Design inference

“Born under a lucky star” or design inference?

Spread the love

In “’Lucky’ woman who won lottery four times outed as Stanford University statistics PhD” (Daily Mail, 7th August 2011), Rachel Quigley recounts the curious tale of a very lucky lady from Texas:

Ms Ginther won four lots of vast sums on lottery scratch cards, half of which were bought at the same mini mart

blockquote>First, she won $5.4 million, then a decade later, she won $2 million, then two years later $3 million and finally, in the spring of 2008, she hit a $10 million jackpot.

Which proves that darn well anything can happen.

The odds of this has been calculated at one in eighteen septillion and luck like this could only come once every quadrillion years. More.

Yes, just like Darwinism. It’s impossible, but it happened, so there.

Some say that she gamed the algorithm. The Texas Lottery Commission thinks she was “born under a lucky star.”

More than one could say for other players.

21 Replies to ““Born under a lucky star” or design inference?

  1. 1
    Eocene says:

    This is comparable to those “God wouldn’t have done things this way on Earth, so therefore Evolution is true” statements.

    In the end, it’s all religious FAITH anyway.

  2. 2
    dmullenix says:

    There’s a much better secular explanation for her luck:

    “Cracking the Scratch Lottery Code” by Jonah Lehrer in the January 31, 2011 Wired Magazine.

    http://www.wired.com/magazine/.....tery/all/1

    “Of course, it would be really nice if the computer could just spit out random digits. But that’s not possible, since the lottery corporation needs to control the number of winning tickets. The game can’t be truly random. Instead, it has to generate the illusion of randomness while actually being carefully determined.”

    And the article goes on to say how he cracked the code and was able to tell good cards from bad cards without scratching them.

  3. 3
    dmullenix says:

    “According to Forbes, the residents of Bishop, Texas, seem to believe God was behind it all.”

    A secularist immediately thinks of cheating.

  4. 4
    Ilion says:

    DMullenix:… Instead, it has to generate the illusion of randomness while actually being carefully determined.

    Just like every other computer program that has ever been written or ever will be written.

  5. 5
    Ilion says:

    DMullenix:“According to Forbes, the residents of Bishop, Texas, seem to believe God was behind it all.”

    A secularist immediately thinks of cheating.

    Our culture (unlike, say, Islamic cultures) is geared toward discouraging automatic assumptions of dishonesty.

  6. 6
    DrBot says:

    Just like every other computer program that has ever been written or ever will be written.

    Apart from all the ones that use real random number generators …

  7. 7
    velikovskys says:

    Ilion,

    Our culture (unlike, say, Islamic cultures) is geared toward discouraging automatic assumptions of dishonesty

    Really, Islamic cultures encourage the automatic assumptions of dishonesty?

    To paraphrase the link ” if this happened in Las Vegas they would arrest her first and ask questions later” ,apparently
    Las Vegas didn’t get the memo

  8. 8
    ciphertext says:

    @DrBot

    RE: Post #6

    Apart from all the ones that use real random number generators …

    What algorithm (or library…) contains a “real” random number generator? Too my knowledge, the best we can approximate are “pseudo-random” numbers. With the exception being that you connect a computer program to a device that measures radioactive decay from a sample of an element that is radioactive, and use that metric to generate a “key” (number).

  9. 9
    mike1962 says:

    ciphertext @8

    A middle ground method to generating random numbers is the “entropy” method sourced by relatively stochastic events that occur in a computer such as the network card events, mouse input, keyboard input, clock, other device interrupts, and so forth. Typically these source are strategically fed into a hashing algorithm such as MD5 or SHA. This sort of random number generation is considered “cryptographically strong” and has exceptionally high statistical randomness.

    http://en.wikipedia.org/wiki/C....._generator

  10. 10
  11. 11
    Elizabeth Liddle says:

    heh

    😀

  12. 12
  13. 13
    Ilion says:

    A person who does not want the truth @ 7:Really, Islamic cultures encourage the automatic assumptions of dishonesty?

    Yes. It all goes back to Arab tribalism.

  14. 14
    DrBot says:

    What algorithm (or library…) contains a “real” random number generator?

    None do, but a psuedo random number generator is just a source of numbers that follow a statistically random pattern. If you choose you can use a real random number generator as the source – and get the same pattern of data. One is deterministic and can be determined if you have the random seed, the other is non-deterministic.

    When using a GA it makes little difference if you use a psuedo or real random source. You can seed the psuedo function with the current system time and get different results each time you run the software. Psuedo random numbers can be better when using models because you can capture the seed and re-run the software to get data you failed to capture before.

  15. 15
    Mung says:

    Psuedo random numbers can be better when using models because you can capture the seed and re-run the software to get data you failed to capture before.

    Same input, same output.

    Just like evolution.

  16. 16
    kairosfocus says:

    My vote is for zeners or sky noise.

  17. 17
    velikovskys says:

    Ilion,

    A person who does not want the truth @ 7: “Really, Islamic cultures encourage the automatic assumptions of dishonesty?”

    Don’t mind the truth, just would like a little more info on why.Didn’t see anything about your claim the Islamic encourage automatic assumption of dishonesty in the google

  18. 18

    The real question is what are the odds of any one person (not this specific person), among all the vast numbers of lottery players, winning the lottery 4 times in a lifetime of actively playing.

    Would be interesting to know, however, whether there was some special knowledge or other non-random factor involved.

  19. 19
    avocationist says:

    Eric,

    The article said that it had been calculated as one in 18 septillion. So, what I would like to know, is what are the chances of a person winning the lottery 4 times who is also a PhD in math and statistics?

  20. 20
    DrBot says:

    Same input, same output.

    Just like evolution.

    IF the universe is entirely deterministic then yes. If the universe contains sources of genuine random noise then a GA that uses a genuine source of random noise is just like evolution in that regard.

    If I wrote an air drop simulation I would want to include in the model some stochastic elements (randomness in things like turbulence and weather) – things that are, in nature, unpredictable but constrained. I would use a psuedo-random number generator as the source because for the purposes of the model it makes no difference if the source of randomness is ultimately deterministic – what matters is the pattern (or lack thereof) and that it is a good model of the thing we are trying to model.

    It also helps sometimes that the simulation can be run again under exactly the same starting conditions.

    Same input, same output – just like a real airdrop!

  21. 21
    ciphertext says:

    @Dr. Bot

    RE: Post #14

    None do, but a psuedo random number generator is just a source of numbers that follow a statistically random pattern. If you choose you can use a real random number generator as the source – and get the same pattern of data. One is deterministic and can be determined if you have the random seed, the other is non-deterministic.

    Ahhh yes. Reminds me of an excerpt I saw on the construction (or proposed construction) of a crypto system that had a “trainable”, nondeterministic transformation algorithm. It was a “flexible” encryption algorithm, in that it allowed you to select subkeys for the encryption of different inputs messages.

    CYBERNETICS AND SYSTEMS ANALYSIS
    Volume 34, Number 5, 684-688, DOI: 10.1007/BF02667041>
    .

    (For those interested in the post)

Leave a Reply