How data transformed the NBA

NBA teams are changing the way they play basketball. The Houston Rockets, who boast stars like James Harden, have used data analytics to help them become championship contenders in recent seasons. Film supported by ‪@DXCTechnology‬
Click here to subscribe to The Economist on KZbin: econ.st/2xvTKdy
Sponsored by DXC Technology.
An invisible powerful force is lifting professional basketball to new heights, transforming how this multibillion-dollar sport is played and, crucially how to win.
In elite sport the difference between success and failure is often the finest of margins. The Houston Rockets are one of the top teams in NBA basketball. They boast some of the sport's biggest stars, including the NBA's most valuable player in 2018.
In the past decade the Rockets have risen from mid-table mediocrity to serious NBA championship contenders. But it's not just big names that have fueled this dramatic ascent - it's big data.
The Rockets recent success owes much to their pioneering decision to start crunching data about every aspect of their game - and this is the man responsible. Computer scientist Daryl Morey is the sport's foremost data and statistics guru among NBA bosses. Ten years ago Daryl set out to find that winning formula. The Rockets were one of four NBA teams to install a pioneering video tracking system which mined raw data from games. What they discovered changed the way teams try to win.
In the 1990s, long two-point shots from just inside the three-point line were common but Darryl's data analysis showed that statistically these shots provided the worst return.
In the 2017-18 season the Rockets made more three-pointers than any other team in NBA history and this was a major reason they won more games than any of their rivals.
Professor Rajiv Masheswaran co-founded Second Spectrum. The analytics company gathers and codes a vast range of increasingly granular data for all 30 NBA teams. Cameras now track and record 3D spatial data for every player and ball movement at 25 frames per second. Machine learning technology uses this huge volume of data to produce interactive visualizations allowing teams to analyze the minutiae of their performances and achieve marginal gains on court.
There is a particular focus on all important data around player movement and the probability of making a shot.
Qualitative data analysis has even changed the type of players that successful teams like the Rockets have players today are on average leaner and more agile. When it comes to recruiting new players from the college draft each season, poring over data on player performance has given the Rockets a winning edge.
Basketball has constantly changed but it's about to enter a brave new world where data could be courtside in the hands of coaches, helping to swing a game as it happens.
For more from Economist Films visit: films.economist.com/
Check out The Economist’s full video catalogue: econ.st/20IehQk
Like The Economist on Facebook: / theeconomist
Follow The Economist on Twitter: / theeconomist
Follow us on Instagram: / theeconomist
Follow us on Medium: / the_economist

[@jrco1010]

not a good time to show this. the rockets are now ranked 13 out of 15 in the west and 19 out of 30 with the rest of the league.

[@marcuscrowley6496]

And they lived and died by the 3-pointer in 2018.

[@shtrakt]

actually, their system worked so well, that every other team in the league is starting to shoot 3pts like crazy. and surrounding all players stars with 3pt shooters. just look at the big difference on the bucks since last year or the nuggets two years ago. the only team not using a lot of 3-pointers is Memphis and they still just got their franchise-high in that category. Now teams need to make sure what makes a good defense and a good defender.

[@kyrie4451]

The data shows that Trevor Ariza as one of the most important guy in defense, and they lost him.

[@memorylove8369]

And non of these Teams be winning Championships. Simple math if I got the Best Players more likely then not I will Win especially in Basketball. More important then data, is player development and then player chemistry.

[@kyrie4451]

@@memorylove8369 tell that to LeBron.

[@chewy98ta28]

I remember Danny Ainge explaining to Bob Cousy who was a critic of the three pointer in the 80's that he only had to make 1/3 of the three pointers to get the same points as a 50% shooter. Groundbreaking stuff.

[@michaelxz1305]

so 3 point shots are useful? where can I buy this supercomputer

[@ishanmaheshwari7047]

wooosh

[@cdxer]

woosh

[@lespaul9836]

Guesses based on observations alone are meaningless. Having inferences from statistics of the data back up your guesses makes them mean something.

[@cdxer]

@@lespaul9836 double woosh

[@ShareefusMaximus]

The 3 point line came to the NBA in 1977. If it was so obviously important where were ALL the attempts for the last 30 years or even TODAY!

[@ivanbekavac2960]

Shot probability: 1% Steph Curry: hold my beer

[@walkersmith2791]

Fuck Stephania!

