Analytics doesn't work.

HereWeBoogie

HereWeBoogie

Hall of Famer
For some reason i found myself watching a 3 month old episode of Open Court today, and the entire first segment blew me away. It's like they were speaking directly about the Kings front office (yes i know they were not) but my god it was so relevant to our situation that you could use Pete/Vivek's names in the commentary and it would have been perfect.


watch the first few minutes as they rip on analytics and talk about why people with those mindsets won't succeed vs people with real basketball minds and experience



trying to find a better video because this one cut out about 5 minutes of the analytics bashing
 
Hard to take any video seriously when the only guy on the panel with front office experience is Isaiah Thomas.

Analytics is absolutely a tool that should be used in the NBA. But it has to be a part of the overall strategy. At the end of the day, the eyeball test is what will always separate good GMs from bad.
 
Last edited:
i really wish i could find the full episode online. The analytics discussion goes on for another 8 minutes or so and some great points are made.

yes it's a useful tool, but it should not be relied upon. Nerds with no real basketball experience will never outsmart the rest
 
It's actually astonishing how wrong you are.

There isn't an NBA FO that isn't armed to the teeth with statistical departments and statisticians. And as far as the analytics movement not being successful? Balderdash. At the forefront of every successful team/championship roster in the last 8-9 years (for some longer) is superior statistical data with regards to roster building. Spurs, OKC, Houston, Memphis are all teams who's FO's are known to rely heavily on analytics; and I think it's turned out pretty well for them.

Like it or not, you won't find a team anymore that doesn't rely upon and use analytics as one of the main tools in putting together rosters. This isn't just me saying this, this is knowledge from actual people who work in FO's that see how they actually work in today's NBA.

As for the video, so what? Just a bunch of "good ol' boys" who love to reminisce about their own days in the league and how their version of the league was much better. If you want to go hire Barkley or Thomas, because they're "basketball men," then more power to ya.
 
The_Jamal said:
It's actually astonishing how wrong you are.

There isn't an NBA FO that isn't armed to the teeth with statistical departments and statisticians. And as far as the analytics movement not being successful? Balderdash. At the forefront of every successful team/championship roster in the last 8-9 years (for some longer) is superior statistical data with regards to roster building. Like it or not, you won't find a team anymore that doesn't rely upon and use analytics as one of the main tools in putting together rosters. This isn't just me saying this, this is knowledge from actual people who work in FO's that see how they actually work in today's NBA.

As for the video, so what? Just a bunch of "good ol' boys" who love to reminisce about their own days in the league and how their version of the league was much better. If you want to go hire Barkley or Thomas, because they're "basketball men," then more power to ya.
Click to expand...

Isn't that the only reason we have Mullin?
 
The_Jamal said:
It's actually astonishing how wrong you are.

There isn't an NBA FO that isn't armed to the teeth with statistical departments and statisticians. And as far as the analytics movement not being successful? Balderdash. At the forefront of every successful team/championship roster in the last 8-9 years (for some longer) is superior statistical data with regards to roster building. Like it or not, you won't find a team anymore that doesn't rely upon and use analytics as one of the main tools in putting together rosters. This isn't just me saying this, this is knowledge from actual people who work in FO's that see how they actually work in today's NBA.

As for the video, so what? Just a bunch of "good ol' boys" who love to reminisce about their own days in the league and how their version of the league was much better. If you want to go hire Barkley or Thomas, because they're "basketball men," then more power to ya.
Click to expand...

I'm not sure how anyone is saying anything wrong here. Analytics are a powerful tool. They can't be the only tool. I think we'd all agree with that.

And I'm not surprised that the players rejected analytics. Firstly because they didn't exist during their playing days in nearly the same type of detail they do now. But also because superstar players (and most of the guys on the panel fit that description) were hugely successful without having to look for the type of edge that analytics could provide. That said, LeBron is a big believer in them, for what that's worth.
 
funkykingston said:
Hard to take any video seriously when the only guy on the panel with front office experience is Isaiah Thomas.

Analytics is absolutely a tool that should be used in the NBA. But it has to be a part of the overall strategy. At the end of the day, the eyeball test is what will always separate good GMs from bad.
Click to expand...

I completely agree with you.
 
For a good GM analytics should be at most a minor tool often telling him stuff he should already know/be able to feel from his understanding of the game.
For a bad GM analytics will be mis and over applied because he lacks the innate understanding to properly mount it and bound it.

Either way its never going to make a bad GM good and would be a specatcularly bad GM on its own. Far too often its used in ignorant fashion to become an impediment rather than an advantage.

