The GoP War on the word “statistically”

Some times you just want to put your hand on somebody’s shoulder and say in a sad and weary voice: “Just stop, you are embarrassing yourself”.

Surrounding the current state of affairs in the US Presidential election are various Trump supporters (including some Trump supporters who claim they aren’t Trump supporters) crying “fraud”. The evidence they have produced is so thin that they are often retracting bits of it themselves but claiming that the shear volume of BS that they have produced demonstrates that the claims must be true.

Amid that is the word “statistically”. Here is an example. An article by “Scott Hounsell” in RedState. It has fewer of the more blatant attempts at fake fraud claims but instead attempts to throw a cloud of suspicion over the election. Note the use of the word “statistically”:

“Additionally, turnout in 2016, as a percentage of registered voters, was just 79.80%. This year, the state jumped to a statistically impossible 92.26%, a 12.46% increase over their 2016 numbers. In a state when Democrats statistically lost more voters than Republicans, we are supposed to believe that a 12% increase (largest ever) swung majority to Dems, by a factor that not only overcame the margin by which Trump won the state in 2016 but also erased any gains Republicans had in registration (by losing less) and gave Biden a 20,000 vote lead?”

Fair enough, that paragraph does have statistics in it but quite what the writer thinks the word “statistically” is supposed to add (other than BS) is unclear. And “statistically impossible”? What on Earth is “statistically impossible” supposed to mean? I was already getting annoyed with seeing “statistically improbable” being thrown around on dodgy Facebook pages but at least those words made some sense.

A figure by itself is just a figure. If we are talking about the probability of an actual piece of data then “statistically” i.e. within the methodology of the discipline of statistics, we are comparing that figure with some model of the figures — for example an existing distribution derived from a theoretical understanding of the figures and/or past experience. The US election figures are boringly unremarkable. Turn out is up but outcomes aren’t very different from 2016. That’s the context and that doesn’t point towards data being “statistically impossible” but if you are going to invoke the djinn of statistics then you invoke what comes with it. To claim a figure is “statistically improbable” then you need to show that STATISTICALLY i.e. show that when compared to some valid model that the figure would only very rarely appear UNLESS some key assumption about the model was violated (eg massive fraud). Simply saying “statistically” a lot doesn’t make a number dubious, that’s just magical thinking.

32 thoughts on “The GoP War on the word “statistically”

  1. Those “statistics,” which are all over the RW web, are faulty. The Wisconsin stat for this year is percentage of registered voters who turned out, while the previous years’ stats are percentages of the entire adult population who did. They’re not comparable.

    The QAnon people are extremely fond of the phrase “statistically impossible,” commonly expressing it as, “How many coincidences [have to occur] before [it is] statistically impossible?” I tried pointing out to a couple of them that coincidences occur constantly by comparing it to the number of events that had to go exactly as they did to produce me or them. By their definition, every human being is “statistically impossible.”

    But for many people, believing in a vast, evil conspiracy seems to be more reassuring than accepting that bad shit just happens.

    Liked by 1 person

      1. does math

        Okay, in my town, the million-to-one odds happen about every 4 days. If I take the metro area, it’s almost as often as NYC.


  2. What I’ve seen is that some states failed to account for same-day registration. That caused some precincts to appear to have > 100% turnout.

    When I worked for Jim Simons in New York, he told me that “statistically impossible” means more than three standard deviations from the mean because “one in one hundred can happen, but one in a thousand is impossible.” He was teasing me, of course, but there’s an element of truth to it: if you see a result that seems that unlikely, it’s worth examining it more closely to see if it’s not actually random.

    Liked by 1 person

  3. And as I’ve pointed out to folks, “One in a million chances have probably happened to over seven thousand people on the planet.”

    Frankly, most people are just crap with understanding statistics, especially as you get towards the tail end.

    Liked by 2 people

  4. Ha. Has anyone here considered the fact that the two popes in power during the 1960s were named John and Paul respectively. Now what were the names of the two leading members of the biggest pop group during that era? Talk about statistically impossible! (Maybe I should have written that in all caps).

    Liked by 2 people

    1. When Pope John Paul I announced his papal name, New York magazine wrote a tongue-in-cheek article saying he’d picked the name just because he wanted to cash in on the popularity of the Beatles and that he’d picked the easy name, leaving for his successor the more avant garde name “Pope George Ringo.”

      When John Paul I died after just a few weeks (what were the odds?), one of my officemates said, “Well, now we’ll see if he really does pick George Ringo.” Naturally “John Paul II” was a letdown by comparison.

      Liked by 2 people

      1. George is at least a saint name, so that should at least be a theoretical possibility. I think it would be more difficult for a pope do explain Ringo.


      2. “When John Paul I died after just a few weeks (what were the odds?)”
        Obviously it was statistically impossible.

