what is happening and why it happens that way.
The two schools of thought are not mutually exclusive. But if they aren’t used correctly, they are misleading and unnecessary.
Professor’s Notes: I really wanted to give you a bad grade. The rise of #DausterMath gave me little reason to believe that you would be able to handle this question. But you nailed it. Kudos for the use of words like “advent”, “yeoman”, and the ‘Moneyball’ reference was sublime. You were on pace for an A+, but points are being docked for starting a sentence with “And”.