[@caucandyellow]

@@walkersmith2791 fuck you!

[@st8of1der]

@@walkersmith2791 Boy... That escalated quickly...

[@st8of1der]

@@caucandyellow... I mean... that really got out of hand fast!

[@maddenisrael_yt]

HAHA

[@saitenotoshuitsnaini]

Last 20 seconds of this video showed the bigger loss the rockets are regretting as they found out their data on defensive performances: Trevor Ariza.

[@ILoveCostcoPizza]

They let go of Trevor to pay the aging Chris Paul more money.

[@geleneceline2310]

@@ILoveCostcoPizza lol😂 two years later the rockets are a lottery team

[@energy4128]

I didn't know "top team" meant top 13 out of 15

[@sehz1789]

loll

[@bguice9306]

@@uman230 Science shows they're not as good though. They keep losing, and losing by blowouts.

[@jeembomb]

i hate the rockets but see what the standings are in June man..

[@vashistpatel]

Lmao.

[@john50beach18]

Second spectrum probably says they're top of the league lmao

[@elieobeid77]

I guess they didn't care about data when they signed chris paul and Carmelo anthony

[@ivanbekavac2960]

Chris Paul is one of the most efficient PG in the NBA, arguably second to Curry.

[@elieobeid77]

@@ivanbekavac2960 was like that, not anymore, certainly not this year

[@Yuri-tt4jo]

@@elieobeid77 Rockets are 10-6 with Harden and Chris Paul at the court this season. What the fuck are you talking about bro. They're still the 2nd best team in the West. They're just struggling when Harden or CP3 don't play.

[@TheRedemptionRain]

Yuri Rockets trash give it a rest

[@Fidel_cashflo]

Yuri they just got shat on by the jazz with both cp and harden

[@DJCrispyC]

At the moment the data isnt really working

[@sofiagrafa6711]

Art is missing

[@Cole9559]

low sample size

[@neofusionstylx]

Or maybe all the other teams caught up and they lost their competitive advantage

[@yoij-ov3sd]

They should have modeled the probability is they re-sign Luc and Trevor.

[@stevenl4543]

Employ better data scientists who are actual bball fans

[@103swagg]

The key is to combine analytics and traditional knowledge in order to be successful

[@ivanbekavac2960]

And have superstar players on the roster

[@Oncopoda]

What is "traditional knowledge"? And how is it not a subset of data?

[@moneyboyok]

I love this two of my favourite things basketball and data.

[@captaincrunchamerica]

So how come he didn't predict carmelos slump?

[@doyouseestroberries]

Frankly, he didn't look at the data. Last year, Melo had a PER of 12.7, which is more in line with an average rookie rather than an all-star. Melo also has had a negative Box Plus Minus the past two seasons with it following a downward trend for years. 3 point percentage wasn't high enough to warrant picking him up. The most I can think of is that he was picked up as an experiment to see if he could improve in an iso heavy environment. The only teams that should've picked him up were teams who are trying to get a better spot in the lottery.

[@wildreams]

He probably did. But he is CP3's best pal.

[@ShareefusMaximus]

@STATS DONT LIE LMFAO @"the reason is that he freezes the offense." You're going to subsitute that sh*t for @doyouseestroberries's analysis? LOL Nerds > idiots.

[@pauskie6]

How about Harden's flopping percentage?

[@ShareefusMaximus]

Your hating percentage is at an all time high.

[@john50beach18]

@@ShareefusMaximus everyone hates Harden's flopping. You must be a superfan if you think Harden isn't a flopper

[@ShareefusMaximus]

@@john50beach18 I didn't say he didn't flop. I would never do that sh*t myself, but it's part of the game. He's getting paid, so I'm not trying to touch his pockets. If they are going to let Horry, Manu and Rodman get away with it, James is just doing his job. Jokic is the worst in the NBA right now, but he's not enough of a player to attract player haters. It's just the winners getting clocked. The NBA should go back to allowing more contact and the players will stop playing to it.

