Basketball: Jalen Rose Has a Problem with Analytics (is it being used to exclude Blacks?)

playahaitian

Rising Star
Certified Pussy Poster
https://www.newyorker.com/news/q-and-a/jalen-rose-has-a-problem-with-basketball-analytics

Jalen Rose Has a Problem with Basketball Analytics



By Isaac Chotiner

June 6, 2019


Chotiner-JalenRose.jpg

Jalen Rose talks about statistical analysis, the racial dynamics underlying various sports debates, and the moon landing.

Source Photograph by Rich Barnes / Getty

Jalen Rose, who became famous as a member of the University of Michigan’s Fab Five basketball team, in the nineteen-nineties, is now one of the most recognizable figures in sports media. At Michigan, Rose was part of a team that included Chris Webber and Juwan Howard and came infamously close to winning the N.C.A.A. championship. He went on to have a successful career in the N.B.A., playing most notably with the Indiana Pacers, the Chicago Bulls, and the Toronto Raptors. Since retiring, in 2007, he has been a regular presence at ESPN and ABC, appearing on ESPN’s morning show, a radio show, and the pregame and halftime show (“NBA Countdown”) for this year’s N.B.A. Finals, in which the Raptors are playing against the Golden State Warriors. Outside of sports, Rose is known for co-founding a charter school, the Jalen Rose Leadership Academy, in his home town of Detroit.

I spoke to Rose on Wednesday, before Game 3 of the Finals, which he was covering from Oakland. I had been interested in talking about his basketball and media career, but I started by asking him about the analytics movement, which has revolutionized most major sports, and Rose and I spent most of our conversation discussing it. During the interview, which has been edited for length and clarity, we also discussed the racial dynamics that he sees underlying various sports debates, the good and the bad of the so-called “player-empowerment era,” and whether he was kidding when he recently went on television and cast doubt on the moon landing.

It seems to me that a lot of ex-players don’t like the increasing focus on analytics and advanced stats in the conversation around sports. Do you think that is fair as a generalization, and if so, why do you think that is?

Oh, wow, you are going to start off with a don’t-get-fired topic that I am going to pull you behind the curtain on, if you really want to know the impetus behind the backlash.

I’m all ears.

No. 1, there are many people that feel like it has a cultural overtone to it that basically suggests that, even though I may not have played and you did, I am smarter than you, and I know some things that you don’t know, and the numbers support me, not you. Two, you notice that, when it is a powerful job in sports—whether it is an owner, whether it is a president, whether it is a general manager, whether it is a coach—usually in football and basketball, sports that are primarily dominated by black Americans, it’s also an opportunity to funnel jobs to people by saying that, “I am smarter than you because the numbers back up what I say, and I am more read. I study more. I am able to take these numbers and manipulate my point.” It’s almost like when you hear that a player doesn’t have experience at doing X job. People that normally get the jobs you are describing don’t, either. They didn’t play at most levels, but that suffices as their “experience” and validates their opportunity for power.

Just to be clear, when you say “cultural overtones,” you mean racial overtones?

Correct. And one other point I want to make with that: it is laughable to me when playing experience gets equated to any other type of experience, including coaching. When you play—for example, somebody like me, who has been playing my entire life—for some strange reason that experience gets diminished when it’s time to talk about powerful positions in sports—like, He doesn’t have experience. There is no bigger experience than being in the foxhole, in the huddles, and out on the floor—being a part of the game plan and being game-planned against. But also all the people you learn from: your teammates, the coaches, how to navigate with the media, how to navigate with the fans. Instead of it being, He doesn’t have experience, it really should be, He has more experience than almost anybody walking the earth.

When you say “manipulate,” do you think that analytics people are doing something dishonest for their own ends, or do you just think they don’t have the experience? “Manipulate” is a loaded word.

I am saying it becomes an entry point, a validation.

And let me say this: I give myself a forehead slap like the Three Stooges when I hear players and ex-players say that somebody can’t cover the game that didn’t actually play the game. That is also extremely incorrect. Just because you played the game, that doesn’t mean that you are best at analyzing the game, or coaching the game, or working in the front office. It’s the totality of the individual that allows special people to be able to juggle and be successful at both.

