Poll

Are you a believer in advanced metrics/analytics?

Yes
21 (77.8%)
No
6 (22.2%)

Total Members Voted: 27

Author Topic: Are you a believer in advanced metrics/analytics?  (Read 13263 times)

0 Members and 1 Guest are viewing this topic.

Are you a believer in advanced metrics/analytics?
« on: March 31, 2015, 05:28:47 PM »

Offline Beat LA

  • NCE
  • Don Nelson
  • ********
  • Posts: 8338
  • Tommy Points: 896
  • Mr. Emoji
I realize that quite a number of people on here cite various statistical measurements that are supposed to reveal who is a better defender, who has a bigger impact on the outcome of a game, etc., but personally, I have to agree with Chuck - they're a load of crap and a waste of time.  Basketball isn't like baseball - there's no on-base percentage or another revealing stat that has completely revolutionized the way that executives and scouts evaluate players, and, imo, this stuff is getting completely out of control.  I wish that there was some magic formula or whatever which would show which guys will turn into all stars, but it doesn't exist, and the more formulas and categories that are created only serves to illustrate this outstanding quote by Jon Stewart.  He was talking about the problems associated with 24 hour news networks, of course, but I think that it can also be applied to advanced metrics and analytics, "If we amplify everything, we hear nothing."

https://www.youtube.com/watch?v=2asGeItzGWM

https://www.youtube.com/watch?v=agq-pL6UHDs

Re: Are you a believer in advanced metrics/analytics?
« Reply #1 on: March 31, 2015, 05:33:01 PM »

Offline D.o.s.

  • NCE
  • Cedric Maxwell
  • **************
  • Posts: 14061
  • Tommy Points: 1239
There was a great article on Grantland on this subject about a month ago. I've quoted some of the relevant bits below, but you can find the full thing here:

http://grantland.com/the-triangle/moneyball-advanced-statistics-charles-barkley-sports-media-daryl-morey-al-leiter-rob-neyer-nba-mlb-nfl-nhl/

Quote
Barkley wasn’t just wrong about advanced statistics. Speaking weeks before the ninth-annual Sloan Sports Analytics Conference (which kicks off tomorrow), he seemed to be fighting a rearguard action. “The war’s over,” CBSSports.com’s Matt Moore declared. “The nerds make the decisions whether Barkley likes it or not.” Keith Olbermann concurred: “Most of the dinosaurs like Chuck don’t even realize the war is over … ”

To which I’d ask: What war is that? The war that pitted writer versus writer, and GM versus GM, to prove once and for all that advanced stats are valuable? Sure. That war — let’s call it Moneyball I — is over.

But Barkley was firing a shot in a second war. Let’s call it Moneyball II. This clash doesn’t pit a blogger versus a newspaperman in a debate over the value of PER. It pits media versus athletes in a battle over who gets to tell the story of basketball. “I viewed Charles Barkley’s comments as being completely about media criticism, not about how a team is run,” said Craig Calcaterra, who blogs at HardballTalk. “If Barkley were still playing and a coach came to him and said, ‘Here’s something we discovered in our analytics department,’ I’m sure he’d be receptive to it. But he doesn’t want to hear someone in the media second-guessing his authority about basketball.”

Quote
In the 1960s, the sportswriter Leonard Koppett was toying with then-radical concepts like adjusting for different eras of baseball history. “Where others approached statistics as black-and-white,” David Stern once noted, “Leonard uncovered the shades of gray that gave them meaning and context.” One day, Koppett was lugging his satchel full of reference books into a press box. Jimmy Cannon, the New York Post’s resident sourpuss, smirked, “Whatcha got in there, Lennie, decimal points?”

And so it went for the next four decades. The sabermetrics-wielding writer said, “I just want information. And I’ll follow that information wherever it leads.”

The denialist said: “You’re a nerd.” Or: “You’ve never seen a ballplayer naked.” The former was a permanent condition, but statistically inclined writers would eventually be lucky enough to accomplish the latter.

There were industry changes that led to the flowering of advanced stats in the media: the crumbling of the newspapers that gave food and shelter to old-line, skeptical sportswriters; the emergence of Baseball Prospectus, FanGraphs, and Football Outsiders as incubators of young talent; the data supplied by leagues like the NHL that essentially blessed the writers who’d long been using numbers to understand sports.

There was also a philosophical shift. “I’m 41,” said Calcaterra, “and people my age or younger switched from saying, ‘When I grow up, I want to be the center fielder for the Yankees,’ to, ‘When I grow up, I want to be a general manager.’ It wasn’t because of Moneyball. It was because of computer simulations, Strat-O-Matic, and Lance Haffner baseball and football. They interested people in sports who weren’t good at them, and they made you think about sports in a different way.”

