Statistical Analysis. Humor. Knicks.

Tuesday, September 23, 2014

Knicks Morning News (Monday, Aug 06 2012)

  • [New York Times] Off the Dribble: 30 Seconds With David Robinson (Mon, 06 Aug 2012 03:33:58 GMT)
    “I give Kobe a lot of credit for standing up for his team,” David Robinson said of Kobe Bryant’s claim that this year’s Olympic basketball squad could beat the 1992 Dream Team.

  • 34 comments on “Knicks Morning News (Monday, Aug 06 2012)

    1. elo

      A lot of ppl are sleeping on us..but I feel like JR Smith feels…

      http://www.insidehoops.com/blog/?p=10696

      We got the championship pieces needed to make this happen…btw this is my first post ever..you guys are great..been lurking on the site for a while lol..had to post something on one of the greatest fansites ever created..knicks forever!

    2. The Honorable Cock Jowles

      elo:
      A lot of ppl are sleeping on us..but I feel like JR Smith feels…

      http://www.insidehoops.com/blog/?p=10696

      We got the championship pieces needed to make this happen…btw this is my first post ever..you guys are great..been lurking on the site for a while lol..had to post something on one of the greatest fansites ever created..knicks forever!

      Thanks for posting. Here are my friendly suggestions for being the awesome KB poster that I know you can be.

      -no more ellipses (…)
      -capitalize and punctuate the requisite shit out of respect for your fellow bro (we read a lot of words on this site; it’s better when you can tell where a sentence begins and ends)
      -never cite PER, FG%, NBA EFF, or any other statistic that is total bullshit
      -don’t allow ruruland to perform metaphoric fellatio on a particularly All-Star SF without due recourse

      Welcome!

    3. BigBlueAL

      Just read this tweet:

      ATL hired NYK assistant Kenny Atkinson. Fantastic on-court knowledge, smart on analytics.

      Coaching staff is now all Mike Woodson people (and Herb). Ugh.

    4. Juany8

      The Honorable Cock Jowles: Thanks for posting. Here are my friendly suggestions for being the awesome KB poster that I know you can be.

      -never cite PER, FG%, NBA EFF, or any other statistic that is total bullshit

      Welcome!

      You forgot Wins Produced! Unless we’re just going by statistics that only 1 person in this board believes in

    5. The Honorable Cock Jowles

      Juany8: You forgot Wins Produced! Unless we’re just going by statistics that only 1 person in this board believes in

      Oh, you’re right. Let’s use WS48 because no one seems to argue that it’s got a flawed methodology either. Or maybe let’s use none and let ruruland do his anecdotal, “expert” vision thing to tell us how great Carmelo is.

    6. er

      Real hard on for melo and ruruland today no?

      The Honorable Cock Jowles: Oh, you’re right. Let’s use WS48 because no one seems to argue that it’s got a flawed methodology either. Or maybe let’s use none and let ruruland do his anecdotal, “expert” vision thing to tell us how great Carmelo is.

      Also: hey Juany8, suck a big ol’ bag of dicks.

    7. Juany8

      The Honorable Cock Jowles: Oh, you’re right. Let’s use WS48 because no one seems to argue that it’s got a flawed methodology either. Or maybe let’s use none and let ruruland do his anecdotal, “expert” vision thing to tell us how great Carmelo is.

      No Wins Shares sucks too, really any linear metric should be treated roughly the same as something like PER. Also, learn science before you bash anyone else’s “methodology”, the vast majority of scientific learning has been done by observation, math isn’t involved until you have an appreciation for what you’re studying. If you start your analysis by looking at data, you’ve already missed half the scientific process.

    8. Open43

      The real news was NBA TV re-running all the Linsanity games back to back this weekend. I couldn’t help but watch that over the Olympics – because you know the Knicks were going to win. Probably the most fun time I’ve had as a sports fan.

    9. er

      yea i took away how bad the teams were that they beat lol……..but no it was pretty cool

      Open43:
      The real news was NBA TV re-running all the Linsanity games back to back this weekend.I couldn’t help but watch that over the Olympics – because you know the Knicks were going to win.Probably the most fun time I’ve had as a sports fan.

    10. Jafa

      Hey guys go easy on the new guy elo. I didn’t know anything about advance metrics for basketball analysis until I came to this site in 2008 so give him time and he’ll be good.

      By the way, when are we signing Chris Anderson? He’s going to come in very handy off the bench when Camby or Thomas (or both) are injured over this 82 game season.

