Health Care, Medical Insurance, and some related stuff...

(A lot of this post was inspired by, and many ideas in it come from, Kevin Williamson's Wanderland newsletter dated 12/9/2024. It's available at https://thedispatch.com. I heartily encourage signing up for the newsletters there, and subscribing if you can, regardless of your political persuasion. They'll make you think, which is never a bad thing.)

OK, there has been a huge uptick in media (social and traditional) about insurance and health care, thanks to events in New York recently. I have some thoughts on the topics.

First, medical insurance is NOT health care. It isn't even "health insurance" because it certainly doesn't insure that you get or stay healthy.

Health care is eating properly, getting regular exercise, taking supplements and medication when needed, and seeing your doctor regularly for a checkup.

Medical insurance (like all insurance) is a financial instrument to manage risk. You pay an insurance company to absorb the risk of you needing medical intervention of some sort. To the extent they pay for aspects of health care as I've defined it here, that is them managing the risk you asked them to take for you. When they deny a claim, that is also them managing their risk (perhaps unfairly and stupidly, but more on that later.)

You have car insurance (at least the law requires you to have a minimum liability coverage of some sort if you operate a car.) Your car insurance doesn't insure that you have a working car at any particular moment, and it doesn't pay for regular maintenance. It doesn't pay for tune-ups, oil changes, alignment, tires, brake pads, and a host of other things that you need to do to have a reliable car. That's on you.

There are a lot of reasons why the services of a doctor are crazy expensive in America, but one of them is certainly that Americans, as a whole, are terrible about health care. We build communities that can't be walked in, we avoid physical work whenever possible, and we eat terribly. People who try to maintain their health are considered "fringe". That's a cultural issue I could write about at excessive length, but I won't rant here.

Going back to the car analogy, Americans essentially treat our bodies like cheap used cars: We drive them until the wheels fall off, then get another. Except we can't get another, so we end up paying a "mechanic" insane amounts to make the vehicle serviceable again. That's obviously stupid when put that way, but that's what we, as Americans, mostly do.

To the extent we take personal responsibility for health care, we reduce the risk that we're trying to pay someone else to cover, and reduce the cost of insurance. 

OK, back to insurance as a financial instrument for a bit.  Certainly there are problems with how we manage that risk transfer service, so let's talk about that. First, insurance is heavily regulated. That regulation is a problem, not the amount or type of regulation per se, but the fact that it regulated in 52 very different ways. Each state (and territory) regulates insurance a little differently, which is a knock-on effect of other aspects of the Constitution. That network of differing regulation drives up insurance costs in two ways:

• It makes meeting regulatory requirements difficult and expensive, by imposing a lot of duplication of effort.
• It makes the pool of insured people a lot smaller, since every plan is state-specific, and so the risks are state-specific, which means the pool of insured people becomes state-specific. The smaller a pool of insured people, the fewer people that risk is statistically spread across, and the more uncertainty about the risk.

Both of those could be mitigated, without Federal action, by States developing compacts in which they regulated insurers the same way, using the same forms, rules and procedures. (Hot tip, we do something like that currently with the companies that own and operate the electric transmission system, pardon the inherent pun.)

A lot of people advocate for "single payer" or "medicare for all". In fact, only a few other countries (the UK and the countries that were once their territories) do that. The other developed nations that people point to as having better health care and lower costs than the US (a) have populations that generally take care of themselves better, and (b) have a single, very stringent, regulation system for the whole country. Look at the Swiss, for example.  I would point you to Kevin Williamson's discussion on his Wanderland newsletter.

Now, I had promised that I would talk about unfair and inappropriate denial of coverage, which clearly happens. I would venture that most of us have experienced it, or know someone who has. I have two amazing and related hacks for y'all.

THE FIRST HACK: One of the things that is pretty consistent across the US is your right to be given the following information on request:

1. Name, License Number and Board Specialty of the physician who make the decision that treatment was not necessary.
2. Copies of all materials used to make that determination.
3. Proof that the physician is current with their registration in your state and up to date on all their continuing education.
4. The aggregate rate at which that physician approves or denies similar treatments for peer review.

Most insurance companies will just approve the claim, rather than admit that the decision was made by a physician who is not licensed in your state, not current on their training, or practicing medicine outside of their specialty (or is just someone with a GED and a checklist, or an AI system, or...)