“The cult of the general manager,” as Neal Pollack called it. Front-office types became nearly as big and heroic as the players themselves. A group of sportswriters became shadow GMs — or even real executives like John Hollinger. As Calcaterra saw it, the question that animates much of today’s media is, “Can I build this team better than the way it’s being built?”

Quote
Barkley belongs to an interesting demographic: players who were rewarded by one set of statistics during their careers, but who lived long enough to see (and then confront as TV panelists) the rise of a different set. They are the Jack Morris generation: witnesses to their own reevaluation. “I have a lot of experience with guys in his position,” said MLB Network’s Brian Kenny. “They were exceedingly productive players in the ’80s and ’90s. They had success in the game. They understood the game. Their views were validated every stop along the way, in playing and broadcasting.

“Now, there’s this new wave of people telling them to learn a new language, and there’s bound to be some resistance.”

In his initial salvo, Barkley’s language was striking. “They never got the girls in high school,” he said of stats junkies. “And they just want to get in the game.” It’s a variant of the very old joke about a critic being a eunuch in a harem.

“That really stood out, didn’t it?” said Brian Kenny. “It was a lead balloon.” But you can find similar slurs from pitcher David Price, who in 2013 dismissed two TBS reporters as “nerds.” And in the puckish title of Howard Cosell’s 1985 memoir, I Never Played the Game. And in the 1970 quip of Detroit sportswriter Joe Falls. When informed by a fan that he had no right to opine on baseball, Falls replied, “My office lets me write obits and I’ve never died.”

Barkley’s problem with reporters isn’t that they’re using the wrong tools. It’s that they’re reporters. “I don’t think it’s an analytics thing,” said Andy McCullough. “It’s the Kevin Durant point about us not knowing Edited.  Profanity and masked profanity are against forum rules and may result in discipline..”

More proof comes from an analytics panel held at the NBA’s All-Star Weekend. Sitting next to Mark Cuban, Phil Jackson, and others, Barkley reprised his act as an analytics troll. But the panel ended with a surprise: The host put an article by FiveThirtyEight’s Neil Paine on the big screen. Paine had argued that advanced stats love Barkley. Barkley got a big smile on his face. Analytics were suddenly OK, even helpful, when they confirmed something Barkley already knew: He was great.
At least a goldfish with a Lincoln Log on its back goin' across your floor to your sock drawer has a miraculous connotation to it.

Re: Are you a believer in advanced metrics/analytics?
« Reply #2 on: March 31, 2015, 05:34:01 PM »

Online Roy H.

  • Forums Manager
  • James Naismith
  • *********************************
  • Posts: 58470
  • Tommy Points: -25640
  • Bo Knows: Joe Don't Know Diddley
There are dozens of advanced metrics; you can't paint all of them with a broad brush.

For instance, do you "believe" in ranking team offenses and defenses per game, or on a per-possession basis?  Do you take statistics at face value, or do you account for pace?  Do you only look at FG%, or do you consider "advanced" stats such as eFG%?

I think that an appreciation of the above stats is a necessary part of analyzing basketball.

At the same time, I'm fairly skeptical of the mathematical created formulas, such as PER, the Wages of Wins stuff, etc.  These are simply another way of packaging statistics, and they don't always generate results that appear legitimate.  I don't care how many stats are cited, I'm not accepting that Brandan Wright is a top-20 player.


I'M THE SILVERBACK GORILLA IN THIS MOTHER——— AND DON'T NONE OF YA'LL EVER FORGET IT!@ 34 minutes

Re: Are you a believer in advanced metrics/analytics?
« Reply #3 on: March 31, 2015, 05:36:38 PM »

Offline D.o.s.

  • NCE
  • Cedric Maxwell
  • **************
  • Posts: 14061
  • Tommy Points: 1239
To summarize my own thoughts on the matter : Analytics are just a way of measuring the game. They're no different from the "basic" box score statistics that, presumably, no one has a problem with. The difference, then, comes from the way people perceive how advanced stats affect the way people talk about the game. Some of that comes from a lack of transparency on behalf of the people working on the new metrics: it's basically a crime against intelligence that Hollinger never explained how PER worked while insisting that it was a better way of understanding the game, for example.

I'm pro-literacy. I believe that people should understand things. I think that the more things available, the more complete a picture can be made. There are nuances that fans, writers, athletes and thinkers can avail themselves to that were unavailable in eras prior -- but that's not a bad thing, it just means that people need to put in the work to understand how to use the new tools in the toolkit.
At least a goldfish with a Lincoln Log on its back goin' across your floor to your sock drawer has a miraculous connotation to it.