    11. The Honorable Cock Jowles

      So Juany8, why haven’t you written the ultimate rebuttal to linear regression, dude? Apparently there’s a whole host of pseudoscientists doing microeconomic work in a peer review format that apparently operates under an enormous fallacy.

      Tell me: When scientists find that the data does not support their hypothesis, do they assume that the data is wrong?

    12. ruruland

      The Honorable Cock Jowles:
      So Juany8, why haven’t you written the ultimate rebuttal to linear regression, dude? Apparently there’s a whole host of pseudoscientists doing microeconomic work in a peer review format that apparently operates under an enormous fallacy.

      Tell me: When scientists find that the data does not support their hypothesis, do they assume that the data is wrong?

      There’s scientific consensus in macroeconomics, too.

    13. iserp

      The Honorable Cock Jowles:
      So Juany8, why haven’t you written the ultimate rebuttal to linear regression, dude? Apparently there’s a whole host of pseudoscientists doing microeconomic work in a peer review format that apparently operates under an enormous fallacy.

      Tell me: When scientists find that the data does not support their hypothesis, do they assume that the data is wrong?

      Try to fit an exponential growth with a line … and then see what happens when you try to extrapolate the results.

      And yes, microeconomy might work in a totally unscientific way. How do the brokers compare to random number generators?

    14. knicknyk

      BigBlueAL:
      Just read this tweet:

      ATL hired NYK assistant Kenny Atkinson. Fantastic on-court knowledge, smart on analytics.

      Coaching staff is now all Mike Woodson people (and Herb).Ugh.

      Yeah i saw that. I am surprised he didn’t go back to Houston to be honest. They are going to be the youngest team in the NBA so there are a lot of players there for him to develop.

    15. DDH

      jon abbey: exactly, which is about as much of a science as astrology.

      http://www.youtube.com/watch?v=bAH-o7oEiyY

      It’s not like whole hosts of Phds operating under enormous fallacies is unheard of.

      FWIW, I think Berri has a lot of good stuff to offer. I just wish he would spend a little less time addressing low hanging fruit like inaccurate judgments in ESPN headlines and contracts given out by bad GMs. Not to mention purposely misunderstanding the term “shot creation” so he can spend more time addressing Steven A.’s view of the world instead of Hollinger’s or Jeff Sagarin’s.

    16. Z-man

      All due respect, guys, I think it is an inane semantical argument to go back and forth over whether WP48 is sufficiently scientific, or scientific at all. As an interested observer, all I want to know is whether it works more consistently and accurately as a predictive and analytical tool than other available tools (including “expert” opinions based on direct observations w/o quantitative analysis.

      I have my problems with WP48 (overvalues rebounding, undervalues “average efficiency” high volume scoring, ignores the impact of steals and blocks on the psychology of the game). I also have issues with the dogmatic way that THCJ uses it, and its cousins such as PAWS40. However, saying that it is not scientific because it is not based on direct observation is total horseshit. If meteoroligists use data gathered from satellites and remote weather stations, does that make them less likely to be accurate than some aboriginal medicine man who has observed the sky for ahis whole life and had the wisdom of ages of predecessors handed down to him? I’m going with the meteoroogist.

    17. Juany8

      Z-man:
      All due respect, guys, I think it is an inane semantical argument to go back and forth over whether WP48 is sufficiently scientific, or scientific at all.As an interested observer, all I want to know is whether it works more consistently and accurately as a predictive and analytical tool than other available tools (including “expert” opinions based on direct observations w/o quantitative analysis.

      I have my problems with WP48 (overvalues rebounding, undervalues “average efficiency” high volume scoring, ignores the impact of steals and blocks on the psychology of the game). I also have issues with the dogmatic way that THCJ uses it, and its cousins such as PAWS40. However, saying that it is not scientific because it is not based on direct observation is total horseshit. If meteoroligists use data gathered from satellites and remote weather stations, does that make them less likely to be accurate than some aboriginal medicine man who has observed the sky for ahis whole life and had the wisdom of ages of predecessors handed down to him? I’m going with the meteoroogist.

      Z-man, those satellites are making the observations for humanity. At it’s core, it’s still an observational science. There is a lot of data you can gather from space, without proper knowledge of what specific data is meaningful, so unless you know what you’re doing (you’re an “expert”) the data is completely meaningless.