VIDEO FROM THE NEW YORKER
America’s Street-Ball Mecca

For many years, political reporters would go out to some diner in Pennsylvania, and they were really experienced at travelling and meeting people and reporting their opinions. More recently, statistics experts and polling experts came in—like Nate Silver, who works at your company—and they basically said that, rather than going to diners, we can learn a lot about the American electorate and voting patterns by analyzing data in a complex way. Even though a lot of these people have no experience reporting, they have incredible insight into the American electorate. I wonder if you think that is a fair comparison.

I think what you say does make sense. I appreciate the example. It’s a great one. However, numbers can be manipulated. Look no further than our current election and how it played out. If you were paying attention to all the polls, we would not have thought that the result would end up being [what it was], if you are going only by the numbers. Numbers can be manipulated. Statistics for me are things you can count: points, rebounds, assists, blocks, steals. Analytics are things that you quantify: player-efficiency rating (P.E.R.), usage, net whatever. I think analytics should be a tool in the box, not the actual toolbox.

Right, but everything except points, such as rebounds, assists, blocks, steals—all these things are ways of analyzing the goal of the game, which is scoring more points than your opponent. So even if it is a more complex formula like P.E.R., it’s still the same thing. It’s just a more complex and accurate version of what people have been doing forever, which is looking at box scores.

The stats that I just described, for example, for me, they physically take place. I can physically watch a game and see them. By the way, I am not a big stats guy. I am an impact person.

Get the best of The New Yorker every day, in your in-box.
Meaning?

I know just because you had five steals doesn’t mean you are a great defensive player.

You better not be talking about James Harden.

Good example. He led the league in steals. [Harden ranked fourth in steals per game.]

I’m a Rockets fan. I was just trolling you a little bit.

O.K. But with those stats—points, rebounds, assists, blocks, and steals—I can still take that information, watch the players perform, and see how they got those numbers.

It seems to me that if you go back ten years, a lot of former players were really skeptical about the increasing reliance on the three-point shot. And now we are in a place where basically everyone thinks shooting lots of threes is good. There may be some debate about whether you want to shoot as many as the Rockets or the Bucks do, but basically everyone thinks you should take threes. And if you can take a shot from twenty-three feet that is worth the points, that is better than a shot from nineteen feet worth two points. Why do you think that people were so slow to adopt this idea?

So, you ready for this?

Yeah.

I don’t agree with that either.

O.K.

And by the way, I love [the Rockets general manager] Daryl Morey, for example. I have had this conversation with him. I have gone to [Morey’s] Sloan Sports Analytics Conference in Boston. I am not anti-information at all. As a founder of a charter high school, I understand that young people are judged by their G.P.A. and I understand that should represent their will. They are also judged by their SAT and ACT scores, which should represent their skills. When John Thompson was walking off of games protesting things like Prop. 48, people felt like he was overstating that the test was culturally biased—but now we are coming full circle to realizing that there was some truth there. [In 1989, Thompson, the head basketball coach at Georgetown, walked off of the court to protest an N.C.A.A. rule that tied athletic scholarships to a formula that included standardized-test scores.]

It’s the same with analytics and the three-point shot. I remember watching your Houston Rockets play and you guys had—and please help me with their roster—you had Corey Brewer, you had J.R. Smith—

Josh Smith.
 

4 Dimensional

Rising Star
Platinum Member

@4 Dimensional

you agree with this?











Yes, basic level stats can be very beneficial if you have a good teacher that is willing to also be philosophical with it.

The reason why stats stands out from the rest of math classes is because it’s the least exact math out there. It’s mostly about interpretations. My favorite example I use in class is this:

You’re nearing the end of a basketball game. You have two 90% free throw shooters. Player A has made 90 out of 100 throughout the season and player B has made 9 out of 10. Which do you put on the free throw line to win the game for you and why?
 

ifuno

Rising Star
Registered
Yes, basic level stats can be very beneficial if you have a good teacher that is willing to also be philosophical with it.