Re: Are you a believer in advanced metrics/analytics?
« Reply #4 on: March 31, 2015, 05:38:16 PM »

Offline Evantime34

  • NCE
  • Ed Macauley
  • ***********
  • Posts: 11942
  • Tommy Points: 764
  • Eagerly Awaiting the Next Fantasy Draft
Numbers should be used in conjunction with eye ball scouting. Those advanced stats should confirm what you see on the court.

I'm not sure if i'm an analytics guy, I just like to have as much information as possible.
DKC:  Rockets
CB Draft: Memphis Grizz
Players: Klay Thompson, Jabari Parker, Aaron Gordon
Next 3 picks: 4.14, 4.15, 4.19

Re: Are you a believer in advanced metrics/analytics?
« Reply #5 on: March 31, 2015, 05:42:41 PM »

Offline LooseCannon

  • NCE
  • Ed Macauley
  • ***********
  • Posts: 11833
  • Tommy Points: 950
The more useful analytics come from analyzing data not found in box scores.
"The worst thing that ever happened in sports was sports radio, and the internet is sports radio on steroids with lower IQs.” -- Brian Burke, former Toronto Maple Leafs senior adviser, at the 2013 MIT Sloan Sports Analytics Conference

Re: Are you a believer in advanced metrics/analytics?
« Reply #6 on: March 31, 2015, 05:50:49 PM »

Offline saltlover

  • Frank Ramsey
  • ************
  • Posts: 12490
  • Tommy Points: 2619
I believe they exist, sure.

Like most forms of evaluating people, they're going to catch some things that other observations miss, and miss some things that others catch.  Good analytics will try to figure what it got wrong and why, but everything has its blind spot.  Smart organizations use many evaluative tools, and some times these tools will disagree about a player.  The key is in reconciling those differences.  In the old days it used to be two scouts having opposite opinions.  Now it might be the scout vs. the computer, or even two different predictive models.  If you're drafting a player, or signing one or trading one, it's a binary decision.  Either you get him or you don't.  Unfortunately the data is always probabilistic, so you're going to get it wrong sometimes.

Analytics is used to get things wrong less.  Of course, if all the teams are using analytics, then you're right back where you started, because now it's about having the best analytics, or, more likely, getting lucky with your guesses.

Re: Are you a believer in advanced metrics/analytics?
« Reply #7 on: March 31, 2015, 06:12:39 PM »

Offline fairweatherfan

  • Johnny Most
  • ********************
  • Posts: 20738
  • Tommy Points: 2365
  • Be the posts you wish to see in the world.
Anyone who criticizes the usefulness of analytics overall (not just one or two specific flimsy metrics) should be required to provide a superior alternative and show how it's demonstrably more accurate. 

They're not perfect, and shouldn't be the only tools used, but they're a lot less imperfect than the alternatives.  It's not a coincidence or a massive fraud that every single front office in the league has bought in.

Re: Are you a believer in advanced metrics/analytics?
« Reply #8 on: March 31, 2015, 06:41:11 PM »

Offline colincb

  • NCE
  • Rajon Rondo
  • *****
  • Posts: 5095
  • Tommy Points: 501
Yes, with some reservations.  I have a strong background in the subject, so I'm pretty attuned to their use and abuse.  I do think their usage can go too far, but I don't trust other people's eyeball tests or my own for that matter.

Re: Are you a believer in advanced metrics/analytics?
« Reply #9 on: March 31, 2015, 06:48:16 PM »

Offline e4sym0de

  • Sam Hauser
  • Posts: 158
  • Tommy Points: 12
Very interesting topic. I am close to finishing my research paper about how the use of metrics can improve performance. In general, it has been empircally proven that the use of metrics indeed can improve the overall performance. This goes basicly back to the information theorie which suggests, that a larger amount of information should lead to making better decisions. Analytics use data to generate metrics which can be used as information. With that beeing said, a single metric can only show information of a limited topic and it is therefore necessary to use multiple metrics to make them usefull. Futhermore, analytics and metrics alone  should only be used as a decision making aid not as a decisionmaker themself.  The problem with highly advanced metrics like PER is, that there are way to many variables which can influence the output. It would be necesary to run complex simulations to take those variables in regard and the benefit from that could not outweight the costs.
To make it short: yes metrics are usefull, but you need to know how to use them and put them in the right context.

Re: Are you a believer in advanced metrics/analytics?
« Reply #10 on: March 31, 2015, 07:03:52 PM »

Offline GreenFaith1819

  • NCE
  • Reggie Lewis
  • ***************
  • Posts: 15402
  • Tommy Points: 2785
No, and here's why:

Metrics/Analytics do not always tell the story. For all of the number-crunching, trend analysis, stat-hounding that is apparent in sports (and even non-sports), numbers (and computers) don't measure human emotion.

Numbers don't measure toughness, guile, locker room leadership.

Advanced metrics/analytics are subject to - no matter how good they appear - human error. computers, supercomputers, anything technologically driven - is made by humans.