[@john50beach18]

@@ShareefusMaximus I wouldn't mind if there was actual illegal contact by the defenders. But if you watch him drive to the basket, he hooks defenders arms and throws it up, which is an offensive foul if refs could see it. And this isn't even doing his job. If you watch the playoff games where he ends up shooting 20%, it's not because he's missing, it's because he chucks up a shot from 30 feet cos he's trying to get free throws for no reason, and he ends up giving up a transition dunk. It's an embarrassing way to play and doesn't work in playoffs

[@srw6772]

In fairness, flopping is ridiculously efficient

[@blackmayb3]

This video couldn't have come at a worst time lol. Take THAT for data

[@R47PH]

Michael Jordan PERFECT comment 😂😂😂😂

[@juniorzagar7354]

Analytics is so fucked up, Chris Paul is one of the leagues great mid range scorers and they want him to settle on threes and layups, they had Melo who is a mid range god and they forced him to chuck up 3s every game, dude even apologised to the bench for shooting a mid range jumper ... this team is fucked

[@juniorzagar7354]

And the analytics don’t even make sense for the players they have, They have Michael Carter-Williams who can’t shoot threes, Marquesse Chris ... Players that this data won’t agree with 🤣🤣 The rockets played themselves

[@st8of1der]

This is the best comment I've seen all day. It could only be better if it were on a video about the Grizzlies or anything pertaining to their former head coach.

[@juniorzagar7354]

B. Arinze AzoroDavis 🤣😂... It’s funny that David Fizdale actually got fired because he tried to encourage Marc Gasol to shoot more threes and he declined. Memphis winning mad games this year, why ? Because Marc Gasol is shooting threes 😂🤣

[@patrickkeaveney9472]

Incredible, I wonder which sports will be next.

[@mooreideas]

I currently work for Second Spectrum and I can say it’s truly amazing to see what we do in the background to provide data to MLS & NBA. We have a lot in the world so expect a lot from SS in 2022!

[@senrichirdesstatistiques9396]

hello I am studying data science and i am very passionate in Basketball could you telle me how can i get in this field? what studies did you do? thank you

[@punitdodwal5615]

Still they missed 27 3s in conf. finals game 7 against warriors.

[@leanalcantara4976]

punit dodwal They shot 20-30 3 pointers every game during that season. Multiply that by 82 plus playoff games. At some point, you are going to miss 27 in a row(one sequence of 27 missed shots out of about 2000 3 point shots). I’m a Warriors fan, btw.

[@punitdodwal5615]

@@leanalcantara4976 but they choked when mattered the most

[@leanalcantara4976]

No argument there

[@davilip]

@@leanalcantara4976 Your stats are slightly off there. By Houston's shooting percentages, the odds of missing 27 in a row is 1 in 118,757. Based on the average number of 3 point shots attempted in the NBA this shouldn't happen for another 90 years give or take. It was extraordinarily unlucky.

[@punitdodwal5615]

@@davilip or extraordinarily poor

[@saumini9]

When everyone has access to the same data there's no competitive advantage

[@Ilovepizza1010]

but not evernyone can use that data the same way

[@HoldLeadersAccountable]

Interesting video thanks

[@pikminlord343]

what a fascinating development

[@kam3410]

Pat Riley said this 30 years ago, rebounds equal rings. It's not that complicated but they want you to think it is, it still boils down to having 2 great players, 1 top tier 3rd player, defense and now 3 pointers. Nothing changes except how you score.

[@s44504620]

It's like a Money ball in baseball

[@rukawa23gold]

If this video came out 6 months ago, it would be have been very convincing. When Rockeys almost knocked off Warriors, they were seen as the successful story in technology innovation. Now, with Rockets struggling to make playoffs, ....... However, statistics and quantitative data analysis will continue to impact basketball decisions. All other teams has their own data analysis team. That probably neutralize the competitive advantage of Rockets.

[@desertmoon39]

I wonder if they can factor in the effect the crowd has on the player, whether its booing or cheering or nonexistent.

[@amrithdas3091]

While I like the probability percentages, it comes down to two simple things, first, how does the team outscore the other and second how does the team stop the other team from scoring. It can give teams an edge to identify winning strategies and targeted plays tailored for success against teams but when the players are humans there are always intangibles and more often than not these play a bigger role in wins. Of course, even those could be data points but we haven't yet reached the ceiling yet on how data analytics and AI can transform teams. For all the success of Rockets in the 2017-18 season, poor decisions on personnel have been their downfall this season. Someday, if this technology can effectively predict and prevent player injuries, then it can truly become transformational.

