Print

A Primer on Advanced Statistics

Written by Craig Meyer on .

For those of you that have read the blog for a while or have seen some of the pieces I’ve done, you know I like statistics.

It seems like there’s always talk about the best way to analyze games – whether it’s through astute observation or through obsessive statistical research. It’s a debate that began in baseball and has transitioned over to basketball at both the pro and college level.

Personally, I think the answer lies in between the opposing camps. Stats don’t mean way too much if you don’t have fundamental basketball knowledge, but you also can’t put forth opinions through game analysis alone. It’s like just writing a thesis statement and calling it a paper.

With the wealth of resources that exist – like Ken Pomeroy’s ratings and the work being done at Basketball State – college basketball followers now have access to numbers that present a more comprehensive picture of the game and heightens our understanding. It aids writers like me and makes fans smarter. A true win-win.

However, the best statistics aren’t the ones commonly floated around. Per game numbers (be it points, assists or rebounds) have their place, but they’re largely inadequate because they are a product of pace. Instead, more contemporary basketball analysis revolves around tempo-free statistics that focus on a possession-by-possession basis, not a game-by-game one.

Teams that often score a lot of points per game are seen as offensive juggernauts when that sometimes isn’t the case. An example that’s been used a lot is Billy Tubbs’ 2003-04 Lamar team. That season, the Cardinals averaged 79.1 points per game, which was the 15th-best mark among Division I teams and would lead some to believe they were an offensive juggernaut. But they also used about 80 possessions per 40 minutes, the fastest pace in the nation. Because of that, they were an inefficient offensive team that averaged just about 0.963 points per possession (242nd in DI), something that might explain why the team just wasn’t very good (finishing 11-18).

The Lamar example helps illuminate the importance of judging a team, both offensively and defensively, on a points-per-possession basis. More often than not, the most efficient teams in college basketball are the most successful.

If there’s any threshold to judge a good team from a bad team on offense and defense, it’s an even one point per possession. Generally speaking, if you’re holding teams under one point per possession on defense and scoring more than one point per possession on offense, you’re in pretty decent shape.

Below is a tempo-free aerial of the NEC last year to expand on this point:

Updated Tempo Free Aerial

(Image snipped from Basketball State)

I’ll follow this up with an explanation of some of the individual statistics I use. I’ll add this to the blog roll at some point for easy reference or you can feel free to individually bookmark it.

Behind many of these stats are mathematical formulas that I won’t attempt to explain or pretend to fully understand beyond the concept of them. But I’ll explain the end product.

Percentage of possessions used: Pretty self-explanatory, but it’s basically how much the ball is in a player’s hands when his team is on offense.

Percentage of shots taken: The stat says it all: what portion of a team’s overall shots are made by a particular player?

Effective field goal percentage: A field goal percentage that puts a greater emphasis on 3-point shooting. Roughly, they’re given about 50 percent more credit. Since 3s are (obviously) worth more points and are harder to make, you generally see strong 3-point shooters at the top of this statistical category. Not too surprisingly, Karvel Anderson led Robert Morris in this category last season.

Assist rate: I use this one a lot when talking about guards (especially those that run the point), but it’s the number of assists divided by the number of field goals made by teammates when a player is on the court. In short, it measures how active a player is in getting his teammates good scoring opportunities.

Offensive rebounding percentage: The percentage of available offensive rebounds a player gets when he’s on the court. Anything over 10 percent is doing alright.

Defensive rebounding percentage: Same thing as offensive rebounding percentage, but on the defensive end. Anything over 20 percent is good.

Block percentage: The percentage of two-point shots that are blocked by a player while he is on the court. For an RMU reference, Vaughn Morgan was a savant in this category before he left the team this season.

Free throw rate: Basically, the number of free throws a player takes per 100 field goal attempts. This is an important thing to consider because a player that gets to the line a lot is often a very capable and efficient scorer, since free throws are pretty easy to make. It also can indicate how aggressive a player is on offense (at least for guards), as they are generally fouled in closer proximity to the basket. Anything over 50 is good, 70 or more is excellent.

Steal percentage: Percentage of possessions where a player records a steal while they are on the court. Any number greater than 5 percent is pretty, pretty good.

Turnover rate: The percentage of personal possessions used on turnovers. Ken Pomeroy warns that these are highly dependent upon context. Building on that, point guards generally have higher turnover rates than their teammates simply because the ball is in their hands more.

Join the conversation:

To report inappropriate comments, abuse and/or repeat offenders, please send an email to socialmedia@post-gazette.com and include a link to the article and a copy of the comment. Your report will be reviewed in a timely manner. Thank you.