The Sports Illustrated Cover Jinx

Discussion in 'Road Cycling' started by Time Cop, Jun 23, 2004.

  1. Time Cop

    Time Cop New Member

    Joined:
    May 14, 2003
    Messages:
    91
    Likes Received:
    0
    Tags:


  2. Ksmanning

    Ksmanning Guest

  3. Sam

    Sam Guest

  4. Robert Chung

    Robert Chung Guest

    Sam wrote:
    > There is no such thing as the SI jinx. Learn a little
    > about a thing called regression to the mean.

    So if they put a really bad rider on the cover, he'll
    improve?
     
  5. Sam

    Sam Guest

    Maybe from a purely statistical standpoint.

    The point is that usually someone who gets on the cover has
    done something extraordinary so the only way for them to go
    is "down", but I assume that with your knowledge base you
    probably figured that out.

    "Robert Chung" <[email protected]> wrote in message news:[email protected]
    berlin.de...
    > Sam wrote:
    > > There is no such thing as the SI jinx. Learn a little
    > > about a thing called regression to the mean.
    >
    > So if they put a really bad rider on the cover, he'll
    > improve?
     
  6. "Sam" <[email protected]> writes:

    > Maybe from a purely statistical standpoint.
    >
    > The point is that usually someone who gets on the cover
    > has done something extraordinary so the only way for them
    > to go is "down", but I assume that with your knowledge
    > base you probably figured that out.
    >

    I skimmed the article.

    I did get the impression though that Mr. Price (author of
    said article) didn't fancy Armstrong's chances.

    --
    le Vent a Dos Davey Crockett No e-patents, pas de brevets
    logiciels P├ętition contre les brevets logiciels :
    http://petition.eurolinux.org Six-Day site:
    http://members.rogers.com/sixday/sixday.html Please address
    all replies to the list
     
  7. Robert Chung

    Robert Chung Guest

    Sam wrote:
    > "Robert Chung" wrote
    >> Sam wrote:
    >>> There is no such thing as the SI jinx. Learn a little
    >>> about a thing called regression to the mean.
    >>
    >> So if they put a really bad rider on the cover, he'll
    >> improve?
    >
    > Maybe from a purely statistical standpoint.

    When you recommended to the OP that he learn a little about
    regression to the mean, what other standpoint did you mean?

    > The point is that usually someone who gets on the cover
    > has done something extraordinary so the only way for them
    > to go is "down", but I assume that with your knowledge
    > base you probably figured that out.

    Oh, I figured that out, as my example may have shown. I just
    wouldn't've described that as regresssion to the mean. As a
    general rule, people who tell others to look up regression
    to the mean in the context you just used should probably
    make that recommendation a bit less frequently. Either that,
    or they should learn a little about a thing called
    regression to the mean.
     
  8. Ilan Vardi

    Ilan Vardi Guest

    "Sam" <[email protected]> wrote in message news:<[email protected]>...
    > Maybe from a purely statistical standpoint.
    >
    > The point is that usually someone who gets on the cover
    > has done something extraordinary so the only way for them
    > to go is "down", but I assume that with your knowledge
    > base you probably figured that out.

    It's your turn to learn something about statistics, which is
    that they are a statement about uncertainty and therefore
    don't apply well when there is a mechanism at work. In
    particular, someone who has done something extraordinary, as
    you put it, is more likely to do it again than someone who
    hasn't. Here's a simpler test for you: In roulette, the
    number 22 comes up 5 times in a row, which number should you
    bet on next (if you are going to bet, that is)?

    -ilan
     
  9. Ilan Vardi

    Ilan Vardi Guest

    "Robert Chung" <[email protected]> wrote in message news:<[email protected]>...
    >
    > Oh, I figured that out, as my example may have shown. I
    > just wouldn't've described that as regresssion to the
    > mean. As a general rule, people who tell others to look up
    > regression to the mean in the context you just used should
    > probably make that recommendation a bit less frequently.
    > Either that, or they should learn a little about a thing
    > called regression to the mean.

