sp0rtsfan
Does anyone have a subscription? Curious as to the findings.

http://insider.espn.go.com/nba/story/_/id/10529488/nba-tracking-referees-next-step-nba-analytics
Quote 0 0
cameroncrazies02
Over the past decade, NBA statistics have made a quantum leap. But as much as we've learned about players and even coaches, the public still knows very little about the other three actors on the court: the referees.

John Ball aims to change that.

A former consultant to the NBA video service Synergy Sports Technology with a background in technology, Ball set out to track the league's officials. Instead of relying on the calls available in box scores and play-by-play, which aren't specifically credited to any individual referee but to the entire crew of three, Ball and his team of loggers have used video to specifically determine which official is responsible for each call.

The results, which will be presented in a research paper at this weekend's MIT Sloan Sports Analytics Conference, reveal individual tendencies that are unique to each referee. And that information could be very valuable to teams.

Variation among referees
Despite the NBA's intensive referee training program, which now uses the D-League and the WNBA as minor leagues for the NBA's reffing standards and philosophy, officials are human. That means they all interpret the subjective aspects of the league's rulebook in different ways, or are better at spotting some calls than others. In practice, that means variation among referees in how often they call violations.

The most variance, unsurprisingly, comes in terms of handing out technical fouls -- probably the least objective aspect of refereeing, as well as one with a high degree of variance because technicals are called so infrequently. (In statistical terms, they have a high "coefficient of variation.")

According to Ball's data from the 2011-12 and 2012-13 seasons, including the playoffs, the difference between the most trigger-happy official (0.577 technicals per game) and the most gun-shy (0.067) represents more than 40 technicals over an 82-game season.

It's easy for fans to notice which referees are prone to technicals because all attention goes to the play. Other calls, such as defensive three-second violations, tend to be more subtle. Yet Ball's tracking still shows large differences among referees in calling these penalties. His database of these calls demonstrates skewed distributions, where a handful of officials are much more likely than their peers to make these calls.

Perhaps most interesting in this regard is the choice between a blocking foul and a charge. While on average, referees do tend to call about an equal number of the two -- making this a true 50-50 call -- that figure varies between officials. One referee called more than twice as many charges as blocks in the tracked sample, while another had the inverse ratio.

Applying referee data

While this data is interesting for fans and neutral observers, its true value is to NBA teams. In the past, teams have maintained their own databases of referee tendencies. Dallas Mavericks owner Mark Cuban described the value of tracking officials at the 2011 Sloan Conference in a panel on referee analytics.

"Anything that can influence the outcome of a game is important," Cuban said. "I wanted to be aware of each official's skill set to the best of our ability, so if somebody happens to call three seconds more often than another, someone else calls more falls than another, it can and sometimes does influence our game planning."

So far, Ball has partnered to provide his referee data to a few teams. As Cuban noted, understanding the tendencies of the three referees working each game can help teams tailor the plays they call, and potentially ultimately the decisions players make on the court. For example, a quick guard might be more willing to drive the lane with abandon if he knows that the officiating crew calls more blocks than charges. Or a center might be willing to stay in the paint a beat longer based on the referees' tendency not to call three-second violations.

A recent finding that could be useful to teams is that certain referees tend to call violations more on one side of the court than the other, presumably because they must focus on one half of the play. That gives teams the opportunity to derive an advantage simply from running a play on that specific side.

In a competitive league, every advantage matters. If a team could turn that understanding into an extra point every three games, on average it would translate to an additional win over the course of the season.

The future of referee analytics
There's one group that has more data on referees than anyone else -- the NBA league office, which tracks every call for monitoring and oversight purposes. For now, they're not sharing. That's also the case with information on referees from the SportVU camera-tracking system, which is now installed in all 30 NBA arenas and monitors the positions of referees as well as players and the ball.

As a result, it's unclear what additional insights might exist outside the public domain. (Cuban did assert, in the Sloan panel, that SportVU data was only useful for determining which referee made each call -- a process Ball's group already does by hand.) As the NBA statistical revolution progresses, more data figures to offer better understanding of the role, importance and effectiveness of referees.
Twitter: @TheSpangover

Check out diener and my new website @ http://www.thirddegreewithcc.weebly.com

Newest additions
-top 15 NBA free agents by position. That's 75 reviews of free agents.
-Diener breaks down the Brewers as a franchise
-2013 NBA Draft Combine Day 1 results and review
-State of the Roster-breaking down 3 best and worst NBA rosters
Quote 0 0
cameroncrazies02
More so just a summary that referee analytics is coming. Won't be released until after the Sloan Conference. I would just like to throw out there that the Sloan conference, in all of its glory, can drive analytics way too far. Analytics are good to a certain extent, especially when it comes to efficiency, but some of these research projects are just becoming shock theater. Kirk Goldsberry came out with a project last year that showed that David Lee was the worst frontcourt defensive player in the league while Larry Sanders came out on top. It blew up into Lee getting slandered for months, many jumping on the "Lee is overrated" bandwagon even though he's not even properly rated and Larry Sanders' value skyrocketed in a contract year which bumped his value by at least $1 or $2 million.
Twitter: @TheSpangover

Check out diener and my new website @ http://www.thirddegreewithcc.weebly.com

Newest additions
-top 15 NBA free agents by position. That's 75 reviews of free agents.
-Diener breaks down the Brewers as a franchise
-2013 NBA Draft Combine Day 1 results and review
-State of the Roster-breaking down 3 best and worst NBA rosters
Quote 0 0
wissportsnet

Boys Basketball Alumni Round-up: February 21st, from @ColtonWilson23 #wisbb -- https://t.co/0K6CZzZWpf https://t.co/715tfpBGVS

wissportsnet

WSN15: Boys Basketball Top Teams #12 -- Two NCAA Division I players at one WIAA Division 4 school = state champions… https://t.co/1MThqzce9L

wissportsnet

Predicting winners of every state wrestling title plus a look By The Numbers, from @Nate_Woelfel -… https://t.co/HgU1bLP7d6

wissportsnet

Join the free Boys Basketball Playoff Pick 'Em Contest; Staff picks coming Tuesday #wisbb -- https://t.co/Y6yTA3OgP8 https://t.co/hA6Uyw9uje