      Box scores wouldn’t even classify as real data for most scientists, there are a ton of inconsistencies and biases present in any game. That’s why front offices go through so much trouble to collect their own data, what’s publicly out there isn’t very good, much less sufficient to make accurate predictions with.

    18. Z-man

      All “raw” basketball stats are based on observation. There are some some issues with these stats (steals and turnovers are notoriously subjective, whereas FT made and attempted) are 100% accurate. It is also the case that things that important things that go on in a game are not recorded in box scores (e.g. deflections, altered shots) and while they are becoming more available, the advanced stats community has not been able to quantify their impact yet. However, if WP48 predicts future performance and/or team success more accurately than “expert” analysis, or PER, or WS48 or whatever, bashing it on the basis of whether it meets some arbitrary definition of “scientific” is misguided and petty.

    19. Juany8

      iserp: Try to fit an exponential growth with a line … and then see what happens when you try to extrapolate the results.

      And yes, microeconomy might work in a totally unscientific way. How do the brokers compare to random number generators?

      This is one of my favorite posts ever, how did all the economic models before 2008 do compared to what actually happened? Linear interpolation is what people use when they don’t have anything else available, it might help if you’re gambling and want to base decisions on as much information as possible, but it doesn’t reveal any fundamental truth about the game. You can correlate anything you can together, until you have a causality based model (think laws of physics, how predictive would a formula for gravitational force be if linear models were used?) you’re not really analyzing the game in any meaningful way. If WP48 helps someone make money gambling, good for them, it doesn’t mean everyone else should base their decisions on the same number

    20. Juany8

      Z-man:
      All “raw” basketball stats are based on observation. There are some some issues with these stats (steals and turnovers are notoriously subjective, whereas FT made and attempted) are 100% accurate. It is also the case that things that important things that go on in a game are not recorded in box scores (e.g. deflections, altered shots) and while they are becoming more available, the advanced stats community has not been able to quantify their impact yet. However, if WP48 predicts future performance and/or team success more accurately than “expert” analysis, or PER, or WS48 or whatever, bashing it on the basis of whether it meets some arbitrary definition of “scientific” is misguided and petty.

      Does it predict future success though? Was Landry Fields a superstar after his rookie season? Would the Mavericks have been better off keeping Kris Humpries instead of Dirk Nowitski these past 2 years (there was an actual article on WoW that argued this very point). Would you rather give minutes to Matt Barnes than Kobe Bryant on the Lakers? Hell, even Tony Parker is supposedly worse than Danny Green and Kawhi Leonard. WP would be valuable if it could predict the future, regardless of how unscientific it is. Unless you think NBA coaches and general managers have absolutely no idea what they’re doing, however, (including guys like Presti and Poppovich) then you can’t argue that WP is very predictive

    21. Juany8

      The Honorable Cock Jowles:
      So Juany8, why haven’t you written the ultimate rebuttal to linear regression, dude?

      Are you going to pay me for it? My time is valuable, and I don’t need the approval of some guy on a message board enough to just write it out for fun. Why don’t you just argue the point instead of trying to make me look like an idiot? Even if I’m totally wrong, clearly you’re not convincing anyone else with the way you “argue”, you’re a college professor, you really should be past constant ad hominem attacks.

    22. jon abbey

      not that I have so much respect for college professors (those who can’t do, teach), but I would like to see proof that THCJ really is one before reading continued references to that fact. as I recall, he was some kind of assistant or TA or waterboy or something.

    23. Nick C.

      jon abbey: not that I have so much respect for college professors (those who can’t do, teach), but I would like to see proof that THCJ really is one before reading continued references to that fact. as I recall, he was some kind of assistant or TA or waterboy or something.

      The NEA and the UFT must love you.

    24. jon abbey

      hey, it’s not my saying. maybe blame George Bernard Shaw? (attribution for the original quote seems a bit unclear, but he is credited in a few places)

    25. Juany8

      jon abbey: wait, players tend to get worse as they age? ALERT THE MEDIA.

      I’m going to do a study charting height against NBA success, I’m guessing it’ll find that most players shorter than 6’2? and taller than 7’2? struggle. HEIGHT IS JUST A VARIABLE. can I get a WoW award now?

      Even more impressive, it seems shooting well, rebounding hard, and limiting turnovers are the key to winning basketball games. Somehow, Berri managed to impress himself by figuring out the “Four Factors” that win games through correlations. I guess he’s never heard a coach talk about getting good shots or fighting for rebounds in one of those dumb sideline interviews between quarters.

    Comments are closed.