Page 1 of 6 1234 ... LastLast
Results 1 to 15 of 87
  1. #1
    NBA Superstar
    Join Date
    Jul 2006
    Posts
    14,382

    Default The Trouble with Data and Statistics

    We as a posting community have become too reliant on certain bits of information to power home our collective points. What's worse is we use a lot of this data to discredit people, their achievements and historical record to make a statement and to make us feel better about our own viewpoints. We are perverting information in the same way conspiracy theorists do to help sell our idea, while we are ignoring or undervaluing other key pieces of data and certain realities of the time.

    Using stats to dissect the 1980's or even the 1990's to a certain degree without having lived through those times leads to false findings and misleading 'facts'. Lets call it the 'Fat Lever Fallacy'. Did you know that for a four year stretch in the 80s and 90s Fat Lever's averages hovered near 19ppg, 9rpg, 8apg? Pretty gaudy numbers for a pretty forgettable player. But this was commonplace in the Western conference in that era as teams tried to run the score up and beat tired teams because they believed the travel visiting teams had to do to get to their venues (when compared to the close proximity of Eastern conference matches) gave them a huge advantage. Road teams rarely lost during this era, unless the match up was heavily tilted in one direction. As a result, the statistical referencing from this era needs to be marked with a giant asterisk because coaches like Paul Westhead and Doug Moe (who coached Fat) employed a run and gun style and their teams averaged near and sometimes over 110ppg. This was a common tactic with Western Conference teams during that era, and only the team that perfected it (the Lakers) won titles. If you look at Western conference stats from the early 80s to the mid 90s many players (especially the PG position) have inflated stats.

    Sometime in the 1990s more and more NBA teams started chartering flights and buying their own planes, there was a time not that long ago where teams flew commercial (delays, cancellations, cramped seating etc...) but with the travel disadvantage becoming minimized and training regiments and fitness levels being improved, it was no longer an easy task to run a tired team off the floor. When the Pistons and then later when the Bulls manhandled the Lakers in the 1991 finals the blueprint was beginning to change. Teams were starting to win with defense. Defense was no longer something you did in only the playoffs, it was beginning to be the main focus of many team philosophies. As a result numbers dipped, coaches like Pat Riley completely changed philosophies and a new era was ushered in. So much so that after Jordan left, many of the best players in the league were power forwards, a position previously held by team goon or low post specialist.

    This is just one example of many and it is what makes cross-era comparisons tricky. It is also what makes disproving a past occurrence with 'facts' (stats) a false positive. Yes stats are facts, but they do not tell the whole story, basketball is more jazz music than math and even if you can read the sheet music it doesn't mean you can articulate the mood of the concert, the level of the performers and the difficulties they faced, the accuracy with which they performed etc...

    let use stats to help paint a picture not frame it. And when we are ranking players in 'all time best' formats, lets not use stat comparisons as the final word, it is just wrong.

  2. #2
    Local High School Star
    Join Date
    Jun 2010
    Posts
    1,606

    Default Re: The Trouble with Data and Statistics

    if we don't rely solely on stats, would Lebron still have a career?

  3. #3
    GiveItToBurrito
    Fan in the Stands (unregistered)

    Default Re: The Trouble with Data and Statistics

    Quote Originally Posted by JtotheIzzo
    We as a posting community have become too reliant on certain bits of information to power home our collective points. What's worse is we use a lot of this data to discredit people, their achievements and historical record to make a statement and to make us feel better about our own viewpoints. We are perverting information in the same way conspiracy theorists do to help sell our idea, while we are ignoring or undervaluing other key pieces of data and certain realities of the time.

