Skip to content
 

Meryl Streep vs. Yakov Smirnoff; LeBron James advances

Yesterday‘s contest was lively. The two best arguments both favor LeBron.

From Manuel:

And King James said unto them, What seemeth you best I will do. And the king stood by the gate side, and all the people came out by hundreds and by thousands.

And from Zbicyclist followed by Dalton:

LeBron has been in the NBA finals for many successive years. This year, the Lakers may not even make the playoffs. LeBron needs something to do this spring. Let’s have him talk.

Coach always said, “Try your hardest. If you don’t succeed, you can always get an ironic invitation to speak at an imaginary seminar as a consolation prize.”

There was one other good comment but I don’t want to mention it again or someone might try to run me over next time I’m in Hyde Park.

And . . . today we conclude round 2 with an unseeded person from New Jersey vs. an unseeded person whose name ends in f. Whaddya want: a fake but convincing Polish accent, or a real but exaggerated Russian accent? Your call.

Again, here’s the bracket and here are the rules.

18 Comments

  1. zbicyclist says:

    “And from Zbicyclist followed by Dalton:

    LeBron has been in the NBA finals for many successive years. This year, the Lakers may not even make the playoffs. LeBron needs something to do this spring. Let’s have him talk.

    Coach always said, “Try your hardest. If you don’t succeed, you can always get an ironic invitation to speak at an imaginary seminar as a consolation prize.””

    I am happy my lob was slammed home by Dalton. :)

  2. I can’t forgive Smirnoff for beating Houdini, but I can forgive Streep for beating Waters. Besides, when it comes to seminar leaders, I prefer a fake but convincing accent to a real but exaggerated one, since the former would not distract from the content and proceedings. Moreover, I saw her on stage (in Mother Courage, in Central Park, in 2006) and believe that if her seminar leadership (a form of being on stage) were one-fifth as good, even one-tenth, we would be glad we invited her. Never having seen Smirnoff on stage, I can’t say the same for him.

  3. Dalton says:

    Yakov Smirnoff’s relevance died with the Soviet Union. And he knows it, which is why he’s apparently now giving motivational talks entitled “Happily Ever Laughter: The Neuroscience of Romantic Relationships”.

    You can even watch in on PBS: https://www.pbssocal.org/programs/yakov-smirnoffs-happily-ever-laughter/

    When I want to experience breathlessly credulous regurgitation of bullshit science on public media, my drink of choice is Shankar Vedantam. So, please, advance Meryl Streep just so I don’t have an aneurysm by chasing my Vedantam with a shot of Smirnoff.

    • Anonymous says:

      “When I want to experience breathlessly credulous regurgitation of bullshit science on public media, my drink of choice is Shankar Vedantam. So, please, advance Meryl Streep just so I don’t have an aneurysm by chasing my Vedantam with a shot of Smirnoff.”

      I like these type of sentences. I feel today’s science could use more of them on a regular basis.

      I believe they may fit well with what i believe professor Gelman wrote about this blog, which went something like “Snark is encouraged on this blog” if i am not mistaken.

      It also reminds me of some debates in the British parliament.

    • Jeff says:

      Wait, we’re using the quality of real-world seminars delivered by the competitors as an input to this decision?

      Is that allowed?

      • Andrew says:

        Jeff:

        This reminds me of an issue that a colleague of mine raised the other day. Suppose you have a dataset with missingness and you do some imputation, so you’re working with a dataset that has some observed data and some imputed data. One can expect the observed and imputed data to look different in some ways—after all, the imputation model won’t be completely correct—and this could introduce artifacts into the analysis. For some purposes you might prefer to use the fitted model to impute all the data and work with that. Some of these issues are discussed in this paper.

        By analogy, suppose you discovered in your attic a previously unknown Leonardo painting, and it was full of missing patches, and you used some algorithm to reconstruct the missing parts. Then the resulting painting might look odd. For some purposes you’d want the partially imputed painting along with an indicator of what parts are original and what parts are imputed, but for other purposes you might just want to work with a fully imputed painting as it would have internal consistency.

        In this seminar competition, most of the data are missing, but it seems that in some cases we have direct data. So here we are with the question of whether to compare modeled (imputed, hypothetical) seminars of all the speakers, or compare hypothesized seminars of some speakers with observed seminars of others.

        • Jeff says:

          Well, that was far more substantive a response than I had a right to expect. The details of your paper are beyond my comprehension but I understand your point. I would just add a caution not to focus so much on direct data (observed seminars) that you fail to consider hypothesized seminars of those real-world speakers.

          Other notes:

          “For some purposes you might prefer to use the fitted model to impute all the data and work with that.” To extend your analogy, I propose to call this the Ecce Mono method.

          I observe that between 2004 (the date of the paper) and 2015 (the first seminar speaker competition) your thinking on how to evaluate 64 variables seems to have evolved.

  4. ksmith says:

    Streep as Smirnoff.

    And then as Thatcher.

  5. Arinarmo says:

    I don’t have an argument, but I took the liberty of making this bracket in a website where it’s easier to edit: https://challonge.com/ultsem2019

    I’ll try to keep it updated!

  6. DanC says:

    How did Smirnoff make to this round? He’s soooooo irritating.

    Streep could be in the GOAT group. Certainly the best movie actor of recent decades. She’s not an academic, but she’d give a good talk, be she’s all about excellence. She’d figure it out.

  7. Jonathan (another one) says:

    No one likes Smirnoff. Nether do I. But here’s my best shot….

    Smirnoff is the Russian Standard. Streep is the Absolut.

    Russian Standard is much better.

  8. J Storrs Hall says:

    So are there going to be any refreshments served at this seminar? I note you still have Streep, Ephron, and Julia Child in the draw, which means you could at least make a movie of cooking 524 dishes. Being hungry, I vote for Streep.

  9. Chucky77 says:

    Why are you wasting electrons on this pointless exercise?
    It has no connection to statistical modeling.

    • Andrew says:

      Chucky:

      I dunno. I do lots of things with no connection to statistical modeling. The other day I even went to the movies! You can feel free to skip the posts on this blog that don’t interest you.

      P.S. Oddly enough, this seminar competition just led to a statistical discussion! See here.

Leave a Reply