The reason why stats stands out from the rest of math classes is because it’s the least exact math out there. It’s mostly about interpretations. My favorite example I use in class is this:

You’re nearing the end of a basketball game. You have two 90% free throw shooters. Player A has made 90 out of 100 throughout the season and player B has made 9 out of 10. Which do you put on the free throw line to win the game for you and why?
Stats can tell any story you want them to tell. In this case you go with the player who has attempted more shots. 90% of one hundred is better than 90% of 10
 

playahaitian

Rising Star
Certified Pussy Poster
Yes, basic level stats can be very beneficial if you have a good teacher that is willing to also be philosophical with it.

The reason why stats stands out from the rest of math classes is because it’s the least exact math out there. It’s mostly about interpretations. My favorite example I use in class is this:

You’re nearing the end of a basketball game. You have two 90% free throw shooters. Player A has made 90 out of 100 throughout the season and player B has made 9 out of 10. Which do you put on the free throw line to win the game for you and why?

I'm going with A because that player has a much larger reliable sample size.

Sidebar...

How in the hell to you factor in the hot hand?
 

4 Dimensional

Rising Star
Platinum Member
Stats can tell any story you want them to tell. In this case you go with the player who has attempted more shots. 90% of one hundred is better than 90% of 10

I'm going with A because that player has a much larger reliable sample size.

Sidebar...

How in the hell to you factor in the hot hand?

Exactly, you go with the highest sample size. While the percentages are the same, they don’t really tell the whole story.

Stats are best used when there are raw numbers versus percentages only. Percentages can hide lack of sample size.

However, stats is also great when dealing with probability, so it really depends on how it’s used and what you use it for.

You can’t really factor in a hot hand. That’s not measurable statistically. You say a player is shooting well over a stretch of games, but statistically you can’t immediately factor in a hot hand.
 
Last edited:

4 Dimensional

Rising Star
Platinum Member
So what is your overall opinion of the rise and take over of analytics in sports?

And what sports do you think analytics is the MOST AND LEAST useful?

It’s not just sports. It’s society overall that emphasizing analytics more. Being a data analyst is one of the most lucrative fields out right now. It’s everywhere.

Anything that can be quantified can be used statistically. The use of analytics in sports have always been there. There difference is that today we use it to state some type of truth behind the numbers.

For example, simplicity say you either make or miss a shot. That assumes 50% either way, right? However, we know there is a huge difference between a layup and a 3 pointer, so we can further quantify the probability of making or missing a shot from certain distances, especially depending on the player.

If a team shoot 40% from the 3 point line, does that mean the entire team has the ability to shoot 40%? Of course not, but these are how stats can be used to justify jacking up 60 3 pointer in one game.
 
Last edited:

SpiritualPorn

Rising Star
BGOL Investor
It’s not just sports. It’s society overall that emphasizing analytics more. Being a data analyst is one of the most lucrative fields out right now. It’s everywhere.

Anything that can be quantified can be used statistically. The use of analytics in sports have always been there. There difference is that today we use it to state some type of truth behind the numbers.

For example, simplicity say you either make or miss a shot. That assumes 50% either way, right? However, we know their is a huge difference between a layup and a 3 pointer, so we can further quantify the probability of making or missing a shot from certain distances, especially depending on the player.

If a team shoot 40% from the 3 point line, does that mean the entire team has the ability to shoot 40%? Of course not, but these are how stats can be used to justify jacking up 60 3 pointer in one game.
You make great points

Do stats account for referee calls?

Unspoken injuries?

VanFleet?

In game adjustments?

Stats tell what happens, very rarely why. It's Monday morning quarterbacking at best.
 

Eclipse99

Rising Star
BGOL Investor
It’s not just sports. It’s society overall that emphasizing analytics more. Being a data analyst is one of the most lucrative fields out right now. It’s everywhere.

Anything that can be quantified can be used statistically. The use of analytics in sports have always been there. There difference is that today we use it to state some type of truth behind the numbers.

