This discussion takes a critical dive into Bayesian versus frequentist probabilities, questioning Scott Alexander's arguments and their implications for communication. The hosts explore the intricacies of probability modeling and the pitfalls of overconfidence in probabilistic reasoning, especially in high-stakes situations. They also tackle controversial therapies, reflecting on societal views toward individuals with pedophilic attractions. Additionally, the nuances of super forecasting and the challenges within AI discussions reveal the complexities surrounding probability interpretation.
The hosts debate Bayesian statistics versus frequentism, underscoring the importance of understanding probability as a tool, not a definitive measure.
A discussion on mental health care reveals significant regional disparities, emphasizing the necessity for improved systems and advocacy for better support.
Interactions with 'micro-celebrities' highlight the value of networking and collaboration in generating new insights within niche communities.
Skepticism around the effectiveness of super forecasting challenges its perceived accuracy, raising questions about the reliance on subjective probability assessments.
Deep dives
The Struggle with Video Transition
There is a discussion about the challenges of moving from audio podcasting to video content. One speaker expresses skepticism about the potential success of this transition, suggesting that their audience may not have the same enthusiasm for video. The other speaker insists on the need for listeners to engage with their content on platforms like YouTube to boost visibility and support. The exchange highlights a tension between differing views on how to reach and connect with their audience.
Networking at Vibe Fest
The hosts share their experiences from a recent festival where they interacted with several 'micro-celebrities,' including individuals with niche audiences. The atmosphere of this gathering fostered interesting conversations among attendees, highlighting the spirit of collaboration and shared learning. One notable interaction involved a discussion about mental health care practices, emphasizing how certain regions have significantly better systems than others. Such conversations point to the importance of networking with diverse individuals who challenge perspectives and share valuable insights.
Institutionalization and Mental Health Care
A conversation unfolds about the stark differences in mental health care experiences in various locations, specifically contrasting regions in the U.S. with British Columbia. One individual recounts their experience being institutionalized in a supportive environment, which contrasts sharply with the more commonly depicted harsh realities. The importance of understanding and advocating for better mental health systems is underscored, especially as one guest shares a personal story of loss related to inadequate mental health support. This exchange calls into question the adequacy of current approaches to mental health issues across different jurisdictions.
Dating a Therapist for Pedophiles
One host shares a surprising encounter on a date with a therapist specializing in working with individuals convicted of sexual offenses. This revelation leads to broader discussions about the peculiar nature of such therapy work and societal perceptions surrounding it. The dynamics of therapy, especially involving controversial subjects like pedophilia, prompt a deeper consideration of ethical boundaries. This conversation illustrates the complex interplay between personal experiences and societal issues, shedding light on a rarely discussed area within mental health care.
Debate on Bayesian Statistics
A significant portion of the podcast is dedicated to discussing Bayesian statistics and its comparison with other probability interpretations. One host critiques the reliance on Bayesian methods, arguing for a more nuanced understanding of probability as merely a tool rather than a definitive measure. The conversation highlights the pitfalls of jargon in statistical debates, particularly when discussing subjective probabilities versus objective data. This discourse is reflective of broader conversations within the academic community about the efficacy and applications of differing statistical methodologies.
The Challenges of Super Forecasting
The podcast delves into the relevance and limitations of super forecasting, evaluating its credibility in predicting future events. One host expresses skepticism about the perceived accuracy of super forecasters, suggesting that the numbers they produce are more about their opinions than grounded empirical evidence. This skepticism raises questions about the validity of super forecasting as a practice, especially regarding high-stakes predictions in areas like AI development and policy. The discussion underscores the need for rigorous scrutiny of forecasting methods and the implications of placing heavy reliance on subjective probability assessments.
Comparison with Historical Predictions
The conversation touches upon how historical predictions, particularly in forecasting contexts, can inform current decision-making practices. The hosts critically evaluate the predictive capabilities of experts and the historical context surrounding high-profile predictions, particularly around technology and regulations. They note the discrepancies between expert opinions on issues like AI and the actual outcomes of past predictions, emphasizing the importance of critical thinking in evaluating forecasts. This comparative analysis calls for a more grounded understanding of the limitations that expert predictions can impose on public discourse.
After four episodes spent fawning over Scott Alexander's "Non-libertarian FAQ", we turn around and attack the good man instead. In this episode we respond to Scott's piece "In Continued Defense of Non-Frequentist Probabilities", and respond to each of his five arguments defending Bayesian probability. Like moths to a flame, we apparently cannot let the probability subject slide, sorry people. But the good news is that before getting there, you get to here about some therapists and pedophiles (therapeutic pedophelia?). What's the probability that Scott changes his mind based on this episode?
We discuss
Why we're not defending frequentism as a philosophy
The Bayesian interpretation of probability
The importance of being explicit about assumptions
Why it's insane to think that 50% should mean both "equally likely" and "I have no effing idea".
Why Scott's interpretation of probability is crippling our ability to communicate
How super are Superforecasters?
Marginal versus conditional guarantees (this is exactly as boring as it sounds)
How to pronounce Samotsvety and are they Italian or Eastern European or what?
During the pandemic, Dominic Cummings said some of the most useful stuff that he received and circulated in the British government was not forecasting. It was qualitative information explaining the general model of what’s going on, which enabled decision-makers to think more clearly about their options for action and the likely consequences. If you’re worried about a new disease outbreak, you don’t just want a percentage probability estimate about future case numbers, you want an explanation of how the virus is likely to spread, what you can do about it, how you can prevent it.
- Michael Story
Is it bad that one term can mean both perfect information (as in 1) and total lack of information (as in 3)? No. This is no different from how we discuss things when we’re not using probability.
Do vaccines cause autism? No. Does drinking monkey blood cause autism? Also no. My evidence on the vaccines question is dozens of excellent studies, conducted so effectively that we’re as sure about this as we are about anything in biology. My evidence on the monkey blood question is that nobody’s ever proposed this and it would be weird if it were true. Still, it’s perfectly fine to say the single-word answer “no” to both of them to describe where I currently stand. If someone wants to know how much evidence/certainty is behind my “no”, they can ask, and I’ll tell them.
- SA, Section 2
Socials
Follow us on Twitter at @IncrementsPod, @BennyChugg, @VadenMasrani
Come join our discord server! DM us on twitter or send us an email to get a supersecret link
Help us calibrate our credences and get exclusive bonus content by becoming a patreon subscriber here. Or give us one-time cash donations to help cover our lack of cash donations here.