THE SECOND HACK: ProPublica has a website that will generate an appropriate letter to make that request, based on information that you enter. Bookmark this: 

https://projects.propublica.org/claimfile/

Figures don't lie, but...

 I want to compare two charts, one is famous, the other is not, yet...

First, let's look at the chart that Donald Trump was pointing to when someone took a shot at him. This is a little fuzzy, because it's from a screenshot.

The first thing I want you to notice is that the chart has a big arrow at the bottom pointing to the end of Donald Trump's admnistration, marked "TRUMP LEAVES OFFICE". Note where it is pointing on the time scale, around March of 2020.

Except Trump didn't leave office in March of 2020. (If he did, then who was acting as President?) He left office on January 20, 2021. March of 2020 was 7 months before the election, and almost 10 months before Trump left office.

 That's odd.  You would think that if there was any date that the Trump campaign would get right, it would be the date he left office.

The second thing I want you to notice is the date marked "TRUMP Tariff threat leads to Mexican cooperation."  That's May of 2019.

Let's look at the second chart, the less famous one:

Yeah, this one's also a screenshot, so it's fuzzy too.  I'll update with clearer images when I can source them. This chart does two things that the first one doesn't:

• Corrects the date when Trump left office to January 2021.
• Adds data through September of 2024 (so far.)

Notice the rise in illegal immigration since the low point the original chart called the end of the Trump administration. That's a roughly 85,000 person increase in the monthly rate of illegal immigration from that low point. Even if you attribute some of that increase to the November election, and Biden becoming President-Elect, most of it happened before November, and the rate of increase tapered off in November.

So the decline in Illegal immigration started around May of 2019 and reversed around March of 2020. It's possible that Trump's actions and policies caused that dip, but those policies were still in force during the rise in most of 2020. Could there be another cause at work?

How about a global pandemic? It may be coincidence, but the decline started around the time that lockdowns began, and reversed around the time that vaccines became widely available in the US. Latin America was about a year behind the US in vaccine availability and the rate of vaccination.

Could the decline in illegal immigration have been a pandemic response, and the rise in illegal immigration late in Trump's return in fact a return to what the rate of illegal immigration have been without the pandemic? Could the Trump campaign be knowingly taking credit for something the Trump administration had nothing to do with? Stranger things have been done and said in politics.

Now let's look at the decline late in the Biden administration. What could have caused that?  Well, in 2023, in response to the surge in immigration, the Biden administration got Mexico to station troops to stem immigration into the US. It worked.

By the way, that cooperation was established in a 2022 "Summit of the Americas" in Los Angeles. A Declaration from that Summit (released June 10, 2022) did the following:

• Got Costa Rica, Colombia and Ecuador to offer legal immigrant status for Venezuelans crossing their borders. The situation in Venezuela drove a lot of immigration to the US, that is now being stopped in other countries.
• Got Mexico, Belize and Costa Rica to place tighter restrictions on Venezuelans flying into their countries.
  

 It was the Biden administration that got Mexico to pay for a wall, by stationing Mexican troops to enforce the border from their side. It was the Biden administration that got other countries to create a series of smaller "walls" to do what the Trump administration promised, but failed, to do.

Yes, a lot of people get into this country bypassing the legal process.  Why is that? Well, the legal system is swamped, so thousands of immigrants a day are crossing the border, asking for asylum, and getting briefly detained and, if there isn't an obvious issue, released, bypassing the Asylum process.

Why is the legal system swamped? Well Congress failed to act. Constitutionally, it is Congress' job to write laws and fund the enforcement process. The Congress had a bipartisan fix, or the start of one, before them last year.  It didn't pass, because the Republican support for it pulled out at the last minute.

That kept illegal immigration a live topic for this year's election. There was a way to start resolving the issue, and the Republican party chose politics over signing off on legislation that they co-wrote. They did the political thing, rather than the right thing.

You can build a wall, or you can work on better solutions. Take your pick.

In simple fact, the immigration situation is far more complex than a simple graph is ever going to show. 

Especially when that graph has carefully selected wrong data.

Figures don't lie, but they sure can be manipulated to sell a false narrative.

The 2020 Election: Was there a "Steal" to Stop?

