$30k challenge to Interpreter’s “Team Bayes”

[drumdude]

Kyler needs to take an introduction to logic class and learn what a tautology is. If you assume the conclusion in the premises of course the conclusion is true.

This is the bread and butter of apologetics, working backwards and trying to fool people into thinking you're drawing conclusions when you're actually just fuddling with premises and prior assumptions to see what you can get away with without being caught.

[Dr Moore]

It isn’t a red herring to ask for a demonstration of statistical independence.

Probability multiplication of the sort he and the others on Team Bayes are engaged in, requires proof of independence. Not special pleadings. Not made up analogies. Not “it can’t be done.”

His refusal to acknowledge this fundamental flaw exposes the bankrupt pornographic core of his math process. That each of the conditional probabilities he’s trying to multiply suffer from small sample size error and common causality in the mind of Joseph Smith. Statistical independence must be shown, or else it MUST be acknowledged by Team Bayes that every point probability suffers from errors of such high magnitudes that the “best honest answer” they can give us that the odds of Joseph authoring the Book of Mormon “might” range from incredibly small up to and including “he did it.”

One doesn’t have to read his arguments to know these flaws exist. This critique comes with the territory.

Now it’s possible to measure and account for sampling size error. Check the literature for dealing with sample sizes of 1 or 2, etc. But has anyone on Team Bayes done that? It’s possible to construct arguments that might outline ways of controlling for Joseph’s surroundings and his mind whilst considering point Bayesian conditionals - and then doing that analysis, collecting the data, showing the work. Have they done it? No and No. They have offered essentially zero controls at any point in the process. As Midgley would say, if he were at statistician, “utter rubbish.”

Kyler’s response is like the crank scientist who submits a long convoluted paper that describing a perpetual motion machine. The experienced physicist will know something is wrong without reading a single word. My challenge is a cousin to that — Team Bayes claims astronomical probabilities in favor of Joseph Smith’s prophetic success, but their statistical process deploys the most powerful tools of probability analysis without addressing the most basic questions of sample size error and independence.

To reiterate: without that rigorous treatment, it all amounts to garbage-in/garbage-out. Mopologetic porn.

But here is a dare. Don’t take my word for it. Walk over to the stats department at BYU and get 3 professorial opinions on a simple question: “did we treat this analysis properly regarding statistical independence?”

Have at it.

[Dr Moore]

Actually, let me take myself out of the equation in this $30k challenge. And make it even more good-faith, despite having already been misleadingly attacked for misleading the public by Kyler Rasmussen.

If Team Bayes will have their papers reviewed and given publicly-signed statements of “clean process” when it comes to the statistical treatments, each by 2 current professors of statistics at BYU (or a higher ranked university), I will consider that good enough to award the prize for each paper.

Clean process in terms of:
- Proper setup and evaluation of the Bayesian conditionals
- Properly addressing statistical independence of multiplied probabilities

Prediction: they won’t do it, or else they will thank me for helping them out of the Mopologetic porn industry.

[Doctor CamNC4Me]

Let's pretend for a second that it does make sense. Let's build an imaginary dataset of a couple hundred fictional books that deal with real-world cultures migrating to the other side of the world. For each of those books, I could estimate the likelihood of that culture's DNA being lost to time. I could then estimate the probability that each book would, on the basis of chance, demonstrate that culture's key literary forms. I could then conceivably be persuaded (probably at the point of gun) to calculate the correlation between those two variables.

Hopefully it's clear by now that this is a fool's errand …

You don’t say.

- Doc

[Dr Moore]

That’s funny doc. He’s starting to grasp why Team Bayes has produced nothing more than Mopologetic porn.

If you’re endeavoring a statistical analysis, but realize the underlying dataset is statistically invalid, what is the intellectually honest thing to do next?

[Dr Moore]

Let's pretend for a second that it does make sense. Let's build an imaginary dataset of a couple hundred fictional books that deal with real-world cultures migrating to the other side of the world. For each of those books, I could estimate the likelihood of that culture's DNA being lost to time. I could then estimate the probability that each book would, on the basis of chance, demonstrate that culture's key literary forms. I could then conceivably be persuaded (probably at the point of gun) to calculate the correlation between those two variables.

Hopefully it's clear by now that this is a fool's errand …

You don’t say.

- Doc

Actually it’s harder. He is trying to work with conditional probabilities that incorporate Joseph Smith’s bricolage into a statistically controlled framework. What Kyler lays out above is just one piece.

[Lem]

It isn’t a red herring to ask for a demonstration of statistical independence.

Absolutely not. What is a red herring, though, is the example KR gave to try to mock the idea of independence:

...In the majority of cases, since it would make no sense to try to calculate correlations on the basis of data (what's the process for calculating the correlation between the probability of me contracting COVID and the probability of Dr. Moore having a had a haircut on the last week?)