    Using stats to dissect the 1980's or even the 1990's to a certain degree without having lived through those times leads to false findings and misleading 'facts'. Lets call it the 'Fat Lever Fallacy'. Did you know that for a four year stretch in the 80s and 90s Fat Lever's averages hovered near 19ppg, 9rpg, 8apg? Pretty gaudy numbers for a pretty forgettable player. But this was commonplace in the Western conference in that era as teams tried to run the score up and beat tired teams because they believed the travel visiting teams had to do to get to their venues (when compared to the close proximity of Eastern conference matches) gave them a huge advantage. Road teams rarely lost during this era, unless the match up was heavily tilted in one direction. As a result, the statistical referencing from this era needs to be marked with a giant asterisk because coaches like Paul Westhead and Doug Moe (who coached Fat) employed a run and gun style and their teams averaged near and sometimes over 110ppg. This was a common tactic with Western Conference teams during that era, and only the team that perfected it (the Lakers) won titles. If you look at Western conference stats from the early 80s to the mid 90s many players (especially the PG position) have inflated stats.

    Sometime in the 1990s more and more NBA teams started chartering flights and buying their own planes, there was a time not that long ago where teams flew commercial (delays, cancellations, cramped seating etc...) but with the travel disadvantage becoming minimized and training regiments and fitness levels being improved, it was no longer an easy task to run a tired team off the floor. When the Pistons and then later when the Bulls manhandled the Lakers in the 1991 finals the blueprint was beginning to change. Teams were starting to win with defense. Defense was no longer something you did in only the playoffs, it was beginning to be the main focus of many team philosophies. As a result numbers dipped, coaches like Pat Riley completely changed philosophies and a new era was ushered in. So much so that after Jordan left, many of the best players in the league were power forwards, a position previously held by team goon or low post specialist.

    This is just one example of many and it is what makes cross-era comparisons tricky. It is also what makes disproving a past occurrence with 'facts' (stats) a false positive. Yes stats are facts, but they do not tell the whole story, basketball is more jazz music than math and even if you can read the sheet music it doesn't mean you can articulate the mood of the concert, the level of the performers and the difficulties they faced, the accuracy with which they performed etc...

    let use stats to help paint a picture not frame it. And when we are ranking players in 'all time best' formats, lets not use stat comparisons as the final word, it is just wrong.
    I'm a pretty big stats guy and I agree with this, even though the title had me psyched up to write some rant about people being too lazy to bother to understand what they do and don't do. It's hard to look at simple rating, w/s, WARP, +/-, etc. and do a strict comparison between two guys due to there really being no way to account for a guy's role and the context of his actions. Also, comparing guys from different eras can be hard due to the rise in offensive efficiency and decline in pace throughout the years as well as the three point shot opening up the middle for guards who can now get to the line at a much easier rate.

  4. #4
    phal5 catch24's Avatar
    Join Date
    Mar 2009
    Posts
    6,213

    Default Re: The Trouble with Data and Statistics


  5. #5
    Bringer of Rain AlphaWolf24's Avatar
    Join Date
    Jul 2010
    Posts
    8,696

    Default Re: The Trouble with Data and Statistics

    Quote Originally Posted by JtotheIzzo
    We as a posting community have become too reliant on certain bits of information to power home our collective points. What's worse is we use a lot of this data to discredit people, their achievements and historical record to make a statement and to make us feel better about our own viewpoints. We are perverting information in the same way conspiracy theorists do to help sell our idea, while we are ignoring or undervaluing other key pieces of data and certain realities of the time.

    Using stats to dissect the 1980's or even the 1990's to a certain degree without having lived through those times leads to false findings and misleading 'facts'. Lets call it the 'Fat Lever Fallacy'. Did you know that for a four year stretch in the 80s and 90s Fat Lever's averages hovered near 19ppg, 9rpg, 8apg? Pretty gaudy numbers for a pretty forgettable player. But this was commonplace in the Western conference in that era as teams tried to run the score up and beat tired teams because they believed the travel visiting teams had to do to get to their venues (when compared to the close proximity of Eastern conference matches) gave them a huge advantage. Road teams rarely lost during this era, unless the match up was heavily tilted in one direction. As a result, the statistical referencing from this era needs to be marked with a giant asterisk because coaches like Paul Westhead and Doug Moe (who coached Fat) employed a run and gun style and their teams averaged near and sometimes over 110ppg. This was a common tactic with Western Conference teams during that era, and only the team that perfected it (the Lakers) won titles. If you look at Western conference stats from the early 80s to the mid 90s many players (especially the PG position) have inflated stats.