P.S. No, the Lakers won their titles by acquiring a second star to put along Kobe. The Celtics won their titles, and so did Miami, by gambling all out to acquire Big Threes of superstar talent. It wasn't abut analytics. It was about the biggest talents available. Even Daryl Morey, the supposed analytics GM of analytics Gms has pretty much in order to turn things just resorted to: "you be star me sign!" That's how he turned things in Houston. Not through some silly Billy Beane foolishness.
 
Bricklayer said:
For a good GM analytics should be at most a minor tool often telling him stuff he should already know/be able to feel from his understanding of the game.
For a bad GM analytics will be mis and over applied because he lacks the innate understanding to properly mount it and bound it.

Either way its never going to make a bad GM good and would be a specatcularly bad GM on its own. Far too often its used in ignorant fashion to become an impediment rather than an advantage.

P.S. No, the Lakers won their titles by acquiring a second star to put along Kobe. The Celtics won their titles, and so did Miami, by gambling all out to acquire Big Threes of superstar talent. It wasn't abut analytics. It was about the biggest talents available. Even Daryl Morey, the supposed analytics GM of analytics Gms has pretty much in order to turn things just resorted to: "you be star me sign!" That's how he turned things in Houston. Not through some silly Billy Beane foolishness.
Click to expand...

There's book by Daniel Kahneman "Thinking. Fast and Slow". I recommend to read it (while even myself not agree with some of ideas he states there... for example that there's no "hot hand" in basketball according to statistics collected by one of his many years coathor Avmos Tversky's assistants:)) as there're many interesting concepts concerning way we are thinking could be found there.
When Kahneman speaks about "intuition" he makes following point. There's certainly valid intuition which enable experts to make good desicions without solid quantative base for it... But to be "good" intuition it had to comply to two requirements: there are should be repetative operations which expert perfrorms "all the time" and they should have feedback which is direct and immediate enough... For example, most of coach's job is ok with those limitations as coach's work both in training and in games is generally quite repetative and feedback can be figured out quite easily... Not so much for GMs. Every single decision of trade/draft/extanding/signing FA is quite singular... at least important ones. And the feedback is way from be immediate - it will take months to understand whether move was success or failure and it depends on all kinds of unknown parameters...
The real problem according to Kahneman is that as rule we're not able to understand whether our "intuition" has good chance to be the good one (mostly because of evolution which shaped us - we're overconfident in our decisions made when there's severe lack of data... as otherwise we would just stall which is very bad policy if some predator, for example, attacks). In many cases common sense/ common experience will help (e.g. if you bring together 3 superstar players and make them play together - 2 maybe not enough, 4 is too much as someone will be underused and frustrated - you'll probably have contender team) but in many cases GM's intuative decisions will be as good as flipping coin. So using advanced stats can really help to improve desicion making from flipping coins to somehow more intelligent guesses and it's good and used by all for now.
I also don't see any proof that Kings GM and advisers are using adv. stats more than others... But there's something else in hi-tech mentality which may be related to what has happened... As every hi-tech employee knows "we should be in touch with current trends in our industry" (GSW!) and "we should be innovative" (4 on 5 defence !)... It's mostly correct that in hi-tech to be "old-timer" is eventually to be bankrupt... but in basketball it's probably not so much...
 
Last edited:
The_Jamal said:
It's actually astonishing how wrong you are.

There isn't an NBA FO that isn't armed to the teeth with statistical departments and statisticians. And as far as the analytics movement not being successful? Balderdash. At the forefront of every successful team/championship roster in the last 8-9 years (for some longer) is superior statistical data with regards to roster building. Spurs, OKC, Houston, Memphis are all teams who's FO's are known to rely heavily on analytics; and I think it's turned out pretty well for them.

Like it or not, you won't find a team anymore that doesn't rely upon and use analytics as one of the main tools in putting together rosters. This isn't just me saying this, this is knowledge from actual people who work in FO's that see how they actually work in today's NBA.

As for the video, so what? Just a bunch of "good ol' boys" who love to reminisce about their own days in the league and how their version of the league was much better. If you want to go hire Barkley or Thomas, because they're "basketball men," then more power to ya.
Click to expand...

Analytics may be a tool for every front office, but it is either a starting point or used to confirm intuition. It's not a decision maker.

Analytics is the culprit in the "faster pace" talk and in the firing of Malone who was clearly tired of hearing about the analytics and why one guy should play over another.

