# A Lesson in Probability to a Twit

A Lesson in Probability to a Twit November 24, 2020

A long time commenter here (See Noevo Twit), who has contributed precisely nothing to any conversation he has been involved in bar annoyance, is not long for these boards. It finally came to a head and I announced he would be banned after the election given: a) if Trump won, he would be incorrigibly smug; b) if Biden won, we could laud it over him given his claims Trump would win in a landslide. After a short period of time of post-election analysis, I would then ban him. I have very lax commenting rules and get lots of hassle off regular commenters here for not banning people. I am trying to foster free speech and good discussion, but there is a limit and when someone consistently derails every conversation they are in and does not foster good discussion: for the good of the collective, they are ejected. He is a troll.

Yesterday, he produced his single longest comment. I will fix it interlinearly:

Johnny/Johnno,

This is another reason why he will be banned. For some time now he has for some unknown reason introduced each comment to me with this, I think as a veiled insult of just juvenile annoyance, and, if so, why would you do that to your host? Would you come round their house and just insult them, expecting to remain there, encouraged? I have explained he is spelling one of those incorrectly, but… He’s just weird.

Let’s look at the presidential election results from another angle, the angle which is your crème de la crème – probability.

For you have recently said that

JP: <<<Granted, this is my interpretation of truth (my truth about truth…). The problem for having a True Christianity (TC) is knowing that your TC is the TC. This is the case for any truth proposition. I defer to Descartes in that the only thing we can know is that you, the thinking entity (whatever that is) exists. Beyond that, it’s all probability. (emphasis added)
https://www.patheos.com/blo…>>>

[We’ll put aside the problem of your knowing probabilities when the only thing you admit to really knowing is your existence.]

So, in other words, you would rely on the truth claims of X if they have some (unstated) probability that you feel is higher than the (unstated) probability of not-X.

Not quite. And since this is his axiom, to get it wrong means that the rest of his post is arguably wrong. What does “rely on the truth claims of X” mean? What this means is that something that I assess as having a higher probability of happening has…a higher probability of happening, ceteris paribus (where all things include my/the most knowledgeable sources’ lack of knowledge of the variables). But, and this is a big but, this will not take into account a whole host of hidden variables. When all variables are known, there is no probability other than 100% unless a true (quantum) random chance is introduced.

Or:

“Under the cosmological hypothesis of determinism, there is no randomness in the universe, only unpredictability, since there is only one possible outcome to all events in the universe. A follower of the narrow frequency interpretation of probability could assert that no event can be said to have probability, since there is only one universal outcome. On the other hand, under the rival Bayesian interpretation of probability there is no objection to the use of probabilities in order to represent a lack of complete knowledge of the outcomes.”

Let’s consider some probabilities.

Would you say the probabilities of the following happening are high or low?

Oh, I’ve looked at them. What a deep well of disappointment I can feel has opened up in my psyche. What was I expecting?!

– That 2020 would be the first time since 1888 that a president seeking re-election would get more votes than in his first election (even over 10 million more votes), yet lose.

That 2020 would be the first time since … that a president seeking re-election would be outvoted by the margin he was by someone who was much more popular nationally, given how deeply unpopular he has been over his term and given his approval rating has never one (NEVER ONCE) sat above 50%, yet nincompoops like See Twit would still contest it. Answer? 1.

– That in 2020 a presidential candidate would get more votes than anyone in U.S. history (except for one shaky, senile 78-year old with 50 years of Swamp growth with a running mate who is the most liberal of any Senator), in fact, over 4 million more votes than the previous record holder (O in 2008), yet lose.

That in 2020, a presidential race would polarise people so much, and invite such distaste for the incumbent, that you would get the highest vote haul in history for both candidates, and that a shaky 78-year-old candidate (who can still run, walk down a slight incline unaided and hold a glass of water with one hand, and who already sounds more presidential and eloquent than elementary school Trump) would still win in an utter national landslide by over 6 million votes (helped by selecting one of the most liberal Senators for his running mate, now VP-elect), irrespective of how well the losing candidate did. Answer? 1.

– That nineteen (19) counties which went with the winner of the presidential race in ten consecutive elections (1980-2016) would in 2020 drop to one (1). That is, 18 of the 19 missed this year by going for the “loser”, Trump.

That the distaste for Trump was so pervasive (sexual assault, fraud, lack of professionalism, nepotism, running the US by Twitter, racism, anti-environment, etc etc), that though he consolidated his core base, he lost moderate Republicans and fiscal conservatives to the point that records were broken. Because, you know, records get broken all the time. Democrats win when they mobilise and get more people voting. Obama broke records and he won. Biden broke records and he won. [There are 3,007 counties in the US, you can slice and dice them statistically to get you all sorts of different conclusions.] Answer? 1.

To add, with no link to those counties, asking to assign a probability without any other information about the reference set is pointless. Knowledge is power, or at least, the greater the knowledge, the better the probability. As it improves, it gets closer to 1; when knowledge is complete, then the probability is 1, as above.

– That Indiana’s Vigo County would go with the winner sixteen (16) consecutive times, beginning in 1956, but in 2020, it went with the “loser”, Trump.

That Vigo County would return to its original predictive status of 1952, because we can all cherry-pick dates; that is would do what it has done twice before. Answer? 1.