        Liked by 1 person

      3. I think we all made that joke at the time. Even my mother thought of that. And it was indeed a letdown.

        Going by statistics — both 20th century papal and 20th century male European life expectancy –JPI should have lived many more years. Therefore, it’s impossible that he died so fast.

        So JPII didn’t happen, and neither did Pope Emeritus Palpatine, nor Frankie I.


        Liked by 1 person

  5. I also have to wonder for many of those pointing out all these fake conspiracy theories if they even know how the whole vote counting process works. Or do they know and just not care because they want to stir up shit. So far from all accounts none of any of the voting results or the process have been all that unusual or shocking. What we’re not used to is having something being so very close that it actually matters. Typically the race has already been determined so we ignore the rest of the process, but rest assured even in races that have already been called they still count every vote.

    Liked by 2 people

    1. Yes, I remember seeing the final results on all the county races we voted on back at the end of the Before Times. They trundled in and were certified sometime in late April, I think.

      One of the very close races took a week of counting before it could be settled.


    2. Yeah, the UK system of looking at the margin, looking at the number of votes left, and just quitting there unless the provisional votes (“pinks”) are greater than the margin (which they almost never are) is not applied in the US.

      Liked by 2 people

  6. “The Party told you to reject the evidence of your eyes and ears. It was their final, most essential command.”

    Liked by 2 people

  7. I feel obligated to note that the ‘o’ in ‘GOP’ is always capitalized, because it’s an acronym for ‘Grand Old Party’. (Much like Gladstone’s nickname “GOM”.)

    The Republicans are of course, the younger of the two major American parties, with the Democrats being the oldest organized political party continuously in existence in the world. But the nickname endures.


    1. I feel obligated to point out that an acronym is pronounced as a word (eg laser, or as nobody calls them, light amplified by the stimulated emission of radiation) and nobody says “gop”, so I would call this an initialism. If I’m being pedantic. Which I am.

      Of course if we did say “gop” it’s got a nice onomatopoeic quality to it, the sort of sound something slimy might make as it slops onto a hot a paving stone. Feels very fitting for now.


      1. I’ve seen that, and goopers as well. My own choice is Gops. He is a Gop. They are Gops. It is no longer an initialism. It doesn’t stand for anything, and in that way represents present-day Republicans. As has been noted, it also sounds a little bit gross. I capitalize it to say “Yes, it’s intentional. See! I am being respectful!”

        Liked by 1 person

  8. I’m not sure how widespread it is, but I’ve seen a few tweets (including one in my timeline from someone I thought knew better) claiming that vote counts in some urban precincts are rigged due to “Benford’s Law”, an observation of the pattern of digits in certain random distributions (such as a high occurrence of “1” as the first digit, and a somewhat lower but still high occurrence of “2” as the first digit). That’s been sometimes used to flag possible fraud in things like expense accounting, where the amounts spent on a wide variety of items tend to have random distributions over many orders of magnitude, when they’re not being manipulated.

    Folks who’ve only heard of the phenomenon but don’t know much about it are trying to apply it to cases where it doesn’t work. In particular, approximate precinct sizes are *not* random in most cases, but are usually designed to be within certain manageable and balanced ranges. So the first digits of their size will generally *not* follow a Benford-like distribution. And therefore, neither will vote totals that are a fairly reliable proportion of district size. (In many precincts in my city, for example, Democrats typically get 80-99% of the general election vote.) On the other hand, unpopular candidates whose vote totals are down in the statistical noise will tend to have a more random first digit more likely in aggregate to follow Benford-like patterns.

    Apparently the “Benford’s Law” argument for spurious “voter fraud” has been around long enough that there’s been at least one thesis dedicated to debunking it. This one was done 8 years ago:

    Liked by 1 person

    1. Yes, Larry Correia was pushing this for awhile. Actually he was pushing an even more nonsensical version in which somebody had been tallying the least significant digits and had found that “00” was the most common in some set of results. Of course SOME number has to be the most common, you aren’t going to get a tie, and whatever it is will be ‘unlikely’ …and anyway it didn’t turn out to be right. Larry then backpedalled.


  9. Unfortunately the GOP, or very large numbers of them, are impervious to reason. You can argue with them until you’re blue (or red) in the face, and it will make no difference. Like their leader, most Republicans think, “Which argument will get me what I want? That’s the one that’s clearly correct. And even if it’s not, Democrats are evil, so it’s ok to use it.”

    trump has made this a fine art. His brain must be the weirdest place: whatever he wants to believe is true, even if it contradicts another thing he wants to believe is true. So you get “Count all the votes!” in some states, and “Stop the count!” in others.

    Liked by 2 people

    1. Thanks to the Puppies, SF/F fans got a two-year preview of post-rational politics. It continues to baffle me that people can think that way–I keep thinking they don’t really believe anything they’re saying. But our preview didn’t help us find any solutions other than outvoting them and forcing our will on them.

      I wonder if anyone’s tried antipsychotic drugs on them?

      Liked by 3 people

Comments are closed.