OK, I generally try to not stray too deep into the miasma that is politics these days, but there is something that is worrying me. I’m seeing people, bright, intelligent people whom I respect, still buying into the “voter fraud” argument. If you’re one of those people: I hate to say this, but I truly do love you, and love sometimes has to speak the truth:

Y’all have been gaslighted. I don’t say that lightly, and I’m trying not to be mean about it, but that’s the only thing I can find to explain the phenomena.

I have two different explanations for why the 2020 Election couldn’t have been stolen: One Logistic and one Statistical. We’ll take the Logistic first.

The Logistics problem begins with this simple fact: Every voting precinct has a different slate of candidates, offices, and issues. Not every state, or every county, every voting precinct. For example, in my precinct, I vote for a different school board than the one for the city I live in, because I’m part of that school district. The school district doesn’t follow city boundaries.

Each voting precinct has a different ballot, which means that the encoded data will be different. Each precinct’s mail in ballots will be different as well. Wikipedia reports that in 2020, there were 176,933 precincts in the United States. That doesn’t have to be an exact number, though. Managing to fraudulently manipulate the results from anything over a few thousand different voting precincts rapidly becomes a logistical nightmare. Anything over 100,000 precincts in a 24 hour period? Logistically impossible. What about with a computer? Well, you would have to find a way to program a computer to invisibly manipulate different ballots, captured by different means, all within 24 hours.

What about AI? Well; (1) AI was not that advanced in 2020, and (2) AI requires a huge sample of text and images (multiple millions of samples) to be able to even recognize a ballot reliably, let alone decipher hundreds of thousands of different forms and formats, and successfully manipulate them. I'm pretty sure that someone would have spotted all those samples being accumulated.

Now, on to Statistics: I’m deep down a “numbers and logic” guy, so let’s take a logical look at some numbers. I’m using two sources of information: A 20 year MIT database of observed voter fraud as of 2020, and a table of 2020 state-by-state vote counts from Dave Lep’s Atlas of US Presidential Elections.

First, the MIT database shows, over the past 20 years, just over 1,200 cases of voter fraud, 204 of which involved mail-in ballots. That is a rate of voter fraud of 0.00006%.

Now let’s look at Dave Let’s Atlas. I didn’t pick this site because of any particular political bent, it just happened to be the first one I found that had a handy table. First, let’s look at the popular vote. The margin of popular vote victory for the Biden/Harris ticket is 3,078,287 votes. In order to make it a “dead heat”, half of those (1,539,543) would have to be fraudulent. There were 158,590,015 total votes cast across all 50 states, so a “dead heat” popular-vote election would require that there was a fraud rate of just under 1%

That is 16,000 times the rate of voter fraud observed at MIT, just to get to a dead heat. A victory for Trump/Pence by the same margin would require a fraud rate of 1.94% of all votes cast, about 32,000 times what MIT observes.

Ah, but a few key states for the Electoral College were close! What about those? Basically, we’re talking here about Arizona (11), North Carolina (15) and Georgia (16). For the sake of discussion, let’s assume that all 3 actually should have gone to Trump/Pence. The next closest state, by a “% of fraudulent votes to make it a dead heat” measure is Wisconsin, with 0.31%. That is still 5,000 times the MIT observed rate. Let’s assume that Wisconsin should have gone to Trump/Pence by the same margin (10,000 times MIT’s rate of vote fraud). Now you’re at a dead heat in the Electoral College at 269 each.

That’s what you have to assume to imagine that voter fraud made it a dead heat. 10,000 times the observed rate of voter fraud.

Not a clear victory for Trump/Pence, just a dead heat in the Electoral College. You have to assume that every one of the 4 closest states to a dead heat should have gone to Trump/Pence.

For a clear victory (by the same vote margin) you would have to throw in Pennsylvania (the next closest state after Wisconsin) with an assumed a rate of fraud in of 12,000 times the MIT observation.

To be clear: To buy the argument that voter fraud manipulated a loss for Trump/Pence, you have to assume that either; (1) a campaign to affect the vote through fraud at 12,000 times the observed rate of fraud has so far gone totally undetected by the people responsible in both parties, or (2) MIT’s data only reflects unsuccessful fraud, and the successful fraud isn’t caught. (That’s called “ascertainment bias”, where you don’t know reality because your sample size misses too much.)