I want to just dwell on this because it shows that See Twit hasn’t got a clue. Here is a quote:

<<<It’s a streak that puts the spotlight on our county every four years. Vigo County has only missed voting for the winning presidential candidate twice since 1888 and has experienced a perfect record since 1956. That’s 60 years, 15 straight presidential elections, and makes for the second-longest streak in the entire United States only trailing Valencia County in New Mexico.>>>

From which we could conclude that is really is rather overdue for getting it wrong. Indeed, given some of the knowledge we had about the election going into November 3, we would probably be in a good position to say that Vigo was more likely to get it wrong than right, given Indiana, but knowing the rest of the country. It had happened twice before, what’s to say it wouldn’t happen again since it was statistically overdue, depending on how you slice and dice it. There were a shed load of bellwether counties that got it wrong this year.  The question is why. If you are interested in an answer, then you wouldn’t be See Twit, who has the curiosity of a brick.

The thing is, typical bellwether county status is no longer fit for purpose, so any point See Twit makes about probability here is not taking into account the evolution of the US political landscape. Or, See Twit is out of date and ignorant of the variables. Here is an excerpt from a really good analysis as to why old bellwethers (focusing on Vigo here) are no longer bellwether at all, and by political scientists actually look to other places for more accurate predictive value. Well worth a read.

<<<Niven [political scientist] notes these bellwethers were born when political battle lines tended to be drawn more cleanly along economic lines. These middle-class communities were in the center and up for grabs. But as national politics become less about economics and more about culture wars and identity, Democrats have lost their grip in places such as Vigo County that are overwhelmingly white, he said.

Now the places emerging as possible new bellwethers have more racially diverse populations. Delaware’s Kent County last missed in 1992. Its population is 60% white and 27% Black. Blaine County in Montana, which last missed in 1988, is more than 50% Native American.

Vigo County doesn’t look much like America, and its place as its foremost presidential predictor relied on a certain degree of luck, said Matt Bergbower, a political scientist at Indiana State University. It is not as diverse as the nation, with a population that is 85% white. It is not as wealthy or highly educated, either.

But for generations its conservative tilt on social issues was balanced by left-leaning idiosyncrasies. There are four colleges in Terre Haute, a remarkable number for a city its size. It is the birthplace of Eugene V. Debs, a champion for workers’ rights who ran for president as a socialist five times in the early 20th century. The county’s blue-collar workforce was heavily organized and union halls dotted the city.>>>

See Twit would do well to look at the impact of deunionisation on the working class and how this is affecting the demographic makeup of the electorate (and, in turn, previously accurate bellwethers). My opinion is that, if the Democrats want to win back working-class rural/town votes, they need to address this (exactly the same for Labor in the UK).

– That the state of Ohio picked the presidential winner every election since 1896 except for three times – Dewey in 1944, Nixon in 1960, and the third time being Trump in 2020.

That the state of Ohio would fail to pick the election winner a third time since 1896 because the state is losing its bellwether status as it trends red due to an older, less educated population shift. Answer? 1.

C’mon, See Twit, it’s like, as a religious nut, you’ve given up even looking for answers! Who knew!

– That Florida, before 2020, picked the winner in six straight, and 13 of the last 14 starting in 1964. But in 2020 would go with the “loser”, Trump.

That Florida, due to massive (Spanish language) disinformation campaigns, huge rural success, massive successful grassroots Republican voter registration drives, consistent campaigning there in person by Trump, with almost half a million extra Hispanics in Florida than 2016 (who skewed independent, who skewed GOP), with Trump threatening to fire DeSantis if they didn’t win, Trump won in his adopted home state. Answer? 1.

…………….
Wouldn’t you say that the probability is very low of ALL those things happening?

No. No, I wouldn’t. See above, see evidence, see variables, see data, see reasoning, See Twit.

And conversely, that the probability is very high that the actual truth is that Trump in fact won, but for massive fraud?

No. No, not at all. The evidence for widespread fraud being the reason Biden won is so unbelievable tiny and the evidence so shockingly poor that the cases are being thrown out of court without going to trial, lawyers have fled the sinking ship, Trump has distanced himself from a conspiracy theorist nutjob (Powell) and has now started questioning his decision to have Sweatyhair-dyehandsdownpants Man in charge because it is making him look bad. I would qualify that to say it adds to the many reasons that already exist as to why he looks bad.

Conspiracy theories are pretty much the lowest probability explanation of a given event. See:

The last (journal) article finds:

<<<The results of all experiments support the hypothesis: The lower the probability of an event, the stronger participants embrace conspiratorial explanations. Conspiratorial thinking, we conclude, potentially represents a cognitive heuristic: A coping mechanism for uncertainty.>>>

What do you say, Mr. Probability?

I say that you are an ignoramus.

Tick tock.

Friday. Just long enough to see Trump deflate and implode a little more, then toodlepips. For the time being, run along.

[Apologies, I just needed some catharsis.]

Stay in touch! Like A Tippling Philosopher on Facebook:

A Tippling Philosopher

You can also buy me a cuppa. Or buy some of my awesome ATP merchandise! Please… It justifies me continuing to do this!

"Now that you can no longer ‮etallef‬ See, you've hooked up with Karenjasha. I hope ..."

The Nonsense in Spades We Are ..."
"And you are shooting the messenger as usualWhen the messenger is a liar and Nazi, ..."

The Nonsense in Spades We Are ..."