Humans make mistakes. Always have. Always will.

I bet that if Sam Presti would've went solely on numbers in acquiring Kendrick Perkins from BOS he wouldn't have made the trade. Perk was just removed from his injury (the same one that probably played a huge role in us losing game 7 to the Lakers in 09-10), and clearly had limitations to his game - even before his injury.

But Perk was clearly a factor in matching up with LA's twin towers of Bynum and Gasol at the time. He and Serge neutralized those two.

Perk went on to anchor that OKC defense up until maybe last year, where further injuries and wear and tear made him even less effective.

Advanced Metrics/Analytics play a role in society. But if you come at me with just numbers I will tune you out.

As should we.

If we become dependent on just numbers then we will truly lose our humanity. And that, sadly, is occurring in society as we speak.

Re: Are you a believer in advanced metrics/analytics?
« Reply #11 on: March 31, 2015, 07:11:05 PM »

Offline mctyson

  • Rajon Rondo
  • *****
  • Posts: 5087
  • Tommy Points: 372
I realize that quite a number of people on here cite various statistical measurements that are supposed to reveal who is a better defender, who has a bigger impact on the outcome of a game, etc., but personally, I have to agree with Chuck - they're a load of crap and a waste of time...

I will start of by saying yes, absolutely, because there is lots of sound evidence to support that they are meaningful.

I will also say that - like any simple statistic, you have to use advanced metrics to supplement what you see on the court.  Metrics don't define everything but they do help you examine problems better.

The Celtics use them more than most teams and I think they have used them wisely (Perkins, Rondo, Green)


Re: Are you a believer in advanced metrics/analytics?
« Reply #12 on: March 31, 2015, 07:23:50 PM »

Offline fairweatherfan

  • Johnny Most
  • ********************
  • Posts: 20738
  • Tommy Points: 2365
  • Be the posts you wish to see in the world.
No, and here's why:

Metrics/Analytics do not always tell the story. For all of the number-crunching, trend analysis, stat-hounding that is apparent in sports (and even non-sports), numbers (and computers) don't measure human emotion.

Numbers don't measure toughness, guile, locker room leadership.

Advanced metrics/analytics are subject to - no matter how good they appear - human error. computers, supercomputers, anything technologically driven - is made by humans.

Humans make mistakes. Always have. Always will.


What's the non-human alternative then?  Because it seems like every possible alternative is also human or made by humans, so I don't see why this is something unique to analytics.

Re: Are you a believer in advanced metrics/analytics?
« Reply #13 on: March 31, 2015, 07:25:41 PM »

Offline Lucky17

  • DKC Commish
  • JoJo White
  • ****************
  • Posts: 16021
  • Tommy Points: 2352
No, and here's why:

Metrics/Analytics do not always tell the story. For all of the number-crunching, trend analysis, stat-hounding that is apparent in sports (and even non-sports), numbers (and computers) don't measure human emotion.

Numbers don't measure toughness, guile, locker room leadership.

Advanced metrics/analytics are subject to - no matter how good they appear - human error. computers, supercomputers, anything technologically driven - is made by humans.

Humans make mistakes. Always have. Always will.


What's the non-human alternative then?  Because it seems like every possible alternative is also human or made by humans, so I don't see why this is something unique to analytics.

Someone needs to create an animatronic version of Billy King out of tinkertoys and an old Commodore 64 to save the Nets from a decade of failure. Just need to wait for it to become self-aware. In the meantime, he'll still do a better job.
DKC League is now on reddit!: http://www.reddit.com/r/dkcleague

Re: Are you a believer in advanced metrics/analytics?
« Reply #14 on: March 31, 2015, 07:37:40 PM »

Offline GreenFaith1819

  • NCE
  • Reggie Lewis
  • ***************
  • Posts: 15402
  • Tommy Points: 2785
No, and here's why:

Metrics/Analytics do not always tell the story. For all of the number-crunching, trend analysis, stat-hounding that is apparent in sports (and even non-sports), numbers (and computers) don't measure human emotion.

Numbers don't measure toughness, guile, locker room leadership.

Advanced metrics/analytics are subject to - no matter how good they appear - human error. computers, supercomputers, anything technologically driven - is made by humans.

Humans make mistakes. Always have. Always will.


What's the non-human alternative then?  Because it seems like every possible alternative is also human or made by humans, so I don't see why this is something unique to analytics.

For me, Faith.

Why do we feel the need to be able to analyze everything? To be able to quantify everything?

Do we want to become God? Not possible.

Because to believe in what we cannot see is crazy, psychotic - to some analysts.

But Faith is enduring, faith can see you through the tough times. Faith is believing in what you cannot see - what we cannot quantify.

Faith, to some - is crazy.