Dealing with the first point, some may argue that it is all a product of a massive conspiracy to cover up the fraud. In response, I would point to Chuck Colson’s comment about his belief in the Resurrection:

“I know the resurrection is a fact, and Watergate proved it to me. How? Because 12 men testified they had seen Jesus raised from the dead, then they proclaimed that truth for 40 years, never once denying it. Every one was beaten, tortured, stoned and put in prison. They would not have endured that if it weren't true. Watergate embroiled 12 of the most powerful men in the world-and they couldn't keep a lie for three weeks. You're telling me 12 apostles could keep a lie for 40 years? Absolutely impossible.”

Regardless of how you may feel about the Resurrection, Mr. Colson’s point about conspiracy is valid: The Watergate conspirators couldn’t collectively keep a secret for 3 weeks. The more collaborators there are in a conspiracy, the more possibility that someone would break, especially when their careers and even their lives are being threatened. A cover-up on this scale would have to involve thousands of people, at many levels, any of whom would be able to bring forward evidence to prove the conspiracy, any time in the past almost 4 years. Nobody has come forward, nobody has confessed, despite the threats, and despite the opportunity to be a hero by saving some court cases.

Didn’t happen. That dog don’t hunt.

Now, let’s think through that second possibility for a moment, because “Survivorship bias” and Ascertainment bias both happen. MIT’s data when I referenced it covered 20 years of observations (1999 – 2019)*.

The assumption that the data is that massively flawed, and that the real rate of fraud is underreported by a factor of 12,000, calls into question every election in the past 20 years, regardless of who won.

So, if you want to believe that Trump’s loss in 2020 was a result of undiscovered voter fraud, you would have to accept that his victory in 2016 was a fraud as well.

Either; (a) there has been massive fraud for the past 20 years, that the MIT survey doesn’t pick up on, or (b) the 2020 election had a rate of voter fraud that exceeded the historical precedent by a factor of 12,000, and that with all the tens of thousands of people responsible (see my logistical argument) either missed it, or lied, and nobody has ‘fessed up.

That’s why I’m saying that, if you, bright person whom I respect, buy the “voter fraud” argument about the 2020 Election, you have been successfully gaslighted.

The numbers just don’t work. The logic is faulty. I’m sorry to say it, but your perception of reality has been distorted.

* I originally wrote this in late 2020, and have only minimally updated it since.

An argument for Intelligent Design from an surprising source

Neil Degrasse Tyson just posted an incredible argument for Intelligent Design on TikTok. 

He didn’t think it was an argument for Intelligent Design, but it is.

Here’s how it goes, he starts talking about shuffling playing cards, and gets to the fact that there are 52! (52 factorial) different possible sequences of cards after a proper shuffle. For those who don’t remember that part of math, 52! is:

52 x 51 x 50 x 49 x 48 x 47 x 46 x 45 x 44 x … x10 x 9 x 8 x 7 x 6 x 5 x 4 x 3 x 2 x 1

which works out to 8 x 10⁶⁷. 8 with 67 zeroes after it. A really big number.

To put that into context, he explains that if you took a trillion people, in each of a trillion civilizations, in a trillion different universes, handed each a deck of playing cards, and gave each person instructions to shuffle their deck a trillion times a second, and had that go on for a trillion years, you would only have a 40% chance of one of those shufflings giving the same order as a deck you just shuffled.

So how is that an argument for Intelligent Design?

How old does Science say the Universe is? The longest time that I have found in scientific research is 26.7 Billion years.

Let’s assume for the sake of discussion that the development of intelligent life only requires 52 steps, in the correct order. Each step is represented in our thought experiment by a playing card. Only one sequence out of all the possible sequences results in life as we know it. Neil argues that hitting the correct sequence would take;

• a trillion universes,
• times a trillion civilizations,
• times a trillion people,
• times a trillion shuffles per second,
• times 31,536,000 seconds per year,
• for a trillion years,
• with only a 40% chance of success at hitting the correct sequence.
  

Here we are having this (presumably) intelligent conversation in (at most) 2.67% of the time, in just one universe!

The counter argument is that this isn’t the only cycle the universe (or multiverse) has taken.

If all that is, is on infinite repeat, anything could happen.

Even a famous atheist accidentally posting an argument for Intelligent Design on TikTok.