    Sometime in the 1990s more and more NBA teams started chartering flights and buying their own planes, there was a time not that long ago where teams flew commercial (delays, cancellations, cramped seating etc...) but with the travel disadvantage becoming minimized and training regiments and fitness levels being improved, it was no longer an easy task to run a tired team off the floor. When the Pistons and then later when the Bulls manhandled the Lakers in the 1991 finals the blueprint was beginning to change. Teams were starting to win with defense. Defense was no longer something you did in only the playoffs, it was beginning to be the main focus of many team philosophies. As a result numbers dipped, coaches like Pat Riley completely changed philosophies and a new era was ushered in. So much so that after Jordan left, many of the best players in the league were power forwards, a position previously held by team goon or low post specialist.

    This is just one example of many and it is what makes cross-era comparisons tricky. It is also what makes disproving a past occurrence with 'facts' (stats) a false positive. Yes stats are facts, but they do not tell the whole story, basketball is more jazz music than math and even if you can read the sheet music it doesn't mean you can articulate the mood of the concert, the level of the performers and the difficulties they faced, the accuracy with which they performed etc...

    let use stats to help paint a picture not frame it. And when we are ranking players in 'all time best' formats, lets not use stat comparisons as the final word, it is just wrong.
    your post is a masterpiece......I think most fans view basketball and sports like this...


    it's only a very small percentage of (mostly online) elitists self described "hardcore" fans that want to use stats to move their views....




    well played ...

    __________________________________________________ ____________

    Edit :I suspect many people here will stay far far away from this thread
    Last edited by AlphaWolf24; 09-17-2011 at 12:25 PM.

  6. #6
    Local High School Star
    Join Date
    May 2008
    Location
    Cloud 9
    Posts
    1,769

    Default Re: The Trouble with Data and Statistics

    Quote Originally Posted by GiveItToBurrito
    I'm a pretty big stats guy and I agree with this, even though the title had me psyched up to write some rant about people being too lazy to bother to understand what they do and don't do. It's hard to look at simple rating, w/s, WARP, +/-, etc. and do a strict comparison between two guys due to there really being no way to account for a guy's role and the context of his actions. Also, comparing guys from different eras can be hard due to the rise in offensive efficiency and decline in pace throughout the years as well as the three point shot opening up the middle for guards who can now get to the line at a much easier rate.
    I'll continue that most advanced stats were not designed to compare players on different teams. They were designed to compare a players personal output in comparison to either himself or others on his team in the same role. Anytime I see somebody whipping out an advanced stat like PER or +/- to compare two players who played on different teams in different eras I cringe. It just shows their lack of understanding of the matrix and the purpose it was created for.

  7. #7
    ... iamgine's Avatar
    Join Date
    Nov 2007
    Posts
    18,108

    Default Re: The Trouble with Data and Statistics

    The Trouble with Data and Statistics

    1. It's not always accurate. An inaccurate data leads to false conclusion.

    2. Not many people can understand and use stats for what it's measuring in a meaningful way.

    3. A LOT of missing useful stats. i.e how many end of quarter prayer shots a player take per game. This could lead to the player seen as very inefficient even though he's not.

  8. #8
    Local High School Star
    Join Date
    May 2008
    Location
    Cloud 9
    Posts
    1,769

    Default Re: The Trouble with Data and Statistics

    Quote Originally Posted by iamgine
    The Trouble with Data and Statistics

    1. It's not always accurate. An inaccurate data leads to false conclusion.

    2. Not many people can understand and use stats for what it's measuring in a meaningful way.

    3. A LOT of missing useful stats. i.e how many end of quarter prayer shots a player take per game. This could lead to the player seen as very inefficient even though he's not.
    That's why teams keep their own stats. So they can throw away meaningless things like that last second length of the court heave, or fouls players commit on purpose that normally gets added to the stat lines seen online and used in advanced analysis.