[@wenlansima537]

really cool stuff comes at 3:40

[@shehanism]

Incredible

[@josenino7140]

Can that data help them play better defense?

[@kyrie4451]

The data shows that Trevor Ariza as one of the most important guy in defense, and they lost him.

[@soniab.2765]

5:03 , which animation device are you using ??

[@Andrew-dh1ws]

1:00 ....aaand then back to mediocrity

[@Sunstirade]

then Houston Rockets missed 27 consecutive 3-pointers in Game 7 and set a all-time NBA record for bricks

[@erickromero1736]

Analytics works great in the regular season but it hasn’t show to really work in playoff situations

[@TampaTayza]

Harden talking about defense gave me a chuckle.

[@saulGoodman655]

Only thing I come to think of watching this is that star wars IV scene where luke, while he is fighting against the death star, get rid of his advanced technology tracking system by closing his eyes and following what he feels and what truly matters : intuition. I'm beginning to really understand what they meant with this and am actually quite surprised to find its echo in nba... But tbh, I've always thought (by both seeing and experiencing it personnaly) that there is this specific moment in basketball where I (or kobe lol) erased all interferences and followed instinct (killer instinct for kobe and semi clutch bastard for me I'll give you that...)

[@tynitty516]

Hey Economist could you show me which computer my Knicks are using? It's probably like a Commadore 64

[@james2otero]

Analytics is going to ruin the nba. Analytics are for the smart people wanting to feel more important or maybe above every basketball player on their team. Those number can’t tell me the heart of a player, if they have that clutch gene. Those numbers can’t predict how some good old practice can make a player great. Or What would they do for a rebound.

[@theonlyalecazam2947]

They can and will, analytics has the potential to account for everything

[@Nbafan2000]

That’s pretty cool

[@samcarney3254]

they missed 37 3's in game 7, and kept shooting, these tech guys trying to sell their shit like it solves basketball, but the human touch and feel remains far better than any supercomputer

[@KoiAquaponics]

you dumb fuck, it's called statistics. A probability isn't an actual value, but a chance of it happening. You don't think that teams are having more success due to metrics done by computer analysis?

[@foreverseethe]

@@KoiAquaponics people are saying currently for the Rockets no. Also the computer can't know when a player is hot or cold, confident or not, tired or no, when to bench them or not, high morale or not, or when to change strategy in play or not.

[@kam3410]

What computer projected the last 8 Eastern Conference finals being led by Lebron James? Yes stats are important but because basketball can be impacted by a singular performer top tier talent is the greatest "statistic" go look at every championship since 1980 and you will find the same formula, two top tier players, a very good 3rd player, good roles players and rebounds and defense. Yes number of possessions have gone up but that doesn't change anything regarding rebounds being the single greatest dictator of performance because a high performing team on average shoots under 50% fg percentage, so making sure you get more rebounds is the greatest dictator of success.

[@alexis6683]

@@kam3410 1994? 2003? 1989? 2011? 2004? Just to name a few.

[@hellfiresiayan]

Probability lmao bro the probability of missing 27 threes in a row (which the rockets did in game 7) on 35% is .65^27, which is 1/9,000,000, meaning it would take on average 9,000,000 shots before they missed 27 in a row. GTFO out of here with your stats like they matter when you have the least clutch player of all time on your team

[@ljs565]

This seems pretty intriguing! But I guess they do not actually reflect the reality given that Houston has not been so proficient among the whole nba teams nowadays

[@HoldLeadersAccountable]

do one on the NHL

[@TheDefenseIsUpNext]

Of course Daryl Morey loves this sort of ass kissing over his analytics bs which has produced 0 championships. You're sitting there with a team that plays no defense and still nothing's been done about that.

[@ivanbekavac2960]

Houston was 7th in the NBA last year in defensive rating, even better than Warriors. They slid this year losing few key players due to the large contracts they had to give.

[@TheDefenseIsUpNext]

@@ivanbekavac2960 They'll sign Ariza back maybe. That might help but the real loss might be that defensive co ordinator they had (forget his name).