    That too.

    -ilan
     
  10. Dan Connelly

    Dan Connelly Guest

    Ilan Vardi wrote:
    > Here's a simpler test for you: In roulette, the number 22
    > comes up 5 times in a row, which number should you bet on
    > next (if you are going to bet, that is)?
    >

    22. And you'll probably lose. No jynx there, though. Same
    with any of the other jynxes (rainbow jersey, SI, ...)

    Dan
     
  11. Dan Connelly

    Dan Connelly Guest

    Dan Connelly wrote
    > No jynx there, though.

    Whoops. Jynx is a band. Lynx, but not jynx or sphynx.

    Dan
     
  12. Dan Connelly <[email protected]_e_e_e._o_r_g> writes:

    > Ilan Vardi wrote:
    > > Here's a simpler test for you: In roulette, the
    > > number 22
    > > comes up 5 times in a row, which number should you bet
    > > on next (if you are going to bet, that is)?
    > >
    >
    >
    > 22. And you'll probably lose. No jynx there, though. Same
    > with any of the other jynxes (rainbow jersey, SI,
    > ...)
    >
    > Dan

    Probability now? Just don't confuse Statistics with
    Probability. They might be often related, but are not the
    same thing at all.......... :)

    --
    le Vent a Dos. Davey Crockett http://petition.eurolinux.org
    ; http://members.rogers.com/sixday/sixday.html Please
    address all replies to the list
     
  13. Ilan Vardi

    Ilan Vardi Guest

    Dan Connelly <[email protected]_e_e_e._o_r_g> wrote in message news:<0%[email protected]>...
    >
    >
    > 22. And you'll probably lose. No jynx there, though. Same
    > with any of the other jynxes (rainbow jersey, SI,
    > ...)

    Actually, it seems to me that beating roulette shouldn't be
    that hard. In particular, the approach taken in the book
    "The Eudaenomic Pie", that is, to use differential equations
    to predict the final placement of the ball, is too heavy
    handed. Why not just use the following method: Watch where
    the ball lands and on the next bet have your team bet on
    numbers in the same half of the wheel (relative to the
    motionless table). This only requires knowing the state of
    the wheel when the ball is thrown.

    -ilan
     
  14. Robert Chung

    Robert Chung Guest

    Davey Crockett wrote:
    >
    > Probability now? Just don't confuse Statistics with
    > Probability. They might be often related, but are not the
    > same thing at all.......... :)

    I used to think I kinda knew a little about probability but
    I'm very confused by the Vice President's use of the word
    "probably" when he said yesterday that he had no regrets and
    felt better after "probably" telling
     
  15. "Robert Chung" <[email protected]> wrote in message news:[email protected]...
    > Davey Crockett wrote:
    > >
    > > Probability now? Just don't confuse Statistics with
    > > Probability. They might be often related, but are not
    > > the same thing at all.......... :)
    >
    > I used to think I kinda knew a little about probability
    > but I'm very confused by the Vice President's use of the
    > word "probably" when he said yesterday that he had no
    > regrets and felt better after "probably" telling

    >
    >

    self-absolvement was even more revealing: he justified his
    incivility by saying that it made him feel real good
    afterwards, and left it for his interlocutor to conclude
    that therefore he had done a right and just thing. And I
    thought the foreskin was ugly ....

    Kirby.
     
  16. Robert Chung <[email protected]> wrote:
    > Davey Crockett wrote:
    > >
    > > Probability now? Just don't confuse Statistics with
    > > Probability. They might be often related, but are not
    > > the same thing at all.......... :)

    > I used to think I kinda knew a little about probability
    > but I'm very confused by the Vice President's use of the
    > word "probably" when he said yesterday that he had no
    > regrets and felt better after "probably" telling

    A couple months ago I went on a little rampage after reading
    an abstract with this sentence, trying to claim that a
    particular phenomenon was worthy of note:

    The random probability of such an occurrence is
    significantly less than 3%.

