Back in the early 2000s, Dean Oliver was sitting at his computer. Oliver, who is the creator of the “Four Factors of Basketball Success” and considered one of the godfathers of NBA analytics, was searching for others who shared his same interest in applying statistical analysis to basketball.
He had seen Bill James revolutionizing baseball with Sabermetrics, but there wasn’t nearly as much being done on the NBA front. In an attempt to fill the void, he created a Yahoo email list to spark a conversation in which he could bounce ideas off his friends.
“I was trying to find other people who were thinking a little bit like me,” Oliver said. “You want to find people who are smart enough to question you, but also naive enough to want to explore all of these crazy ideas.”
Oliver slowly began to build up a community. Toward the end of 2004, ESPN writer Kevin Pelton decided to migrate Oliver’s Yahoo group into a more user-friendly message board format.
And thus, the APBRmetrics message board was born, bringing together some of the most brilliant minds from all over the country and shaping the future of the basketball analytics space.
APBRmetrics (Association for Professional Basketball Research) offered a different medium for discussion from what we see today. It was accessible to everyone, mostly free of trolls and united in a common goal of advancing the state of basketball analytics.
Dan Rosenbaum, an economics professor who left academia for jobs with the Cavaliers, Hawks and Pistons, learned from all of the different walks of life that came through APBRmetrics’ doors.
Rosenbaum found the boards much different than what he was used to in a university setting. Posters were judged “more on the merit of their ideas” rather than their credentials.
“And folks found places where they could contribute,” Rosenbaum said. “Not everybody could contribute to every aspect of every discussion. But I think that was part of the strength of the board.”
Daniel Myers, a bridge engineer by day who had never taken a statistics class in his life, stumbled across the board through a search engine result. He quickly found a welcoming community willing to answer his questions about basketball.
“Everyone was very cordial and non-confrontational,” Myers said. “Even if you asked a really dumb question, these people who had doctorates in statistics would answer your question and not say, ‘That’s a stupid question.'”
Myers was like a lot of members back in those days. He learned an “absolutely massive” amount of information from reading and asking questions to the leaders of the field, then found ways to chip in.
Eventually, he used the information he picked up to create one of the first-generation all-in-one player metrics, Box Plus-Minus.
“Maybe there was a core of 15 or 20 [posters], but these were 15 or 20 of the very best in the world — the very leaders in the field,” Myers said. “[On] Twitter, you’re shouting into this orchestra of other voices also shouting to be heard.
“Because the forum was kind of a hidden gem, you would get really knowledgeable feedback. The amount of feedback wouldn’t be as high, but the quality would be way better.”
Kevin Ferrigan, creator of Stable Player Impact, notes that the popular all-in-one metrics — from ESPN’s RPM to FiveThirtyEight’s RAPTOR, EPM and LEBRON — incorporate knowledge and techniques that were established in the APBRmetrics boards.
“We’re basically iterating, and at this point, the improvements are marginal as compared to what a quantum leap forward those stats gave us initially,” Ferrigan said.
John Hollinger is another pioneer in the analytics space. Hollinger created the Player Efficiency Rating (PER) metric, served as the vice president of basketball operations for the Grizzlies and has written for ESPN and The Athletic. The APBRmetrics board helped Hollinger “learn and figure more stuff out.”
The collegiality, intellectual curiosity and openness of those boards were a stark contrast from what Hollinger later experienced in the front office, where information was treated as an advantage that should be heavily guarded.
“We weren’t working for teams and worried about their proprietary information and all this bulls—, so it was this really kind of pure collegial environment where we could share stuff,” Hollinger said. “People could critique or offer suggestions. It was a safe space almost.
“We still had our arguments and discussions, but it was a cool time, this very innocent new era for analytics and basketball.”
The APBRmetrics board wasn’t just for hobbyists and intellectually curious folks. It opened up avenues for professional advancement, serving as a sort of pipeline into NBA analytics departments during the mid-2000s.
“If you wanted to get in the league back then, you made a good website or a good blog and posted good analysis, posted it or linked it over there in APBR and had a good back-and-forth with the really smart people there,” Myers said. “It didn’t usually take very long before you got hired.”
That all started after his first big break as a freshman in college, posting on the APBRmetrics boards and catching the eye of Oliver, who brought Falk to Denver for a summer internship while he was serving as the director of quantitative analysis for the Nuggets.
“I don’t think I was unique in that way at all,” Oliver said. “I was looking at APBRmetrics back then for identifying people.”
It wasn’t just writers and executives that honed their ideas on the APBRmetrics board.
YouTube sensation Ben Taylor of the Thinking Basketball channel was lurking, occasionally posting and referencing ideas that he picked up from the site. Evan Zamir, creator of NBA WOWY, was posting his thoughts regularly.
And Justin Kubatko got the inspiration to start one of the most influential basketball websites on the internet from his long hours on the forums.
“I remember reading through the messages on the APBRmetrics boards and thinking, ‘Wow, this is kind of cool and, at the time, revolutionary,'” Kubatko said. “I was like, ‘It would be nice if there was a place online where we could see all these figures — not for just the current season, but historically.’
“And that was one of the things that helped motivate me to create Basketball-Reference.”
The format of the APBRmetrics board lent itself to deeper discussions that have gone by the wayside on today’s social media platforms.
“You get to know each other better in a setting like that, as opposed to Twitter where it could be someone without an avatar that you don’t know well wandering into a discussion,” Pelton said.
“And a lot of these are nuanced discussions. It’s difficult to discuss in what at that point was 140 characters, as opposed to being able to post a five-paragraph discussion and go back and forth that way.”
Through that nuanced discussion, the APBRmetrics board established fundamental concepts that remain in place today. Topics discussed included the idea of a true shooting percentage, defining what a possession was, scaling statistics to per 100 possessions and many others.
“This feeling that we actually knew s— that people in the league didn’t know was kind of wild,” Hollinger said. “And that made it a really fun and interesting time. You were part of a select group that was in on the joke.
“That era introduced the entire mindset that, ‘Oh my god, look at all this money you’re leaving on the table!’ It really encouraged teams to go deeper into this in the first place, to learn more and to find out more things. A lot of the stuff seems kind of basic now, but at the time, it really changed the league.
“I think the league was not as open-minded to that kind of thinking, and now, it’s a requirement almost.”
BORG = Big Ol’ Rating. Here’s an explanation, plus an introduction to my free agent values: https://t.co/K26nfe2ArX
— John Hollinger (@johnhollinger) November 2, 2020
The APBRmetrics board still exists today, and Myers volunteers as the site’s administrator. An annual thread soliciting statistical models for a win projection contest still gets some traffic.
But for the most part, traffic has died down. It was a victim of its own success, as some of its brightest members left to go work for teams or pursue other professional endeavors. And yet, its legacy lives on.
“You see true shooting percentage in ‘NBA 2K,'” Pelton said. “It’s still not totally mainstream, but at the college level, you hear advanced statistics as a regular part of broadcasts. All those things that happened is what we were hoping for.”
The legacy also lives on in the work that its members continue to do.
Behind many of the trades that teams make, the websites that you visit, the YouTube videos that you watch, the podcasts that you listen to and the articles that you read, there are strings leading back to the members of the forum.
“Fundamentally, it’s not just a board,” Oliver said. “It is the people who were on it.”