Thursday, November 16, 2023

Post-dated foresight

As we discussed last time, Josh Marshall wasn't too impressed with the latest piece of NYT political analysis.

First a small digression. I don’t know if it was immediately before we started recording yesterday’s podcast or during the recording, but what I said was something like this: There’s this weird pattern with the Times/Sienna poll, the most recent installment of which kicked off the latest freakout. They have a poll which is bad for Democrats. Then there’s an election that’s good for Democrats. Then the Times publishes a story that says the results confirmed the findings of their poll. And the kicker is that when they dive into the inner workings of the numbers it kinda seems like they’re right? Sorta? Maybe?

Here's the central thesis of Nate Cohn's piece:

Put simply: Tuesday’s results don’t change the picture for President Biden heading into 2024.

First off, this is obviously false. A strong performance by Democrats and Democratic issues in purple and deep red States one year before a presidential election pretty much has to shift the priors of any responsible analyst. Cohn can't argue this point and, despite some hand waving, doesn't really try. Instead, he comes up with various hypotheses that tend to reduce the impact of this new data point on the upcoming presidential election. None of these theories are self-evident or even particularly well supported by the election results. As Marshall points out, he's seriously stretching the definition of low turnout election. Not to mention ignoring conflicting data and making the counter-intuitive leap that younger voters are less concerned with reproductive rights.

He also wants us to know that poll savvy folk like him weren't caught off guard in the slightest.

Abortion rights and marijuana legalization prevailed in Ohio. Democrats held the governor’s mansion in Kentucky, took full control of the State Legislature in Virginia and won a Supreme Court election in Pennsylvania. They even were competitive in Mississippi.


The polls showed the Democrat winning Kentucky. They showed abortion rights and marijuana legalization prevailing in Ohio — and showed them to be popular in many red states all over the country. They also show that voters disapprove of Mr. Biden and that Mr. Trump leads in the battleground states.

Funny thing, though. Cohn published two analyses of these polls the weekend before the election and as far as I can tell  he didn't say anything about the big news of these imminent elections. If he really saw it coming, a story about the Democrats being on the verge of a big night would have been a huge scoop. Possibly even a career highlight.

It's not like Cohn and the New York Times are reluctant to make predictions about political events like Trump not getting the nomination in 2016 or Dobbs having limited impact (see our real-time reply here) or DeSantis crushing Trump (by Michael C. Bender with Cohn contributing). If we go back to the previous Nate, we even get this...


 Almost exactly twelve years ago.


Not sure what brought this to mind...

Most billet reading is an example of a generalized class of tricks known as "one ahead" reading. It is accomplished by having the performer know one of the statements beforehand, typically through a plant, or through sleight of hand by opening one of them before starting the act.

To start the act, the mentalist selects the topmost envelope on the stack and pretends to mind-read the contents, typically by holding it to their forehead. Instead of announcing anything related to that envelope, they instead read aloud the memorized statement. The plant in the audience then cries out some variation of "that's mine!" Another variation is to claim to be unable to read the first card due to some problem, perhaps that the audience member's mind is closed or too powerful. In either event, the mentalist then opens the envelope to "make sure they got it right" or perhaps to "see what is so confusing" and is then able to read what a real audience member wrote on their billet.

The trick proceeds to the next envelope. The mentalist pretends to mind read it, but reads aloud the statement from the envelope previously opened. This time a real audience member is impressed and agrees they got it right. The mentalist then reads the contents of the second envelope and repeats this sequence. The trick then continues until the envelopes are exhausted, the last one being empty or the envelope of the plant. Throughout, the mentalist is "one ahead" in the envelope stack, pretending to be reading one while actually reading the next one.

To disguise the reason for opening the envelope, the typical variation used by mentalists has the audience members write questions on their cards, which the magician will answer. The magician then starts by making a statement like "I feel beautiful!", expresses some confusion about why he would say that, and then opens the envelope to read the question, "will the weather be nice tomorrow?" (while actually reading the next card, "what is my shoe size?"). As the questions may be impossible to guess, like a random person's shoe size, comedy or misdirection is often worked into the routine. For instance, "a size larger than last year" makes a reasonable answer to shoe size no matter who asks the question. Mediums may use the question and answer format as well, except that the questions are to be asked of the deceased, or perhaps are simply names of people to be contacted in the spirit world.

No comments:

Post a Comment