[@TheDefenseIsUpNext]

@Solentos That wasn't the point. Point is Morey is hailed as a forward thinker but his teams are often lacking in something as fundamental as defense. The Warrioirs in 2015 when they had Bogut were always a top defense and Durant gave them a lot defensively as well. Why don't you mention that? Analytics are helpful but not when you don't do the basics.

[@Fidel_cashflo]

DEFENSE FILMS last years team was not lacking in defense

[@claoul9451]

French sous-titres please

[@dustinparrish2074]

Thanks

[@dharmaram7527]

Take that for data! - Coach Fizdale 🗣

[@kennytheJETT]

B4 I watch this is this about shooting more 3’s??

[@robertgeoghegan7741]

Where was this last year.

[@energy6830]

I laughed a little when Harden talked about defense.

[@tigerlife1007]

I guess the analytics team was on holiday when they traded for Westbrick

[@NateGPT]

Offensive analytics don't compare to good defence. Check the 2014 Warriors, the 93 Bulls a defensive record is the key to winning championships.

[@ZachSilveiras]

2:08 LOL front end website JavaScript code is really how they make formulas huh?

[@LuthfiRahman]

wow, if this big data and ML did not go south, it'd be something big, very big!

[@shelkonnery]

They just forgot to mention the impact it all has on the fact that now NBA games have become predictable and boring to an extent.

[@abdoudjam6846]

6:20, the scariest phrase yet to be said about the data analysis, ! beware of these guys XD !

[@john50beach18]

I guess Second Spectrum weren't able to quantify the choke factor. Analytics can only take you so far. If basketball was just about putting numbers together, there'd be no point in playing basketball anymore. The fact that Rockets with all their analytics can't win tells you it's a flawed way of looking at sport. Sport is art. And analytics isn't mathematics, it's applied mathematics, which is the furthest thing from art there is

[@user-ue9yl5ec7v]

MVP Morey

[@ohneon1998]

20/20 hindsight

[@shangoat2008]

I wonder if Vegas hired this company to calculate the odds, or they have in-house team to do that on their own...

[@raredreamfootage]

In the clutch is where analytics fails. Just ask RJ Barrett of Duke vs Gonzaga.

[@damnit7045]

wow super computers are connecting the world in so many ways wow

[@sammyndeda]

🔥🔥🔥

[@soniab.2765]

Can anyone Please suggest a free and easy to used 2-D animation software for making presentations .

[@mzia8925]

How long did it take Economist to make this? Cuz it was a long time when Houston was a top team.

[@ldsericb]

"We need to interview a player...how about this Carter-Williams guy? Must be one of those analytically-sound stars they brought in because of their data wizardry. Perfect."

[@waranle961]

How’s it working out for ya ? lol

[@ShareefusMaximus]

LOL Golden State uses advanced data too. Maybe MORE! www.goodreads.com/book/show/34466932-betaball

[@ILoveCostcoPizza]

@@ShareefusMaximus I mean that data might have played a part but Stephen Curry and Klay Thompson are the best shooters in NBA History. Take it step further, how does data explain Curry making Half Court shots consistently, was no pass data on that because only Curry started doing it as a legit shot.

[@themrhead]

What a flawed correlation. Data was for basketball as technique and practice. It was always there. It obviously tries to sell the idea data, data science, and all trending buzzwords are the key. Basketball is transforming and evolving constantly. The same tactics that worked 2 seasons ago adapted and evolved, because data, analysis and new techniques are in practice every year

[@sportingminds3311]

Who the hell wrote this script? And who decided to release this now, would of been way better to do this at the end of last season.

[@simon_patterson]

Well, maybe just before Houston's final game last season, anyway...

[@rammendozapogi]

how did those analytics help the rockets when they missed 24 treys to throw away game 7 of the WCF?

[@milessmiles65]

anyone know how to cite this video

[@jasonhockly8655]

Capture data on mental toughness

[@UnknownStuff0417]

Its there, its about quantifying/formulating it. If they haven't already, teams probably will have means to track it as well. Teams capture all their players biometrics as is.

[@ThePhlox99]

Take that for data

[@jamaln9056]

The current failure with the rockets shows that no matter how much knowledge and information you have, unless you have the pieces to put it into practice, you cant do anything.