Analytics played a role in our last draft as evident by the videos of crowd sourcing and bringing in young analyst who won a contest to prove why we should draft one guy over another. I'm sure it played a role in why we drafted another SG, as the analytics showed our current one was terrible and new guy was off the charts in college. I know our front office got cold feet on Ben because they were unsure of what they were seeing from a basketball stand point. Yet every casual observer of the Kings that I know said to a man that Ben is a good player and they'd take him on their team in a heart beat.

Basketball minds could tell you that while pace is important, you put your best players on the court and play at THEIR best pace. Coaching is an art form. It's not a math equation. Phil Jackson and POP are great with people and getting the most out of them. They can do the x's and O's, but their talent is psychology and that's what puts them over the top.

Our FO wants to think they found a way to use Rudy Gay better and they were smart for getting him and want to tout that as evidence of their analytics. But Gay made a choice to play different after being traded and needing to ressurect his reputation. Now with Malone gone, Rudy is making a choice to play one on one ball and get his. How does analytics account for human emotions, choices, determination, lack of desire,etc. It can't.

What these "old guys" are saying is that talent can spot talent. Basketball guys can see the potential and know what is needed to get the most out of it from motivation to coaching and direction. Stats can deceive if you don't understand what you're seeing. Stats will fail you if you can't address the human side of the equation. Basketball experience tells you when and how to get a guy the ball or when someone is most effective.
 
Come on people, math and analytics work just fine. They are the best tools that will ever exist. Unfortunately, only a few people can use the tool correctly. The Human brain is quite fallible...See some posters on this site for example.
 
Livinthedream said:
Analytics played a role in our last draft as evident by the videos of crowd sourcing and bringing in young analyst who won a contest to prove why we should draft one guy over another. I'm sure it played a role in why we drafted another SG, as the analytics showed our current one was terrible and new guy was off the charts in college.
Click to expand...

Yet, if you watch the video, the crowd-sourcing team preferred Payton over Stauskas.

I know our front office got cold feet on Ben because they were unsure of what they were seeing from a basketball stand point.
Click to expand...

But when did the front office get cold feet on Ben? They never included him in any trade, despite being hot after guys like Rondo or Josh Smith, and they always said that they believed in him as a player, that we didn't see him in practice, he just needed time to put it all together, stuff like that. I don't see any evidence that Ben's poor first year (and it was bad, historically bad in some ways) gave our front office any cold feet at all. He's still here, and he's starting.

Yet every casual observer of the Kings that I know said to a man that Ben is a good player and they'd take him on their team in a heart beat.
Click to expand...

That's like opposite of what the bulk of the casual observers around here said. Kings fans on the web were ripping Ben left right and backwards by the end of last season. He was considered a complete bust and a waste of a pick. Around here we were throwing his name into basically every trade proposal we could think of.

The idea that casual observers knew Ben was going to turn out good and the front office didn't is like the opposite of reality.
 
Capt. Factorial said:
Yet, if you watch the video, the crowd-sourcing team preferred Payton over Stauskas.
Click to expand...

Actually, they were split on Stauskus vs. Payton. Offense vs. Defense.

More from the video if you watch it. Mullin says the basketball guys will tell you who is good and the analytics are used to support it. Sounds familiar.

Also from the video in reference to the future, Vivek says "You don't need to know the why. You simply need to find the pattern." NBA 3.0. There you have it.

Capt. Factorial said:
But when did the front office get cold feet on Ben? They never included him in any trade, despite being hot after guys like Rondo or Josh Smith, and they always said that they believed in him as a player, that we didn't see him in practice, he just needed time to put it all together, stuff like that. I don't see any evidence that Ben's poor first year (and it was bad, historically bad in some ways) gave our front office any cold feet at all. He's still here, and he's starting.
Click to expand...

What a team says and what they do are two different things. I think your first point about the crowd sourcing seeing Payton favorably answers your second question. If data said Payton was equal to Stauskus in value, why go after a SG when PG was an obvious position of need? They didn't trust Ben.

Capt. Factorial said:
That's like opposite of what the bulk of the casual observers around here said. Kings fans on the web were ripping Ben left right and backwards by the end of last season. He was considered a complete bust and a waste of a pick. Around here we were throwing his name into basically every trade proposal we could think of.

The idea that casual observers knew Ben was going to turn out good and the front office didn't is like the opposite of reality.
Click to expand...

I'll clarify quickly. These observers I speak of are casual observers of the Kings, but lifer basketball people. They did not project him to be a star, but none said bust.

