Skip to content
Archive of posts filed under the Decision Theory category.

Yes on design analysis, No on “power,” No on sample size calculations

Kevin Lewis points us to this paper, “Sample-Size Planning for More Accurate Statistical Power: A Method Adjusting Sample Effect Sizes for Publication Bias and Uncertainty,” by Samantha Anderson, Ken Kelley, and Scott Maxwell. My reaction: Yes, it’s reasonable, but I have two big problems with the general approach: 1. I don’t like talk of power […]

Albert Brooks vs. the Japanese dude who won the hot dog eating contest; Jim Thorpe advances

Best argument yesterday came from Jonathan: Thorpe was played in the movies by Burt Lancaster. Caesar was played by Joseph Bologna. Ham vs bologna? Thorpe. Also, Thorpewards. Today the contest is between two people who, a commenter reminded us last round, both have names. Wit or creative eating? Your call. Again, here are the rules […]

Round 3 begins: Jim Thorpe (1) vs. Sid Caesar

Today’s contest is a tough call. Caesar is the king of live TV, a real originator—but he did not write his own material, so it could be risky to invite him without getting Carl Reiner, Woody Allen, etc., as support staff. Thorpe is a legend but probably not much of a performer. Both Caesar and […]

Meryl Streep advances and the second round is over!

Best comment yesterday came from Dalton: 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 […]

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 […]

Ellen DeGeneres vs. LeBron James (3); Pele advances

Not a lot yesterday; maybe Phil‘s right that the formation of the brackets is sometimes more fun than the actual competitions. Anyway, best argument came from Ethan: We’ve not thought at all about language. I think modern Portuguese might do better in Columbia’s neighborhood than classic French. Laplace would write equations in a universal language […]

Pele vs. Pierre Simon Laplace (2); Alan Turing advances

Best comment yesterday came from Manuel: Turing did not know how to train a machine to pass the Turing test. I’m sure Oprah knows how train a person to pass the Oprah test. But there is no Oprah test. So Turing will advance. Maybe next time we do this competition we can include Alison Bechdel. […]

“We’ve Got More Than One Model: Evaluating, comparing, and extending Bayesian predictions”

I was asked to speak at the American Association of Pharmaceutical Scientists Predictive Modeling Workshop, and a title was needed. This is what I came up with: We’ve Got More Than One Model: Evaluating, comparing, and extending Bayesian predictions It’s the Bayesian Workflow stuff we’ve been pushing for awhile. But I like this new title.

Oprah Winfrey (1) vs. Alan Turing (4); Nora Ephron advances

Yesterday Diana gave an eloquent argument in favor of Voltaire, but then I came across this comment from Dzhaughn: I am concened, as Dalton was earlier, about the risk of uniting the Ephron-Streep-Child triumvirate. This could lead to bad things. Julie vs. Julia, courtroom drama. Jules or Julia, cashing in on the franchise. Meatless in […]

Voltaire (4) vs. Nora Ephron; Veronica Geng advances

Jonathan informs us: Geng’s first contribution to the New Yorker was a July 12th 1976 parody of Martin Gardner. With Gardner eliminated, the Gardner partisans among us can jump on a new wagon. (Alternatively, Geng could simply be replaced by Gardner, but the structure of this contest is anarchic enough already.) In any case, here […]

George H. W. Bush (2) vs. Veronica Geng; Mel Brooks advances

Manuel writes: Old Frankenstein vs Young Frankenstein? Like Prior Frankenstein vs Posterior Frankenstein? Posterior Frankenstein would incorporate more information, so in my opinion it would be preferable for the seminar. But who is Posterior here? Old Frankenstein is older, it should incorporate more information. But he came first. If we exclude time travel, Young Frankenstein […]

Boris Karloff (3) vs. Mel Brooks; Riad Sattouf advances

In yesterday’s contest, Dalton asks: Lance Armstrong isn’t even a GOAT. Did he cheat to get included on the list at the expense of Eddy Merckx? But then Jrc points out: Lance isn’t in for Cycling GOAT, he’s in for NGO-bracelet GOAT. I’m pretty sure he didn’t juice the bracelets. Although now that I think […]

Riad Sattouf (1) vs. Lance Armstrong; Bruce Springsteen advances

Best comment yesterday came from Jan: Now we have opportunity to see in the next round whether Julia is really that much better than Python! But that doesn’t resolve anything! So to pick a winner we’ll have to go with Tom: Python foresaw the replication crisis with their scientific method of proving someone is a […]

Statmodeling Retro

As many of you know, this blog auto-posts on twitter. That’s cool. But we also have 15 years of old posts with lots of interesting content and discussion! So I had this idea of setting up another twitter feed, Statmodeling Retro, that would start with our very first post in 2004 and then go forward, […]

Monty Python vs. Bruce Springsteen (1); Julia Child advances

From Jeff: If they meet in the semi-final the Japanese dude will eat Frank for lunch: All vs. Nothing at All. Though it appears she also had a soft spot for hot dogs, if Julia makes it that far it would be a matchup of gourmet vs gourmand, which seems a better contest. Today it’s […]

Julia Child (2) vs. Frank Sinatra (3); Dorothy Parker

For yesterday‘s contest, Jonathan gave a strong argument: First New Yorker showdown, just to see who will be taking on Veronica Geng in the finals. All the other contestants are just for show. I’m going with Liebling, because Parker wasn’t even the best New Yorker writer of her generation, being edged out by Benchley. Liebling […]

A. J. Liebling vs. Dorothy Parker (2); Steve Martin advances

As Dalton wrote: On one hand, Serena knows how to handle a racket. But Steve Martin knows how to make a racket with some strings stretched taught over a frame. Are you really gonna bet against the dude who went to toe-to-toe Kermit the Frog in racket making duel? Today we have an unseeded eater […]

Update on that study of p-hacking

Ron Berman writes: I noticed you posted an anonymous email about our working paper on p-hacking and false discovery, but was a bit surprised that it references an early version of the paper. We addressed the issues mentioned in the post more than two months ago in a version that has been available online since […]

Serena Williams vs. Steve Martin (4); The Japanese dude who won the hot dog eating contest advances

We didn’t have much yesterday, so I went with this meta-style comment from Jesse: I’m pulling for Kobayashi if only because the longer he’s in, the more often Andrew will have to justify describing him vs using his name. The thought of Andrew introducing the speaker as “and now, here’s that Japanese dude who won […]

P-hacking in study of “p-hacking”?

Someone who wishes to remain anonymous writes: This paper [“p-Hacking and False Discovery in A/B Testing,” by Ron Berman, Leonid Pekelis, Aisling Scott, and Christophe Van den Bulte] ostensibly provides evidence of “p-hacking” in online experimentation (A/B testing) by looking at the decision to stop experiments right around thresholds for the platform presenting confidence that […]