  9. #9
    NBA lottery pick Dave3's Avatar
    Join Date
    Apr 2009
    Posts
    5,870

    Default Re: The Trouble with Data and Statistics

    Quote Originally Posted by knightfall88
    if we don't rely solely on stats, would Lebron still have a career?
    Naw, no one would even know who he is...back to back MVP's tend to be ignored pretty often in the NBA, considering there are so many of them...

  10. #10
    Bringer of Rain AlphaWolf24's Avatar
    Join Date
    Jul 2010
    Posts
    8,696

    Default Re: The Trouble with Data and Statistics

    Quote Originally Posted by Dave3
    Naw, no one would even know who he is...back to back MVP's tend to be ignored pretty often in the NBA, considering there are so many of them...

    so he's a bigger Steve Nash then???

  11. #11
    NBA lottery pick Dave3's Avatar
    Join Date
    Apr 2009
    Posts
    5,870

    Default Re: The Trouble with Data and Statistics

    Quote Originally Posted by AlphaWolf24
    so he's a bigger Steve Nash then???
    I would reply, but I make it a point not to argue with people who try to argue 21 y/o Kobe>Shaq, or that Kobe>Jordan. Would love to explain it though if someone smarter asks.

  12. #12
    Banned
    Join Date
    Jul 2011
    Posts
    85

    Default Re: The Trouble with Data and Statistics

    Statistics are misleading but they have a point most of the time. If there were no statistics we would have a bunch of people stating a bunch of BS and their evidence would be because I watched them play. You already see that a lot with people talking about 90s, 80s, 70s, 60s players.

  13. #13
    3-time NBA All-Star
    Join Date
    Dec 2009
    Posts
    9,904

    Default Re: The Trouble with Data and Statistics

    Vegas bookies use VOLUMES of DATA and STATISTICS in their football betting lines...AND, where MILLIONS of DOLLARS are at stake.

    Of course statistics CAN be mis-leading. So what? Every year major professional sports ADD more-and-more of them. Why?

  14. #14
    Bringer of Rain AlphaWolf24's Avatar
    Join Date
    Jul 2010
    Posts
    8,696

    Default Re: The Trouble with Data and Statistics

    Quote Originally Posted by Dave3
    I would reply, but I make it a point not to argue with people who try to argue 21 y/o Kobe>Shaq, or that Kobe>Jordan. Would love to explain it though if someone smarter asks.

    great so you won't argue with the majority of fans who already voted Kobe the greatest player of his generation by a landslide...


    you will only argue with the very very very small percentage of online fans who are "hardcore" and use flawed stats....

    great....you can have em'...we already know Kobe is the best.....and I speak for the people....





    next

  15. #15
    Bringer of Rain AlphaWolf24's Avatar
    Join Date
    Jul 2010
    Posts
    8,696

    Default Re: The Trouble with Data and Statistics

    Quote Originally Posted by jlauber
    Vegas bookies use VOLUMES of DATA and STATISTICS in their football betting lines...AND, where MILLIONS of DOLLARS are at stake.

    Of course statistics CAN be mis-leading. So what? Every year major professional sports ADD more-and-more of them. Why?

    because as Time continues to keep ticking we have numbers to help guide us when discussing (fill in random area )___________...But stats are hardly ever used as the main contributor when trying to pin point greatness...that's reserved for us watching and receiving how "said athlete".... captures our Imagination...and makes us feel.


    No one ever watches a game and doesn't understand what they see....no one ever needs to go back and look at stats and get "informed" to find out who was the best.....Combat sports doesn't do that, either does team sports.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •