[HBR Blog] Success Comes From Better Data, Not Better Analysis

What's your point?

That it's possible to tell talent without data? Sure. And it's even easier when you have data.

That made zero sense. What does Lebron James' education level have to do with... anything. He's not comparing Lebron James playing basketball with Lebron James playing the stock market. He is comparing picking stocks with picking talent on a basketball court. Maybe with a better education you could have understood that.

As opposed to the non-mediocre teams, who get the best players un-available. I'd like to learn how they pull off that little trick.

My grandmother's blind poodle could have told that to you, so what good does it do? Data could have told you Lowry was the better player even when Brooks was putting up his M.I.P. year. Something people who claim to rely on their "eye balls" still can't see.

 

Actually, stat tells you that in the last minute of a close game, one would want the ball in CP3's hands and not Kobe's hands. CP3 will get his team more points than Kobe, all else being equal.

 

You're missing the point.

Superstar players almost never drop out of the top 5 in the draft. This is despite some teams using archaic methods of judging players. Lebron is Lebron regardless of whether you use points/reb/assists or the super-secret-awesome stats Morey's data and analysts provide him.

Data Morey's talking about allows a team to draft someone like Carl Landry with a 2nd round pick, to understand that Kyle Lowry was more than just a "career backup", as a certain clutchfans poster like to call him.

The problem though, is that unlike football or baseball, basketball is superstar driven. So the relative value of extra data is worth much more to the Heat or the Lakers than it is to the Rockets. BECAUSE WE DON'T HAVE A SUPERSTAR.

To put it a different way, Morey's gigantic amount of data would've been awesome in the 2 seasons T-Mac and Yao were healthy(our 3-4 losses to Dallas and Utah). But much less so today for the Rockets.

 

Again, people think I'm criticizing people who enjoy analyzing data, which I'm not.

But to your 2nd point, you hit it on the head here. Spoelstra of the Heat is notorious for using advanced statistics to determine his gameplan and playing rotation. There was a segment on NBATV before a Heat game last year that talked about this. But that was a huge weakness of his as he couldn't adjust his game plan as the game goes along or depending on the opponent.

Consequently, many teams, if not the majority right now, have the access into advanced statistics nowadays, not just the Rockets. They may not be as public as Morey like to be, especially the "younger" generation--like a Morey or a Spoelstra.

Although I think data/statistics do play a bigger part in generating game plans and rotations nowadays, I just think at the end of the day, basketball is a game of instincts and the fundamentals. The teams that usually play the best defense, protect the boards, and having a star win the games. Look at how far the Bulls got to where they are with the basics: a superstar, a great defense/rebounding. We all know Thibs is a clone of JVG who gives all control of the offense to his superstar.

Exactly my point. If you run an NBA team, you're not scratching your head trying to find a Landry or a Lowry. You're trying to find a Lebron James or Kobe Bryant, or even an Amare Stoudamire. Those types of players don't require more than just a pair of eye balls to know that they're really freakin' good.

What Morey does is pretty much useless b/c we will continue to be a mediocre team year in and year out unless he acquires a superstar. And he knows that.

Now go make a poll of all NBA players, coaches, GMs, fans around the world, and I GUARANTEE you more than 85% will choose Kobe. Some things go beyond stats. Kobe has the uncanny ability to create any shot he wants versus any defense in any situation, something only 2-3 other players in the NBA can do.

It's like I'm talking to a brick wall.

 

What does this have to do with the article? I understand we need a superstar. Morey understands that. And it has nothing to do with this thread.

 

1. Spoelstra using statistics inflexibly in a game isn't the fault of statistics, it's Spoelstra's fault.

2. None of the Kobe stuff you are talking about refutes the article. No one is saying there isn't stuff that goes beyond stats. I agree with you that some stuff can't be measured.

3. We keep getting off- topic, which is how to find an edge in today's NBA. Identifying superstars is NOT an edge. Everyone knows who the superstars are.

 

Title of the article: "Success Comes From Better Data, Not Better Analysis"

When the truth is neither plays the most important role in success. Between just data and analysis, Morey's likely correct. And even if he's wrong, it's not something I can disprove. But the title's false when everything's taken into account.

 

I think you're misinterpreting the title. He is talking about data vs. analysis.

 

delete

 

I don't argue with that, but I bet there are a few other teams who are on par with the Rockets in terms of that.

And I don't think we have a huge advantage over the competition anyways.

 

Just cuz you're excluding middle ground in your comments doesnt mean Morey actually is.

You're saying - using NFL as an example - you're not going to win unless you have a good quarterback, so whats the point of putting so much focus bolstering the positions you DO have? Whats the point of having all those big volumes of playbooks?

I think for the most part the quant data approach is "geek p*rn" for board members to give themselves relevance, and to support the fellow geek fraternity in passively-aggressively taking over the world. It doesnt add any enjoyment to the games for me.

But that doesnt mean I overly assume the person is ONLY using that approach.

 

This is actually a perfect example of why stats are so valuable. Everyone just knows Kobe is the most clutch. Fans know it, coaches know it, players know it. Every survey shows it. Before I read the articles, I'd have bet large amounts of money that nobody in the league was better at getting the clutch basket than Kobe.

But, it turns out that the conventional wisdom is probably wrong.

http://espn.go.com/blog/truehoop/post/_/id/24200/the-truth-about-kobe-bryant-in-crunch-time

http://probasketballtalk.nbcsports....-you-think-but-he’s-not-as-bad-as-some-think/

http://www.slate.com/id/2255932

The general consensus seems to be that Dirk, Lebron, and Chris Paul are all quite a bit better in the clutch, based on actual clutch performance rather than perception. I think even the most ardent Kobe supporters would admit that if Kobe wants to keep his crown as Most Clutch, he's got some explaining to do in the face of this data that shows he clearly isn't.

Stats are not the end all, be all (as Morey would agree) but they are valuable because they occasionally allow us to see things that we wouldn't have noticed otherwise, and they allow us to verify assumptions we hold to be true.

 

Using your own approach of regular analysis, you don't think Morey or the Rockets front office notices their own team record WITHOUT Yao and McGrady? Think they're content having 10-15 less wins a year missing the playoffs finishing below the Grizzlies?

Can GUARANTEE you that they've noticed. I don't think they've gotten myopia and all of a sudden forgot what its like to have star players on the team, like they've never seen a star player before and are totally stats dependent now. (Or that Morey knows he's failed and his giving PR spin to buy himself more time).

Semi on-topic, from an article I read years ago: ​

In 1996 in Pat Riley's 1st season with the Heat they finished with 42 wins. They had Tim Hardaway (got injured) and Alonzo Mourning.

After the season, some stat guys went to Riley saying that teams that shoot more three pointers STATISTICALLY have better offenses and better records. Riley was old school and didnt care for the three. But all he cares about is winning so he reluctantly went with it.

Next season '97 with a healthy Tim Hardaway, he let his team launch away threes at a HISTORIC rate.

They finished with 61 wins. At LEAST an extra 5 wins all because someone told him a stat. This isnt to make anyone a "believer" in stats. Like I said, when I watch the game I don't care about that stuff. Just to show that they CAN provide a competitive edge in a "non-traditional" way.

 

Th NBA is unlike other sports, where it is more individual driven. You do not need a good QB to win, as the Bucs and Ravens recently prove. And I think advanced stats would play a huge role in the NFL because coaching and playcalling are much more imperative in that sport.

In the NBA, you really keep it simple. Everyone knows the Lakers run the triangle, and knows Kobe will get the ball in the last 5 minutes. Same with the Heat or other teams with superstars. Those teams do not have huge playbooks; they simply give the ball to their stars and give them freedom.

But your points are valid.

 

Your missing the point of this thread. The reason the rockets have a ton of great role players and one of the deepest benches in the nba is because of our edge in stats that allows us to see what certain players do significantly better than others. Stats only help us get players that other teams do not value as highly. Any moron can tell us we need a star, but since every team recognizes these players as great, it's impossible to gain a competitive edge.

 

Assuming that is true, how intelligent is it to get great role players and consistently be mediocre first round fodder?

DD

 

OMG NO TANK EV4R UR NOT A REAL FAN!!!!!!111

:grin: