MarginOfError

Members
  • Content Count

    6067
  • Joined

  • Last visited

  • Days Won

    15

MarginOfError last won the day on April 2

MarginOfError had the most liked content!

About MarginOfError

  • Rank
    Lord of the Gummi Snacks

Profile Information

  • Gender
    Not Telling
  • Religion
    So Mormon...You Don't Even Know.

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. MarginOfError

    Return to Church Guidelines

    I don't think that was meant in a "I object to mask wearing" kind of way. More like a "I would be physically uncomfortable wearing a mask for that long." I'm not looking forward to wearing a mask that long either. I will do it, but I won't pretend to enjoy it. Then again, depending on what my ward decides to do, I may just refuse to go. We have a few yahoos in our ward that are adamant this is all a conspiracy, and I can see them deliberately not wearing masks and touching every surface in sight just to make a point. Hopefully I'll be able to talk my bishopric into a moderate approach to reopening (one has already floated the idea that we could fully open right now if we wanted to).
  2. MarginOfError

    Return to Church Guidelines

    It is a deliberately vague set of guidelines. It wasn't intended to be a map of how to open church meetings, but a guide to work within local guidance and regulations. The letter I got had a little more guidance than what was in the newsroom. Essentially, if your local health guidelines limit you to less than 100 people, work within the phase 1 guidance. Otherwise work within phase 2. What each stake and ward actually implements will vary.
  3. MarginOfError

    Bad Citizens or Religious Liberty Guardians

    The appropriate response to those accusations is, "No, it is the fault of 1) the legislature for not extending the order, and 2) other idiots who misbehaved" Of course, to prove 2), you have to be willing to participate in contact tracing.
  4. MarginOfError

    Baptisms for the dead

    McConkie, Joseph Fielding Smith, and Joseph Smith Jr seemed to have been of a school of thought that between "restoration of all thing" and "God is unchanging", that any practice of worship that was given to the modern church must have existed in both ancient Christianity and Judaism. On this premise, they assume that if there is no evidence of such practice, it must have been lost from the record. Personally, I don't find that line of thought particularly convincing. I think it's perfectly reasonable for the concepts and principles to have existed throughout history, but the form and practice to have changed with culture and technology. It would seem unlikely to me that the Israelites were practicing baptism for the dead prior to Christ's death. They had some parallels, but nothing quite like baptism for the remission of sins. Instead, their equivalent to baptism was more along the lines of restoring their ability to enter the the temple and participate in rituals there. Under our understanding of the temple, that may not seem like much of a difference, but for them atonement and freedom from transgression was gained through the sacrifices at the temple. Baptism as a symbol of conversion didn't become popular until after the Babylonian captivity. As a symbol of repentance, I doubt it was new or revolutionary by the time of John the Baptist, as not even the Jewish leadership of the time really objected to it. So sometimes between the Babylonian captivity and John the Baptist, it had evolved into a recognized and accepted custom. (https://en.wikipedia.org/wiki/History_of_baptism) But under either premise, there really isn't reasonable way to get to ancient Israelites performed baptism for the dead.
  5. MarginOfError

    Click on this thread if you are an apostate

    Under the new handbooks, I actually fit the definition of apostate. Oh well.
  6. MarginOfError

    Bad Citizens or Religious Liberty Guardians

    I don't consider this an either/or proposition. You can be a guardian of religious liberty and a bad citizen at the same time. I actually agree with the court ruling. I don't object to executives having emergency powers that permit them to act rapidly. But those powers need to be limited and subject to review by legislatures. I think a 28 day review is prudent, and since the legislature didn't uphold the emergency declaration, it should be terminated. It would be nice if the federal government operated similarly (for most emergencies at the federal level, it seems the legislature has to override the declaration, or else it persists indefinitely). But just because I agree with the court ruling doesn't mean I think it's wise or responsible to meet in large groups.
  7. MarginOfError

    Interesting COVID-19 numbers from Southern Utah

    Let's clear up a few things. Estimates for how many people in the US contract the flu each year vary from 5% to 20%. (Source) That includes asymptomatic carriers. In real numbers, that is 16.4M to 65.6M US citizens per year. Flu kills 12,000 - 61,000 in the US each year (source). That puts the upper limit of total infection mortality at 0.093%. SARS-nCOV-2 (Covid) currently has 1.5M confirmed cases with 89,000 deaths (source) as of 16 May. @Plein Air's research indicates that the actual number of infections may be 20 to 50 times higher, or 30M to 75M cases. That yields a total infection mortality rate upper bound at 0.112%. That places the Covid total infection mortality upper bound at 1.2 times the flu upper bound. ------------------------------------ Another way to look at this is look at the number of symptomatic cases only. With the flu, the US sees between 3 - 11% of the population present with flu symptoms (source), or 9.8M to 36.1M cases per year. Covid, by comparison has 1.5 million cases so far (source) as of 16 May. These numbers put the upper bound of symptomatic mortality rate at 0.17% for flu and 6.93% for confirmed Covid. What we are missing here is a symptomatic Covid estimate (total - asymptomatic). Estimates for how many cases are asymptomatic range from 3% to 80% (source). Continuing to use worst case numbers, if 80% of 75M total Covid cases are asymptomatic, then 15M are symptomatic. This would put the upper bound of symptomatic Covid mortality at 0.593%. much lower than the confirmed case mortality, but 3.5 times higher than the flu symptomatic mortality rate. ------------------------------------------ But wait! How can the total infection mortality rates be so similar but the symptomatic infection rates be so different. That difference describes how Covid behaves differently than flu. Covid is more transmissible, more likely to lay dormant, but more likely to be severe when the immune system fails to suppress it. So, mortality isn't everything. Virulence also plays an important role. And in four months, Covid has infected more people than flu does in a year. Or a moderately more fair comparison, we've only been looking at Covid for the last half of flu season. So it's reasonable to expect Covid to infect twice as many people as the flu. Even at comparable total mortality rates, that is twice as many deaths. --------------------------------- Regarding the difference in response between swine flu and Covid, consider the following worldwide metrics: (source). Duration of pandemic: 20 months Total confirmed cases: 491,382 Total suspected cases: 1.4B (upper bound, 70M per month) Total confirmed deaths: 18,449 (922 per month) Total estimated deaths: 575,400 (upper bound, 28,770 per month) Now look at the worldwide statistics for Covid: (source, as of 16 May) Duration of pandemic: 6 months (including December 2019) Total confirmed cases: 4.64M Total suspected cases: 232 M (upper bound calculated as 50 times confirmed, 38.7M per month) Total confirmed deaths: 312,000 (52,000 per month) Total estimated deaths: too early to tell To be frank, swine flu wasn't as big a deal as Covid. It looks like swine flu may have been more virulent, nowhere near as deadly. By the numbers, it seems like the "lack of response" may have been appropriate.
  8. MarginOfError

    Creeping mediocrity masquerading as virtue

    I won't say that I love mediocrity, but it's okay.
  9. MarginOfError

    Kneeling to bless sacrament.

    Just out of curiosity, does anyone know where the "must kneel on two knees" thing comes from? My stake presidency started getting on our case because they would occasionally see missionaries in our ward kneel on one knee and wanted us to correct them. I always refused to do so because I've never seen anything that states two knees is a requirement. Seems like such an odd hill to die on. That being said, @Ironhold, barring anyone being able to provide documented evidence that two knees is the only way that comes with God's approval for performing this ordinance, I'd tell your father to get over himself (or, perhaps more tactfully ask him to show you where two knees is a requirement). Personally, I'm of the persuasion that if kneeling causes a person anything more than fleeting moderate discomfort, they are exempt from the expectation to kneel. My opinions don't carry a lot of weight in Church administration, but they are right!
  10. MarginOfError

    Hydroxychloroquine

    Did you randomize who gets to know you and who doesn't? Maybe we should do a crossover design
  11. MarginOfError

    Hydroxychloroquine

    I can't say I find much sense in this theory. Old people tend to suffer more from common maladies simply because their bodies are run down from having been alive for so long. I forget what the exact number is, but something like 80% of our elderly in the U.S. die within a few weeks of having a surgical procedure in a hospital. They just don't recover. Their deaths are typically classified as cardiovascular death. But that doesn't appear to account for the fact that the recovery time for elderly patients tends to be longer, harder, and riskier than it is for younger patients. When a respiratory disease come around, the elderly have always been the most effected because they tend to have more difficulty getting adequate exercise to have a strong enough respiratory system to withstand a disease. It sucks, but it is reality. The thing that is remarkable about SARS-COV-2 is not its mortality rate, nor its severity, nor its impact on the elderly population. The thing that is remarkable about this disease is its transmission rate. This thing spreads easily, and when combined with it's long incubation period, spreads fast. We're actually rather lucky it isn't a more severe disease or we'd really be hurting. (for flu vs SARS-COV-2 comparison, consider that flu kills about 60,000 each year in the U.S., while in the past two months, SARS-COV-2 has killed almost 50,000). If this thing had the even half the mortality rate among the 30 - 50 crowd that it does among the elderly, we'd be looking at Spanish Flu levels of death.
  12. MarginOfError

    Hydroxychloroquine

    Having only used males is actually a benefit to this study. If they only had 17 females available, it would have been difficult to extract any information of statistical value while adding another layer of noise. The thing that isn't stated clearly in the abstract is that the results should not be extrapolated to the female population. I'm not too concerned about the unrelated medical conditions. When performing the analysis, they adjusted the hazard ratios for "propensity score." What this means is they run the analysis in two models. First, they use all of the demographics and descriptors they can in a preliminary model to predict the probability of each person dying from factors unrelated to SARS-COV-2. Then they use that predicted probability as a moderating term in the primary model. While this methodology is known not to eliminate all of the bias associated with those demographics, it has been found to be successful in significantly reducing those biases*. So what's my take on this study? I wouldn't put too much weight on this one single study. Instead, you add it to the pile. Retrospective, observational studies were my bread and butter when I worked in the health field. They are rarely (never) conclusive on their own. But when studied in conjunction with other similar (or slightly dissimilar) studies, you can actually learn a fair amount. The big thing you look for in concordant and discordant results. If you have six or eight studies that study a similar outcome and have similar premises (they don't have to be identical, just similar) and they all show pretty similar results, then it typically ends up that a proper, randomized trial will confirm those results. In my experience (anecdotal), this will hold when the smaller studies are about 80% concordant. Less than 70% concordant, and you get into the "mixed results" range, which makes it harder to guess what the outcome of a clinical trial will be. Based on the number of studies I've heard of studying this drug and the mixture of results, I'd guess that the drug isn't likely to prove very effective. With enough research, we may find some marginal effect in some portion of the population that will benefit from the right dose. But it will take a pretty long time for us to be able to tease out effects with that kind of granularity. So it seems unlikely this is some kind of miracle drug. At best, it's a tool that can be used to some undetermined benefit when used carefully in the right circumstances that we don't yet understand. * some would argue that you should adjust for those factors directly. Well, it turns out these kinds of models can only support about one predictor variable per death, not per patient. So with only about 50 deaths total in the cohort, you get five terms, or degrees of freedom. With three study groups, you'll eat up two of them. So you've only got three to play with. Using the propensity score gives you the best balance of demographic adjustment while still having a model with good statistical characteristics.
  13. MarginOfError

    Price of Oil Drops Below Zero Per Barrel

    I might be too specific on the details, but I think the general concept is sound. Oil companies aren't able to store all of their supply and are having to pay people to take away the excess. That gets reflected in the stock price as a negative value. Probably a better description here: https://money.stackexchange.com/questions/124269/what-does-it-mean-for-the-price-of-oil-to-be-negative As far as the difference between paying someone to store or paying someone to take it away, I would imagine that in many cases, the people taking the oil away are planning to store it until the price of oil goes back up, at which point they may sell it back to oil companies. So, I may get paid $37 per barrel to take it away from Company A. I sit on it for six months, and when oil is back up to $50 per barrel, I sell it to Company A for $40 per barrel. I've made $77 per barrel, minus my storage costs. Yeah, what's happening right now is a little bizarre.
  14. MarginOfError

    Price of Oil Drops Below Zero Per Barrel

    My understanding is that the price went negative because storage facilities are near capacity. It goes something like Company A placed an order for a barrel of crude oil March to be delivered in April. Shelter in place orders take effect, demand for crude oil plummets Company A reduces refining. Storage capacity for crude oil is above expectation. That barrel of crude oil ordered in March finally arrives in April. Company A is at 100% storage capacity. Company A must either store the crude oil, or pay someone to store it for them So the negative price reflects the going rate to pay someone else to store that barrel of crude oil for you.
  15. MarginOfError

    Is religious liberty threatened in USA?

    huh. I was trying to be melodramatic and so over exaggerated that it would be obvious that I wasn't trying to name any specific entity. And then you had to go and pull up an an entity that might actually fit the description