My name is Dr. Andre’ Snellings, and I absolutely LOVE NBA analytics. And…in case you didn’t know…I’m as black as the ace of spades, like my grandma used to say.
As you probably know by now, that “confession” was in response to Michael Wilbon’s article Mission Impossible: African-Americans & Analytics that has the premise that black folks don’t talk about sports in terms of advanced analytics.
My life is evidence to the exact contrary. I’ve lived my entire life in “Blackworld”, as Wilbon puts it. I love barbershop sports arguments. My friends and I LIVE to argue sports. My best friend Jamal and I once got kicked out of both of our houses because we were yelling so loud at each other on the phone in an hours-long argument about whether Penny Hardaway or J.R. Rider were better players.
And in the internet age, I messed around and discovered basketball message boards, and things were never the same. These days I mostly post on RealGM, in the Player Comparison board, which could be re-named as the “Barbershop Argument” board because that’s all we do there: pick players and compare them, trying to figure out who’s better or worse and why, and trying to convince others that our way is right. If you ever go to that board, you can check out my posts…I post under the name drza.
If you ever read ‘drza’s’ posts, you know what you’ll find? Numbers! Oodles and skads of numbers! All the stuff that Wilbon said you never hear black folks say, I say it. I can talk boxscore stats like win shares or WARP. I can talk about why points per 100 possessions is necessary, often more-so than points per 48 minutes. And I absolutely LOVE talking about the impact stats, which the article says don’t exist.
And everyone that knows me, knows it. You know how I heard about Wilbon’s article? My friend Sellers…well, given the tenor of the response that Wilbon’s article has gotten, I should give his race and credentials: My black friend and former teammate from the University of Michigan, Dr. Sellers, texted me to make sure I saw it. This is because my friends, who also live with me in BlackWorld, know good and well that this is what we do. This is how we talk.
So, how did Wilbon get it wrong? A quick glance at the Twitterverse suggests that there are a lot of people who appreciate the article, but there are many others that are mad, and calling the article irresponsible. I can see their points of view, but I also feel like I can see what he was trying to do, so that’s not where I’m going to focus, here. But there are obviously a lot of people like me out there that Wilbon didn’t poll for this article. So…maybe it’s just a polling error. Wilbon says he observed all the black people he knew, so maybe he just needs to know more black folks? If so…
Wilbon, holla at me! I’ll talk sports with you until your EARS bleed! I’ve got a whole lot to say to you about your coverage of, say, the 2008 Celtics championship run. If you call me, I will change your entire view of what went down. I’ll do the impossible…I’ll use a LOT of numbers, AND I’ll even make it convincing to a self-professed analytics hater like you.
But all jokes set aside, I think the fundamental problem with Wilbon’s message isn’t one of intent (though he should maybe be more careful in “speaking for the blacks”) or polling error (though he really could call a brotha. I think he’d enjoy talking to me). But no, the biggest problem with Wilbon’s article is that he, himself, just doesn’t understand analytics. It’s blatantly clear that he doesn’t, and he doesn’t even really hide it. But that’s a problem, for a person that’s writing an article about how analytics are or aren’t used. Understanding, or lack there-of, can tilt the very tenor of the piece.
RotoWire has the best daily fantasy football tools on the web.
Try Our NFL Lineup Optimizer Now
Take, for example, this excerpt on Wilbon’s and J.A. Adande’s take on “numbers” in sports conversations:
“It was Adande who a few months ago pointed out to me that two guys who have covered sports for a total of 60 years never, ever resort to any analytic, simple or advanced, to discuss the daily ins and outs of sports. It’s not part of any discussion of any game for any reason, ever. And that’s overwhelmingly true of all the black people we know (and older white people, too) regardless of occupation or station in life. No conversation is ever framed or dominated by numbers.”
Whoa, slow your roll. Are you kidding me? I can’t speak for Adande or Wilbon, but literally EVERY sports discussion I’ve EVER heard, or been a part of, relies heavily on some sort of numbers. Check the bolded above, no analytics, “simple or advanced”. Simple, OR advanced? Do you know what analytics are? Analytics are, essentially, numbers. Do you know what a simple analytic for basketball is? Wins. Losses. Record. Points scored. Number of championships. Number of MVPs. All “simple analytics”. I flatly DEFY you, reader, to tell me that your sports conversations don’t include even numbers like these.
Wilbon brought up the barbershop argument. For those of you don’t know what he’s talking about…have you ever seen the movie, Coming to America? When Eddie Murphy and Arsenio Hall are arguing: Joe Louis is the greatest fighter ever lived! He better’n Cassius Clay…he better’n Sugar Ray…he better than…what’s the new boy, Mike Tyson, he look like a bulldog, he better than him, too!
That is the CLASSIC barbershop argument, because arguments just like them happened in every black barbershop in America. Eddie Murphy and Arsenio could have EASILY been G-Man and Jabbo from Blue’s Barbershop in Dayton, Ohio. That’s what I grew up on. But when the arguments were about basketball, instead of boxing, especially in the 80s and 90s, this is the kind of thing you’d hear:
“Kareem is the best that ever lived! He’s the scoring King! No one ever scored more points.”
