Today’s NBA looks alien compared to a decade ago. The conversation and the lens by which we view the NBA has changed with it.
You may have heard some new terminology in this space and go, huh? I don’t want to lose my readers, so with that in mind, I decided to write up a glossary that you can use as a trusty guide next time I bring up an acronym that sounds like a bill proposed on Capitol Hill.
Let’s call it Tom’s Toolbox. Here are some of my favorites:
Points scored by a team every 100 possessions, also commonly referred to as offensive efficiency.
Offensive rating adjusts points per game to give a better sense of a team’s ability to put the ball in the basket. To calculate, you simply take the points that a team scores and divide it by the number of possessions it used and then multiply that figure by 100. Ta-da!
Uh, why 100?
It is roughly the number of possessions in an NBA game. Offensive rating (ORtg for short) also helps to compare teams in the past and is really useful when comparing lineups that don’t play the same minutes.
Where to find it: NBA.com and Basketball-Reference.com. Note: possession data can differ from site to site. Some sites estimate the number of possessions in a game because it’s not historically tracked in the box score. You have to do some back of the napkin math to calculate possessions and so it can vary slightly. I use NBA.com for looking at today’s teams and those dating back to 1996-97, the first season available on the league’s site. If you want to go back further and size up teams with Kareem Abdul-Jabbar on it, I recommend using Basketball-Reference.com.
This is the counterpart to Offensive rating, which adjusts for pace. Defensive rating, also known as defensive efficiency, is simply the number of points allowed per 100 possessions.
You’ve probably seen this a lot in modern NBA writing when talking about lineups or how good the Milwaukee Bucks are. It’s simply offensive rating subtracted by defensive rating. Boom: net rating. Another way to think about net rating is that it is point margin once you control for pace.
Offensive rating, defensive rating and net rating can be super noisy in small sample sizes. As a rule of thumb, I like to look at lineups with at least 100 minutes played before I give it serious weight. Even then, it gets wonky.
There have been studies that show net rating is a better predictor for team success than using win-loss record. That’s because net rating strips out the flimsy randomness that can decide close games.
PER (Player Efficiency Rating)
The most popular all-in-one player metric in basketball. PER set out to measure productivity on a per-minute basis. League average PER was set at 15, which may be its best quality. You can get a pretty good handle of how good a player is by glancing at his PER. Twenty is great. Ten … not so much.
PER has its issues. For example, it likes chuckers more than it should. And it doesn’t do a good job of capturing defense because box scores don’t really measure defense very well. And it uses box score data to inform it. Blocks and steals only go so far.
I still look at PER as a rough thumbnail portrait of a player and probably more than I should. It’s a handy tool that can sometimes get you in trouble if you use it too much and it can’t do as much as you’d like.
This “rate” is the estimated percentage of team possessions that a player uses via field goal attempt, free throw attempt or turnover while on the floor. Translation: How much offense is this player responsible for through his scoring?
It doesn’t perfectly answer that question but it’s a helpful way to see how much a player is chucking or not. High usage players in the 30s (or 30 percent, to be specific) are typically your MVP candidates who put up huge point totals.
I like to look at usage rate to see if a player is more of a role player. Usage rate could also be called role rate and it’d get at the same thing. Turnovers are included because a player “used” up the possession with a miscue. However, assists are not included in usage rate.
This stat can be more informative than using plain old field goal attempts to show how much offense is run through a player.
Find it on Basketball-Reference.com.
Win Probability Added
The beginning stages of the holy grail. Which player raised (or lowered) his team’s chance of winning through his shot attempts and turnovers? This is definitely not a comprehensive metric. It doesn’t see a host of important things like passes, screens and rebounds that help teams win.
This stat is more descriptive than predictive. I like this one for MVP races because that is more about what you did as opposed to what you will do.
Offensive Rebound Rate
What percentage of available offensive rebounds does this player or team collect while he’s on the floor? That’s what Offensive Rebound Rate is after.
It’s hard to get an offensive rebound if your team makes all of their shots. Like, really hard. So hard it can’t happen. Therefore, a team can have a “good” offensive rebounding night even if it’s raw total of offensive rebounds is low. The best teams collect 30 percent of their misses. So if you only have three offensive rebounds in a game (bad), but your team only missed 10 shots (wow!), that’s doing pretty good.
On the flipside, it’s much easier for a poor-shooting team to rack up a bunch of offensive boards simply because there are so many of them to be collected. For coaches, it’s more helpful to look at this number rather than raw offensive rebounds to determine how their team crashed the boards. Oh, and good news for bench players: this stat is playing-time agnostic so you can show you’re a great offensive rebounder without having to log 30 minutes a night.
Offensive rebound rate is also one of the main pillars of advanced stats in basketball, which Dean Oliver called the four factors. If you’re good at collecting your available offensive rebounds, you’ll be pretty successful.
Defensive Rebound Rate
This is the other side of the coin. Defensive rebound rate is why the Golden State Warriors' starting five has been so underwhelming this season.
Win Shares (WS)
Another all-in-one metric that attempts to quantify a player’s contributions to winning. Find it on basketball-reference.com. It’s a counting stat that tries to answer the question: How many team wins is this player responsible for?
Like every all-in-one metric, this one has to be handled with care. If Player A has more win shares than another Player B, that doesn’t necessarily mean he’s better. He could have just played more. The PER equivalent is win shares per 48 minutes or more commonly seen as WS/48. League average WS/48 is .100 which is a little harder to remember than PER’s 15.
Win shares has its blind spots, too. It doesn’t use lineup data, which can help find players that do things that the box score doesn’t pick up. It’s shown to be a tad biased towards players with good team defenses because its defensive component relies more on the team’s output rather than individual box score stats like blocks, steals and defensive rebounds.
Find it at Basketball-Reference.com.
RPM (Real Plus-Minus)
The most ambitious of all the popular all-in-one metrics, this statistic attempts to quantify a player’s impact on team performance, scaled on a per 100 possessions basis. This leans heavily on on-court lineup data and controls for teammates and opponents. It’s a blackbox stat which means the formula is not publicly available and can’t be reproduced. That lends itself to some skepticism in academia.
RPM is a rate stat like PER and can be translated into a counting stat. RPM Wins is the counting stat equivalent of win shares. It’s a fun stat but it can be noisy in small sample sizes. If you want a full, way-smarter-than-me breakdown of RPM, read this.
True-Shooting Percentage (TS%)
One of my favorite stats. How efficient is a player with his shot attempts once you consider free throws and 3-pointers? It’s scaled roughly to your standard field goal percentage, so a .600 is great while .400 is … not great, Bob.
True-shooting percentage derives its name from the concept of true shots. “True” shots are field goal attempts plus fouled shots. Once you account for those free throws, you get a fuller picture of the player’s scoring night and how efficient he was getting to that point total.
Another cool thing: true-shooting percentage is points per true shot, divided by two. The divide by two is to keep it on the same scale as field goal percentage.
Effective field goal Percentage (eFG%)
Also seen as adjusted field goal percentage. The calculation is simply field goals plus 1.5 times 3-point field goals divided by field goal attempts. If you don’t make any threes, your effective field goal percentage will be exactly your field goal percentage.
Whereas true-shooting percentage includes free throws, this stat is just field goal percentage with proper weighting for three-pointers. This stat naturally rewards 3-point shooters who don’t need a lot of shots to score a lot of points.