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?

References

• In Continued Defense Of Non-Frequentist Probabilities
• Article on superforecasting by Gavin Leech and Misha Yugadin
• Essay by Michael Story on superforecasting
• Existential risk tournament: Superforecasters vs AI doomers and Ben's blogpost about it
• The Good Judgment Project

Quotes

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.
• Click dem like buttons on youtube

What's your credence in Bayesianism? Tell us over at incrementspodcast@gmail.com.

Support Increments

So what's the deal? Is it possible to "harness the wisdom of the crowd to forecast world events"? Or is the whole thing just a result of sloppy statistics? We believe the latter is likely to be true with probability 64.9% - no, wait, 66.1%.

Intro segment:

"The Sentience Debate": The moral value of shrimps, insects, and oysters

Relevant timestamps:

• 10:05: "Even if there's only a one in one hundred chance, or one in one thousand chance, that insects are sentient given current information, and if we're killing trillions or quadrillions of insects in ways that are preventable or avoidable or that we can in various ways mitigate that harm... then we should consider that possibility."
• 25:47: "If you're all going to work on pain in invertebrates, I pity you in many respects... In my previous work, I was used to running experiments and getting a clear answer, and I could say what these animals do and what they don't do. But when I started to think about what they might be feeling, you meet this frustration, that after maybe about 15 years of research, if someone asks me do they feel pain, my answer is 'maybe'... a strong 'maybe'... you cannot discount the possibility."
• 46:47: "It is not 100% clear to me that plants are non sentient. I do think that animals including insects are much more likely to be sentient than plants are, but I would not have a credence of zero that plants are sentient."
• 1:01:59: "So the hard problem I would like to ask the panel is: If you were to compare the moral weight of one ant to the moral weight of one human, what ratio would you put? How much more is a human worth than an ant? 100:1? 1000:1? 10:1? Or maybe 1:1? ... Let's start with Jamie."

Main References:

• Superforecasting: The Art and Science of Prediction - Wikipedia
• How Policymakers Can Improve Crisis Planning
• The Good Judgment Project - Wikipedia
• Expert Political Judgment: How Good Is It? How Can We Know?: Tetlock, Philip E.: 9780691128719: Books - Amazon.ca

Additional references mentioned in the episode:

• The Drunkard's Walk: How Randomness Rules Our Lives
• The Black Swan: The Impact of the Highly Improbable - Wikipedia
• Book Review: Superforecasting | Slate Star Codex
• Pandemic Uncovers the Limitations of Superforecasting – We Are Not Saved
• My Final Case Against Superforecasting (with criticisms considered, objections noted, and assumptions buttressed) – We Are Not Saved

Use your Good Judgement and send us email at incrementspodcast@gmail.com.

Support Increments