“Naw, Jordan’s the best. Didn’t you see him drop 63 on the Celtics? He can’t be stopped!”
“Yeah, but Magic is a walking triple-double. You can keep Jordan’s 40 points…give me Magic’s 20 poins and 15 assists and another title.”
And that’s when the old dudes, like G-Man, would jump in: “Triple doubles? Titles? Please, you young boys don’t know nothing about the game. Oscar Robertson averaged a triple-double for a whole SEASON! And Bill Russell got 11 titles, more than anyone ever! And Wilt once scored 100 points! Learn your history before you talk around grown folks.”
Look back at those arguments for a minute… EVERYONE’s argument was based on numbers. The very “simple analytics” that Wilbon and Adande decried.
Now, obviously this is a more simplistic example than Wilbon had in mind. Wilbon specifically mentioned advanced analytics like Win Shares or Point per 100 possession. Where do they fit in? Well, at some point it became clear that the “simple” analytics like the traditional boxscores or championships won just aren’t enough to fully capture the game. There were too many unanswered questions:
What actually IS better, to average 35 points or 20 points with 15 assists?
How much credit SHOULD an individual get for a championship? After all, we can’t just count rings or the much-used Robert Horry and his 7 rings counter-example can be used as a retort. But on that note…just how important WAS Horry to all those titles?
THIS is where the NBA analytics revolution comes from. Wilbon talks about Analytics like it’s something alien to the game…or something that has to be independent of all other basketball knowledge. But it’s the exact opposite! When used right, advanced NBA analytics are just giving us new, better tools to understand the game in a more complete way. And THAT’S what makes them so great in an argument.
What’s better, Jordan’s 35 ppg or Magic’s 22 and 13? Well, we can do regressions on successful players and teams through the years that look at combinations of the entire box score to help estimate what makes good basketball. Lo and behold, scoring efficiency, turnover percentage, usage rate…these are all factors that tend to be important for good basketball. So producing all-in-one boxscore stats like win shares or wins above replacement player (WARP) become good, quick ways to try to group successful player traits and thus enable quick comparisons of players, even across eras. Are these stats bullet-proof? Of course not. Would I advocate using them the exclusion of all else, like watching the game? Of course not. But if you saw my video on Kobe Bryant’s place in history, these types of stats can be important data points that pretty much all of the usual suspects for NBA top-10 lists do excel in. So noting how Kobe does in them, compared to his peers, gives us something to work with beyond just, “But Kobe scored 81 beans in a game!”, which my best friend Jamal once used in one of our hoops conversations.
But NBA analytics extends far beyond the box scores. In Wilbon’s article, Shaun Livingston said, “Our conversations seem to go the other way, away from data and more toward intangible things. Like impact. There are too many areas where the numbers don’t assess the impact.” But this isn’t correct, as there actually ARE analytics approaches that specifically measure for impact and the intangibles that Livingston mentions! The +/- approach, which has evolved into a variety of regressed techniques, are to me the evolution of the “Championships/Rings” arguments from my childhood. Usually, rings are brought up to suggest that a player was great at contributing a lot to his team’s success. But the +/- numbers actually help to measure the correlation between a player’s presence on the court and his team’s scoring margin. It’s noisy and requires good interpretation, but it literally puts numbers to the question of how much credit Robert Horry deserved for his titles versus how much Kobe or Shaq or Duncan might deserve for theirs. And even if a player didn’t win a title in a given year, they may have been contributing more to their teams than anyone on the actual title team!
You think that’s not relevant info? You think black folks, or folks of any color, wouldn’t use that in their arguments if they understood it in that way? Or, to keep it grounded to real examples, do you think that I don’t use many, many types of analytics in my arguments with my friends?
That last point is crucial: how are analytics used? Wilbon treats them like they’re just this scary, monolithic entity, “advanced analytics”, that are trying to take the game away from “purists”. But analytics are just information. Just a good way to put numbers next to things that we either had to guess at before, or maybe didn’t even realize was important.
There are lots of types of analytics too…we’ve talked a bit about boxscore analytics and impact analytics, but there are a rich number scouting-based analytics as well (like the ones that Livingston referred to in the article, that are changing the ways that teams prepare for opponents). And again, those analytics are just added information. Added info…NOT replacement info. Old school former players like Nate McMillan and Larry Bird absolutely know the game in ways that analyst Amin Elhassan won’t. But the reverse is also true…Elhassan’s grasp on analytics gives him some insights that a purist may have missed as well.
These days, the true basketball nerd is able to acccess and use any and all of these strands of information. And yes, Mr. Wilbon, some of the basketball nerds look like me and you. We’re not scared of NBA Analytics, because we understand them and use them all of the time…and Mike, if you holla at me like I suggested, I’ll help you get over your fear of the numbers as well!
If you’re interested in my takes throughout the week, you can follow me on Twitter @ProfessorDrz. Also, don’t forget that you can catch me on the radio on Rotowire Fantasy Sports Today with Chris Liss and Jeff Erickson on XM 87, Sirius 210 on Mondays at 1:30 EST. I also co-host for TYTSports basketball clips on Youtube.