Beyond the box score: An introduction to the world of advanced basketball analytics

Thursday March 11, 2010 9:15 AM

Beyond The Box Score

An introduction into the world of advanced basketball analytics

Jason Friedman Staff Writer

HOUSTON - Not long ago, the Holy Grail of major league baseball was thought to be found within its triple crown statistics: batting average, home runs and RBI. Those numbers served as the foundation of player evaluation, supposedly shedding ample light into everything fans, media and management needed to know about the men who made their living on the baseball diamond.

Then people like Bill James, Rob Neyer, Billy Beane and countless others helped usher in the statistical revolution, forever changing the way the game is viewed. Now we know the critical importance of on-base percentage. Numbers like OPS and WHIP finally occupy a (well-deserved) permanent place in the baseball lexicon. And even difficult to properly assess areas like fielding have seen significant strides in the way we determine the difference between those who are gifted with their gloves and those more prone to gaffes.

None of this is to say that the triple crown numbers have lost all meaning. It’s simply that the palette used to paint the baseball canvas these days now contains an assortment of colors which span the entire spectrum. As a result, today’s picture is far more vibrant, not only allowing us to see certain subtleties which went undiscovered before, but also enabling a greater appreciation of that which is absolutely essential to the beauty of the portrait as a whole.

The same evolution, of course, is well underway in basketball. An army of analysts, writers and executives – including Rockets General Manager Daryl Morey – has helped bring a dizzying array of advanced metrics to the table, shedding light on some of the game’s finer points which had previously been left in the dark by the confines of the traditional box score.

As with baseball’s sacred stats of yore, these new numbers are not meant to completely take the place of the facts and figures which served as basketball’s foundation for so many years, but simply to put them in their proper perspective. Points per game and other such statistics still matter, of course – but they become far more meaningful and impactful when placed within the context of terms like pace and usage.

It’s all part of the process of discovering that which impacts winning and losing the most, and it’s that ongoing search for the truth which lies at the heart of the analysis utilized by not just the Rockets’ brain trust, but a host of brilliant minds behind websites such as Basketball Prospectus, HoopData,, Basketball Reference and many others as well.

If you’re already familiar with the treasure trove of information available at those sites then chances are what you’re about to read will be nothing more than a refresher (though, by all means be sure to check out the highlights from the recently completed fourth annual Sloan Sports Analytics Conference – of which Morey is a co-Chair – which included discussion of such thought-provoking topics like the value of a blocked shot and the price of anarchy in basketball).

If, however, you’re unfamiliar with or not entirely clear what terms like efficiency and true shooting percentage really mean then hopefully this can serve as an introduction into a few of the fundamental concepts which drive basketball analytics, allowing you invaluable insight into not only the minds of Houston’s decision makers but, ultimately, the game itself. After all, even though the Holy Grail isn’t likely to ever be found in statistics, a far greater understanding of the difference between salvation and damnation (or in this case, winning and losing) certainly can be.


1.) PACE

Pace is exactly what you think it is: a way to measure the speed with which a particular team plays. Not only is it helpful when discerning each club’s preferred style of play but it also serves as the first step in figuring out why all scoring averages are not created equal (we’ll get to that one in the efficiency section).

Pace is simply defined as the number of possessions a team uses per game. Not so simple, however, is figuring out the number of possessions each club gets in a single contest since it’s not something which can be precisely calculated just by using the contents of a traditional box score (as explained in detail here). That said, there are several different formulas readily available online which can quickly calculate the number of possessions used by each team per game, subsequently shedding light on the pace of play.

This is a stat which is particularly interesting when examining this year’s Rockets squad, since the team repeatedly stated its intention to push the pace and play at a faster rate before the season began. Sure enough, Houston has generated about 3.5 more possessions (96.2 compared 92.7) per game than last year’s club.

To put that in its proper perspective, consider that A.) The average NBA team uses 95.3 possessions this season and B.) Whereas this year’s Rockets team ranks No. 7 overall in pace of play, last year’s club would have placed 26th. So while a difference of 3.5 possessions per game might not sound like a lot, it’s been enough to transform Houston from one of the league’s slower paced teams into one its fastest.

Of course, it’s not how many possessions you have but what you do with them. Which brings us to…


Team X scores more than 107 points per game, ranking it No. 3 overall in the NBA.

Team Y, meanwhile, tallies a shade under 98 points per game, placing it in a tie for 21st.

Less than a decade ago, the general consensus would have been that Team X possessed an offense which was vastly superior to that of Team Y’s. But today we know better. Or at least we should.

Team X happens to be the Golden State Warriors, a team that epitomizes the term “run and gun.” The Warriors shoot early and often, which is why they lead the league in pace by a significant margin, averaging 102.7 possessions per game. Team Y, on the other hand, is the Portland Trailblazers, who just so happen to be the slowest team in the league from a pace perspective. The Blazers operate almost exclusively out of the half-court, often milking the shot clock which therefore means they typically utilize fewer possessions than does the average team.

So here we have two clubs which operate on polar opposite ends of the spectrum. It’s not a matter of right and wrong – they’re just different. But that variance doesn’t prevent us from comparing the overall offensive effectiveness of the Blazers and Warriors. By simply figuring out the number of points each club would score if given 100 possessions, we can precisely see just how efficient each team is. And what we find when doing that is the Blazers soar into 7th place in the NBA, producing 107.6 points per 100 possessions, while Golden State slides all the way down to 15th, averaging exactly 104 points per 100 possessions. So even though the Warriors might possess the more prolific offense, Portland clearly boasts the more effective one.

That, then, is why all scoring averages are not created equal and why pace of play must always be accounted for when viewing both a team's and individual’s per game stats (defensive efficiency, by the way, is calculated exactly the same except it’s the number of points allowed per 100 possessions).


If Rockets fans weren’t familiar with this term before the trade deadline, they probably are now. That’s because the newly-acquired Kevin Martin is Captain True Shooting Percentage, having posted a TS% over 60 percent in each of the past four seasons.

So what is true shooting percentage and what makes it so meaningful? It really comes down to the fact that the traditional barometer of a player’s shooting ability (field goal percentage) weighs 2-pointers and 3-pointers the same (which clearly they are not, given the fact made 3s are worth 50% more than shots made inside the arc) and ignores free throws entirely. TS% takes all three into account (the formula, for those interested, is PTS / (2 * (FGA + 0.44 * FTA)) and therefore does a much better job of assessing who makes the most of the shots at their disposal.

It should come as no surprise then that Martin shines in this area. Though his .448 career field goal percentage might appear underwhelming to casual observers, his accuracy from deep and incredible ability to get to the line – and sparkling conversion rate once he arrives there – are what make him such a special player.

Again, the key word here is efficiency with the goal being to find out who maximizes the value of their team’s possessions. And when it comes to analyzing high-usage, perimeter players over the past five years, few come close to occupying the same rarified air of Martin in terms of efficient scoring.

Click here to read Part II of our introduction to advanced basketball analytics.

Got a question for Send it to Jason Friedman. And for up to the second news and injury updates follow the Rockets and Jason on Twitter.