    What the hell does that mean? What's significantly less than
    3% ? 2 percent? 2.5 percent? If it was 3%, would they claim
    it was significantly less than 4% ? It's still actually
    uninteresting, because it's an a posteriori probability, and
    it turns out that enough observations (on order of 30) have
    been made that you should expect something with a 3% chance
    to turn up. But damnit, just being wrong does not excuse
    illiteracy. (And that brings us back to the Bush
    administration.)
     
  17. Howard Kveck

    Howard Kveck Guest

    In article <[email protected]>,
    "Robert Chung" <[email protected]> wrote:

    > Davey Crockett wrote:
    > >
    > > Probability now? Just don't confuse Statistics with
    > > Probability. They might be often related, but are not
    > > the same thing at all.......... :)
    >
    > I used to think I kinda knew a little about probability
    > but I'm very confused by the Vice President's use of the
    > word "probably" when he said yesterday that he had no
    > regrets and felt better after "probably" telling

    Well, as we all know, Republicans are the party of family
    values, clean living and morality, while the Democrats
    are potty-mouthed sociopaths. So,, by those standards,
    Dick Cheney has obviously proven himself to be a
    Democrat. Now, the Republicans can't run a split ticket
    in the election. So they're going to have to come up with
    a new VP to run with the Boy King.

    http://webpages.charter.net/micah/newag.jpg

    --
    tanx, Howard

    "The fickleness of fame and fortune's
    caprice Together changed the life of Mason
    Reese" Alice Donut

    remove YOUR SHOES to reply, ok?
     
  18. Benjamin Weiner <[email protected]> writes:

    > A couple months ago I went on a little rampage after
    > reading an abstract with this sentence, trying to claim
    > that a particular phenomenon was worthy of note:
    >
    > The random probability of such an occurrence is
    > significantly less than 3%.
    >
    > What the hell does that mean? What's significantly less
    > than 3% ? 2 percent? 2.5 percent? If it was 3%, would they
    > claim it was significantly less than 4% ? It's still
    > actually uninteresting, because it's an a posteriori
    > probability, and it turns out that enough observations (on
    > order of 30) have been made that you should expect
    > something with a 3% chance to turn up. But damnit, just
    > being wrong does not excuse illiteracy. (And that brings
    > us back to the Bush administration.)

    What does it mean? No Problemo:

    ``Figures don't Lie, but Liars can sure Figure.''

    --
    le Vent a Dos. Davey Crockett http://petition.eurolinux.org
    ; http://members.rogers.com/sixday/sixday.html Please
    address all replies to the list
     
  19. Ilan Vardi

    Ilan Vardi Guest

    "Sam" <[email protected]> wrote in message news:<[email protected]>...
    > Maybe from a purely statistical standpoint.
    >
    > The point is that usually someone who gets on the cover
    > has done something extraordinary so the only way for them
    > to go is "down", but I assume that with your knowledge
    > base you probably figured that out.

    The data shows that of the past 20 tours de France, half
    were won by the previous year's winner.

    -ilan
     
  20. Howard Kveck wrote:

    > In article <[email protected]>,
    > "Robert Chung" <[email protected]> wrote:
    >
    >
    >>Davey Crockett wrote:
    >>
    >>>Probability now? Just don't confuse Statistics with
    >>>Probability. They might be often related, but are not the
    >>>same thing at all.......... :)
    >>
    >>I used to think I kinda knew a little about probability
    >>but I'm very confused by the Vice President's use of the
    >>word "probably" when he said yesterday that he had no
    >>regrets and felt better after "probably" telling

    >
    >
    > Well, as we all know, Republicans are the party of
    > family values, clean living and morality, while the
    > Democrats are potty-mouthed sociopaths. So,, by those
    > standards, Dick Cheney has obviously proven himself to
    > be a Democrat. Now, the Republicans can't run a split
    > ticket in the election. So they're going to have to
    > come up with a new VP to run with the Boy King.
    >
    > http://webpages.charter.net/micah/newag.jpg
    >

    http://www.playapixie.org/past/000291.php
     
Loading...
Loading...