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

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 […]

The Japanese dude who won the hot dog eating contest vs. Oscar Wilde (1); Albert Brooks advances

Yesterday I was going to go with this argument from Ethan: Now I’m morally bound to use the Erdos argument I said no one would see unless he made it to this round. Andrew will take the speaker out to dinner, prove a theorem, publish it and earn an Erdos number of 1. But then […]

More on that horrible statistical significance grid

Regarding this horrible Table 4: Eric Loken writes: The clear point or your post was that p-values (and even worse the significance versus non-significance) are a poor summary of data. The thought I’ve had lately, working with various groups of really smart and thoughtful researchers, is that Table 4 is also a model of their […]

Paul Erdos vs. Albert Brooks; Sid Caesar advances

The key question yesterday was, can Babe Didrikson Zaharias do comedy or can Sid Caesar do sports. According to Mark Palko, Sid Caesar was by all accounts extremely physically strong. And I know of no evidence that Babe was funny. So Your Show of Shows will be going into the third round. And now we […]

Sid Caesar vs. Babe Didrikson Zaharias (2); Jim Thorpe advances

Best comment from yesterday came from Dalton: Jim Thorpe isn’t from Pennsylvania, and yet a town there renamed itself after him. DJ Jazzy Jeff is from Pennsylvania, and yet Will Smith won’t even return his phone calls. Until I can enjoy a cold Yuengling in Jazzy Jeff, PA it’s DJ Jumpin’ Jim for the win. […]

Halftime! And Jim Thorpe (1) vs. DJ Jazzy Jeff

So. Here’s the bracket so far: Our first second-round match is the top-ranked GOAT—the greatest GOAT of all time, as it were—vs. an unseeded but appealing person whose name ends in f. Again here are the rules: We’re trying to pick the ultimate seminar speaker. I’m not asking for the most popular speaker, or the […]

Yakov Smirnoff advances, and Halftime!

Best argument yesterday came from Yuling: I want to learn more about missing data analysis from the seminar so I like Harry Houdini. But Yakov Smirnoff is indeed better for this topic — both Vodka and the Soviet are treatments that guarantee everyone to be Missing Completely at Random, and as statistican we definitely prefer […]

Harry Houdini (1) vs. Yakov Smirnoff; Meryl Streep advances

Best argument yesterday came from Jonathan: This one’s close. Meryl Streep and Alice Waters both have 5 letters in the first name and 6 in the last name. Tie. Both are adept at authentic accents. Tie. Meryl has played a international celebrity cook; Alice has never played an actress. Advantage Streep. Waters has taught many […]

Alice Waters (4) vs. Meryl Streep; LeBron James advances

It’s L’Bron. Only pitch for Mr. Magic was from DanC: guy actually is ultra-tall, plus grand than that non-Cav who had play’d for Miami. But Dalton brings it back for Bron: LeBron James getting to the NBA Final with J.R. Smith as his best supporting cast member is a more preposterous escape than anything David […]

Our hypotheses are not just falsifiable; they’re actually false.

Everybody’s talkin bout Popper, Lakatos, etc. I think they’re great. Falsificationist Bayes, all the way, man! But there’s something we need to be careful about. All the statistical hypotheses we ever make are false. That is, if a hypothesis becomes specific enough to make (probabilistic) predictions, we know that with enough data we will be […]