[@homemark22]

cannot predict the true gift of human being but to detect after move data

[@Sizdothyx]

You can key in all the analytics you want, boys. Daryl Morey still doesn't have a damn Championship season and all he's proven is that while data is a NICE concept to have, at the end of the day, down the fourth quarter and with everyone's arms too heavy to lift, it's going to be down to who wants it more and who's willing to go for it.

[@locosmiley2008]

this gotta be the most enlightening thing ive heard in my life...3 points are better than 2 points...waw...

[@kareemblackston2049]

How was analytics when the rockets took 28 3s in the row and lost semi finals

[@davidjeffs2700]

We feel the Economist is a “ must read “ , particularly when explaining a complex news story on a subject you ( as a Brit ) previously knew little about ... Thank you

[@meijerrc]

Harden interviewed by The Economist . Y 'all been smokin? Not asking me about my beard?

[@NickSal310]

I knew something was up

[@kyosukedeiknives]

Man that data made them sign CP3 to one of the worst deals ever, traded their 2 wing 3 and D players, and now look.

[@welven2321]

james harden - point guard - what?? 0:56

[@MarkoMcShane]

So what changed in this season for the Rockets?

[@TheAJKahn]

About the (analytical) NBA of these days, Pop said it best, "There's no beauty in it. It's pretty boring"

[@reploid001]

Steve Kerr: 3 point trend you say? It's all because of moi

[@obsurity]

0:54 James harden isn't a point guard he is a shooting guard

[@radovanprstojevic1060]

How data ruined the NBA!

[@celsdogg]

This should be the number one comment.

[@sushirolls24]

Data keeping them out the top 8 rn

[@masaodepre9936]

I was wondering if the data could learn to James Harden how to defend ?

[@butchie0415]

In the end it’s not the number that win games. It’s the team combined skills.

[@ElforTheLandstander]

Sadly, this video came out about six months too late. Using the Rockets as an example of success would have been a great idea before the 7th game of the Western Conference Finals, where they missed a ton of 3 pointers. Now they're almost a cautionary tale.

[@dumplings4mee]

did his analytics predict game 7?

[@YisongLi]

Problem with high production video pieces over news: When video finally publishes, news has became the past. The Rockets are at the bottom of West so far this season.

[@MariE-bz2eq]

The piece kind of worked out. Too many injuries for thw rockets, but they still managed to get 4th and could have gotten the 2nd seed of they weren't plagued by injuries. Remarkable feat considering they were second to last

[@TokenBlackman7]

The Rockets are a "top team" team!? As of the release of this video, they are in 13th place (out of 15 teams), in the Western Conference, with a 11-12 season record. Shame on the script writers at The Economist!

[@Prestige_00]

They almost won conference finals last year and had the MVP.. GS are 15-9 and you'd be a fool to say they aren't the best team in the NBA

[@TokenBlackman7]

Those achievements came LAST year, for the Rockets. I think the timing of this video is what rubs people the wrong way about Houston, considering their current season record. Because of that, it's comical to refer to them as a "top team", right now.

[@ivanbekavac2960]

Houston is still top 5 team in the league

[@abakahcoffie]

Don't be a hater man

[@BigPurp9]

I think they shot this before the start of the season

[@_Manimal_]

Morey forgot to calculate how James harden doesn't play defense. Offense is half the formula.

[@jamesber722]

I remember when Charles Barkley used to always say, "Jumpshooting teams never win championships." Not anymore.

[@declup]

Maybe it's time for a 4-point shot too, since 3-pointers have become so popular.

[@declup]

For that matter, maybe dunks should only be worth point each. I'm not an avid basketball fan, but I suspect that (non-flashy) dunking is a stable, controlled way to put the ball through the hoop, and so maybe dunks are too "easy" to be worth two points each. (Of course, the difficulty of the shot itself should be also balanced against the difficulty of a team's movement and positioning during the course of play.)

[@YouCallThataKnife253]

Kind of funny hearing Michael Carter-Williams talking about data. Data is why your ass stay on the bench

[@elobenz4105]

his shot be broke fr

[@clownpenisfart]

Which datum told them getting Melo was a good idea?

[@YT.Abhinav]

I am more curious about the human database collected by Facebook and conclusions from it

[@mapleridgenaturopath]

Yeah but can data stop you from a huge blunder like losing Trevor Ariza?

[@billc.4690]

Harden a SG tho

[@jman360co]

wheres the championship ringz