I also don't think every Kings fan thought Ben was a bust and wanted him traded. I certainly did not and championed his side on this board as best I could. But there were strong voices on the internet and media who didn't like him and to people unsure of what they are seeing, it's understandable to not argue the point and take the side of someone they respect. However, I at one point before Ben had any success, said I was absolutely sure Ben would be a good player. So not everyone thought he was a bust. Negativity usually talks louder and more often is all.
 
The thing with analytics is that the the numbers don't create themselves. They are a result of something else so in looking to the numbers for all your answers you're dumbing down the process by which to manufacture the numbers you want to produce in the first place.
 
lawlman91 said:
Analytics does not take into account personal relationships and the effect of these relationships on the court product.
Click to expand...

Which is far more important in basketball than in baseball. I only knew two people from the best baseball team I played on.
 
Glenn said:
Which is far more important in basketball than in baseball. I only knew two people from the best baseball team I played on.
Click to expand...
It differs from sport to sport, the variables being the nature of the in-game interaction between teammates and the amount of people in a team. Don't know if there are any more, though.
 
I have no doubt that analytics are a very good tool. But I don't think it can supplant the importance.of understanding the psychological aspects of team sports. What creates team chemistry or detracts from it. How best to manage people and get the best from them. How to get disparate personalities to want to play for each other or at least a common goal, not just for themselves.

It's not an entirely a wrong stereotype to say that some people are so numbers, logic and analysis obsessed, that they are tone deaf about people and relationships.

I happen to think maybe Pete and Vivek are among the tone deaf. They can't see the human side beyond the numbers.
 
I disagree analytics most Definetly be used to predict relationships between players.
 
I'm
VF21 said:
Oh, please elaborate.
Click to expand...
I'm sure you have heard of E-harmony. Also at the combine they give personality test and so forth. They could match style of play with personality types that make a cohesive unit/team.
 
Lol. Of course.

This is the same old mindless drivel spewed by the folks who can't tell the difference between an advanced stat and their shoe size. These "old school" folks will repeat the same old straw man about how GMs use advanced stats, vaporizing any confidence they would have from any serious basketball mind. Willful ignorance at best, outright stupidity at worst.

Here is the truth: a statistic is a measurable fact about something. Every statistic has parameters, and knowing those parameters is critical to whether the information you glean from a statistic is useful or not. If you don't know what a statistic is designed to measure (and most of the folks mentioned above frankly don't) and you don't know what contextual information the statistic could possibly obfuscate, chances are the statistic is going to mislead you. If you know both, you can glean invaluable information you wouldn't otherwise have.

Another truth: the evil "advanced stats" that "traditional" folks will demonize are such a vast range of different methodologies and objectives that its idiotic to lump them all into one category. You have statistics that try to measure specific skills (like TS%, rebounding%, assist%, etc.), statistics that try to measure a player's overall impact on the game (PER, Win Shares, VORP, etc.), statistics that try to catch things a player does not available in the boxscore (real +/- and all its iterations), and more.

This is the NBA. This isn't your local YMCA rec league. Every team should be, and in fact are, fighting tooth and nail to obtain valuable information to gain an upper hand against 29 other professional teams. The Spurs use advanced stats to make personnel decisions. The Heat use advanced stats to make personnel decisions. The Mavericks use advanced stats to make personnel decisions. There's every championship accounted for in the last four years. Brushing all of that away is a spectacular mistake.
 
jdbraver said:
I'm

I'm sure you have heard of E-harmony. Also at the combine they give personality test and so forth. They could match style of play with personality types that make a cohesive unit/team.
Click to expand...
Analytics typically describes on-court statistics rather than personality types
 
lawlman91 said:
Analytics does not take into account personal relationships and the effect of these relationships on the court product.
Click to expand...

maybe this explains why our nerdy FO fired a coach that the players loved/respected without thinking of the consequences to team morale, and also doesn't bother to inform players and just lets them find out major team deals via social media
 
You must log in or register to reply here.

Similar threads

Spike
  • Poll Poll
Power Rankings, Week of 12/5
Replies
0
Views
181
Spike
Spike
Spike
What Do The Kings Mean to You?
Replies
9
Views
737
Jdbraver137
J
Tetsujin
  • Locked
  • Poll Poll
[Game] Kings vs. Timberwolves 10/24/2024 7pm Pacific 10pm Eastern
18 19 20
Replies
574
Views
15K
joemama
J
captain bill
Monte Interview in the Athletic
Replies
22
Views
2K
TaxMan
TaxMan
Capt. Factorial
Can we get a quality player in the 2023 draft?
2
Replies
56
Views
5K
206Fan
2
Back
Top