He starts with "in the majority of cases," followed by an example clearly chosen to be independent (which still fails, because, since the pandemic 1) closed down so many barbershops, the odds of 2)a bad haircut when trying to find a barber actually could be correlated, which shows how difficult it is to show independence).

KR, however, illustrates that he does NOT understand that an underlying variable, such as the pandemic closing businesses, could affect interdependence, with this statement:

Kyler Ray Rasmussen noel 16 hours ago

Upon further reflection, the haircut/COVID example is actually an interesting one, because I actually could calculate a correlation for that if I wanted to, because both of those variables could have a workable third variable to structure them around: time. I could try to calculate the probability of both me contracting COVID and the probability of a Moorian haircut for each week in 2021, and then correlate those respective variables.

https://www.patheos.com/blogs/danpeters ... 5443141768

No. The third variable is not time. His imagined correlation of multiple incidents of haircuts/covid contractions, measured per week, completely misses the point that it is the two variables and their relation to the third, i.e. the pandemic closing businesses, that causes the possible dependence.

Notwithstanding KR's very naïve example, there is an apparent preview of the data KR will be considering, posted in his discus comments. For example, look at 1, 2, and 3 here:

1. When you look at books written by first-time authors in Joseph Smith's era, it would be very unlikely for Joseph to write anything near as ambitious as the Book of Mormon, particularly given his relative lack of education. It would be even more unusual for him to write a book of that size and then to never write anything of that scale again, which is what happened.

2. All of the eyewitness testimony of the dictation process (even from Joseph's enemies) agrees that he dictated the entire book without any notes and without referencing what he'd said previously. It would be very unlikely for him to be able to do that while giving us a complex and coherent (and often profound) scriptural text, particularly since he never showed any indication that he had that kind of prodigious memory.

3. There were eight individuals who were able to see and handle the plates themselves, and three additional witnesses who claim to have seen an angel who testified on behalf of the Book of Mormon. Many of these individuals would later fall away from the church and turn against Joseph Smith, including all three of the witnesses who saw the angel. Those people had every reason and opportunity to recant and reveal Joseph as a fraud. But all of them continued to vehemently uphold their personal testimonies of the Book of Mormon until their dying day. Furthermore, Joseph did not have the means or materials available to create a convincing set of metal plates, inscribed with tens of thousands of complex characters. It's extremely unlikely that Joseph could have been able to maintain that kind of conspiracy or to somehow have hypnotized the witnesses into believing they saw what they claimed.

https://www.patheos.com/blogs/danpeters ... 5403176960

Independence between 1, 2, and 3, given the underlying hypotheses (for example: H=ancient document from angels bringing gold plates, vs. Not H=document from the mind of Smith without an angel bringing him gold plates), is just too unlikely to even consider. There is no "independence" of the variables.

[Dr Moore]

Right Lem. And the sample set he can work with, being so incredibly narrowly defined, means he also has to introduce statistical error methods in the analysis. If properly accounted for, error methods will show, in the math, that he cannot make any statistically valid conclusions from much of his “evidence.”

[Lem]

This statement of KR's will be key:

Or I could instead conclude that, since there's no logical reason why those things should be connected, it's reasonable to assume that they're independent. If someone ever points out a logical connection that I've missed, I'll be happy to reconsider.

https://www.patheos.com/blogs/danpeters ... 5443141768

Hmm. If I recall correctly, when it was pointed out to the Dales that by organizing their 131 probabilities into groups with attributes in common, they were implicitly acknowledging dependence, and therefore considering every single item to be independent of every other was illogical, I don't recall that they were happy to reconsider.

It's good to read that logical connections will be considered this time around. Not that there is any obligation, but since KR does read here, apparently, he could start by reconsidering the logical connection he missed with his haircut/covid example, since acknowledging the dependence completely changes the point he was using the example to make.

[Philo Sofee]

Kyler Rasmussen
There are cases where independence can and should be demonstrated quantitatively. Such is generally not the case for the historical Bayesian approach adopted by Carrier and others.

I want to see where in Carrier he gets this. I am skeptical this is Carrier's meaning wherever KR is getting this theme. I have Carrier, I wanna see page number and which book KR is using of Carrier. If he is thinking of Carrier's conditional probability of historical evidences, then this makes sense. In that case, since Carrier shows this really works very well with historical subjects, why is KR not going this route in his Book of Mormon evidences? I suspect it is because he knows he has to analyze all sides of the analysis of the evidences the apologists have given for the Book of Mormon, and the fact that every single one of those evidences without question have been soundly refuted by outside scholars. That is the Bayesian way Carrier is working with. So it is curious that he's throwing Carrier's name around, yet using a supposed Bayesian method Carrier doesn't use! And it appears that KR is not using a proper Bayesian method at all in the first place. We need some honest, open clarification of how Bayes is going to help you show the Book of Mormon is authentic KR. Quit using irrelevant examples of haircuts and get on to the Book of Mormon. Begin with how you have validated Nephites/Lamanites/Jaredites actually exist in order for the evidences to have any actual historic meaning.