For example, simplicity say you either make or miss a shot. That assumes 50% either way, right? However, we know there is a huge difference between a layup and a 3 pointer, so we can further quantify the probability of making or missing a shot from certain distances, especially depending on the player.

If a team shoot 40% from the 3 point line, does that mean the entire team has the ability to shoot 40%? Of course not, but these are how stats can be used to justify jacking up 60 3 pointer in one game.

Great point.

And yes I work in a field of quality improvement and we use statistical data to justify why a certain company should use a strategy to improve their organization.

When it comes down to it, it's all about the probability factor. What's the probability that based on these factors, what is the most likely outcome. There will always be outliers, but we live in world that is based on the averages.
 

4 Dimensional

Rising Star
Platinum Member
You make great points

Do stats account for referee calls?

Unspoken injuries?

VanFleet?

In game adjustments?

Stats tell what happens, very rarely why. It's Monday morning quarterbacking at best.

Crazy as it sounds, stats can account for all of that stuff plus more.

We call it multivariate regression analysis (https://stats.idre.ucla.edu/stata/dae/multivariate-regression-analysis/)

This allows to take into account multiple factors and possibilities, so if I was a data analyst for a particular sports team, and if the data was accessible to me, I could come up with regression models to help predict various outcomes.

Now stats can not account for chaos because injuries are somewhat (even though we know some players are prone to injuries) unpredictable.

However, we can adjust the regression model to account for a teams output if a injury is suffered or if certain refs has a history of making certain calls at certain situations. But we are talking about adding several variables, which could be 100 different factors.

The chaotic factor is what can't be measured. We know it's very likely James Harden is going to dribble the shot clock out and shoot a contested step-back 3 pointer most of his possessions. But for some reason, it still can't be stopped. That because, as you mentioned, stats can't explain why it can't be stopped and those nerds can't either. But true basketball fans know that's 100% skill.
 
Last edited:

4 Dimensional

Rising Star
Platinum Member
Great point.

And yes I work in a field of quality improvement and we use statistical data to justify why a certain company should use a strategy to improve their organization.

When it comes down to it, it's all about the probability factor. What's the probability that based on these factors, what is the most likely outcome. There will always be outliers, but we live in world that is based on the averages.

Bro, your last statement is why all the isms and stereotypes in the world exist today. :smh:

Please don't get me started on that shit :lol:

The university I work at is all about numbers. I had to put all my ethics aside to make the numbers work because that's all they care about.

Not the quality of the education, but can I get Dumb Doe out of college algebra so the fail rates don't look as bad. When I was ethical about it, I suffered consequences for not meeting the numbers even though I can only do so much as an instructor. I can't MAKE a student do work if they simply don't want to do it. Jesus couldn't come down and force these students to do work if they don't want to do it.

But you're right about the probability factor. A lot of what we do in society comes down to that.
 

4 Dimensional

Rising Star
Platinum Member
If a team shoot 40% from the 3 point line, does that mean the entire team has the ability to shoot 40%? Of course not, but these are how stats can be used to justify jacking up 60 3 pointer in one game.

D'Antoni uses the law of averages philosophy, which is the fallacious belief that a particular outcome or event is inevitable or certain simply because it is statistically possible.

Meaning if his team shoots 40% from the three point line as a yearly average, and in game they are currently shooting 25%, then if they keep shooting, then eventually it balance out, which is not true.
 

SpiritualPorn

Rising Star
BGOL Investor
D'Antoni uses the law of averages philosophy, which is the fallacious belief that a particular outcome or event is inevitable or certain simply because it is statistically possible.

Meaning if his team shoots 40% from the three point line as a yearly average, and in game they are currently shooting 25%, then if they keep shooting, then eventually it balance out, which is not true.
I'm learning a lot in this thread

Thanks
 

daking181

Rising Star
BGOL Investor
Analytics are/can be used anyway the person wants them to look. If Mugsy Bogues is playing and Shaq scores 10 straight points that's a +/- 10 for Bogues who not even guarding him. Now when it's time to sign a contract the team will be like "you have a negative +/- so your only worth this amount" not the amount you and your agent think your worth.
 
Top