Even as I switched to a diet of mostly soylent to save money and attempted to adopt an extremely aggressive update schedule for this blog, I was slowly making myself more and more miserable and gaslighting myself about my own emotions.
Be warned: Soylent comes with a trap. It makes total calorie intake extremely legible, which can overshadow peoples’ instinctive eating, causing them to undereat, which causes depression. In particular: for most people, 2000 calories per day is a shortfall. The standard answer to how much one should eat, if not attempting weight loss, is to use the Harris-Benedict formula. A mostly-Soylent diet is also a low-sodium diet, which I believe is also something to look at with skepticism. Whenever mental health issues appear close in time to questionable dietary changes, I believe the diet should be the first suspect, regardless of how compelling other narrative explanations might seem.
I actually ran directly into this after I’d been on soylent for about a month and a half. I found myself feeling consistently awful in a way that had slowly built over time and when I bought myself something to eat that wasn’t soylent I felt so much immensely better I just started crying in relief and from that I pretty much immediately knew I had done something to mess up my diet. I backed off the soylent pretty substantially after that.
Okay, I’ve heard you say “for most people, 2000 calories per day is a shortfall” and other nutrition-related claims like a hundred times (probably literally if secondhand “Jim says X about nutrition” counts), but unless https://www.lesswrong.com/tag/nutrition is seriously failing me, you’ve never written about why you believe any of that. Which seems especially bad to me in a domain where most information is untrustworthy, and we all know that it’s untrustworthy but most of us don’t know how to figure what is true. I feel like I’m just supposed to believe you because you’ve said it so many times with so much conviction for so many years. Can you write a post or something?
I agree that I really need to write a post on this (and a half-dozen other nutrition-related things that I’ve become known for saying in person and in comments, but haven’t written up properly).
For this specific point—about 2000 calories per day being a shortfall—it’s fairly straightforward. There’s a scientific-consensus answer to how many calories people typically need, the Harris-Benedict formula. (This has a few variations with slightly different constants, all of which trace back to linear-regressions on measured energy expenditure.) I typed in my own parameters, and those of a few other people, and observed that, within the demographics of people I know, ie mostly youngish male and with a height reflective of decent childhood nutrition, it was consistently well above 2000. I also traced the history of where the 2000 number used on nutrition facts panels came from, and found that it was never intended as any sort of recommendation, and seemed to have been misinterpreted as one by historical accident. I try to always include a link to the formula or a calculator for it, when I talk about calorie intakes, so that people can get the real number for themselves.
There’s a scientific-consensus answer to how many calories people typically need, the Harris-Benedict formula.
Out of interest, I checked wikipedia, which gives three different equations (Harris-Benedict, Mifflin St Jeor, and Katch-McArdle), saying “Historically, the most notable formula was the Harris–Benedict equation”.
I just put in my details on them, and my daily caloric burn varied over 1000 calories (one was 500 less than the other which was 500 less than the other). One of them was pretty close to 2000. /shrug
Isn’t the link to the Harris-Benedict formula sufficient evidence? Like, that’s the actual formula nutritionists use. I don’t know any specific reason for using the 2000 calories per day target other than that it’s a nice round number.
I mean, not really? Everyone has told me that ‘nutritionists do/believe it’ is not a good reason to do/believe something. I’m also not saying that I stand behind 2000 calories per day; I’m saying that Jim says a lot of things about nutrition all the time and I want to know why.
Are you sure about the Harris-Benedict formula? It seems like Mifflin-St Jeor is the most reliable. Nonetheless, I’m curious if you have any recommended articles/books on diet and mental health?
https://en.wikipedia.org/wiki/Harris%E2%80%93Benedict_equation lists both, and seems to consider the Mifflin-St. Jeor version a revision that sort of uses the same name. The calculator I linked uses constants from Roza-Shizgal, which is technically neither of these. All of the sets of constants were produced by taking linear regressions on measured energy expenditure among different groups of people. I don’t have much in the way of opinions about which of these sets of constants is best; the differences between them are small compared to the difference between “2000 is what it says on the nutrition facts panel” vs actually-using-any-sort-of-formula, and compared to the error in the activity-factor constant and the measurement error in a non-Soylent diet.
Be warned: Soylent comes with a trap. It makes total calorie intake extremely legible, which can overshadow peoples’ instinctive eating, causing them to undereat, which causes depression. In particular: for most people, 2000 calories per day is a shortfall. The standard answer to how much one should eat, if not attempting weight loss, is to use the Harris-Benedict formula. A mostly-Soylent diet is also a low-sodium diet, which I believe is also something to look at with skepticism. Whenever mental health issues appear close in time to questionable dietary changes, I believe the diet should be the first suspect, regardless of how compelling other narrative explanations might seem.
I actually ran directly into this after I’d been on soylent for about a month and a half. I found myself feeling consistently awful in a way that had slowly built over time and when I bought myself something to eat that wasn’t soylent I felt so much immensely better I just started crying in relief and from that I pretty much immediately knew I had done something to mess up my diet. I backed off the soylent pretty substantially after that.
Okay, I’ve heard you say “for most people, 2000 calories per day is a shortfall” and other nutrition-related claims like a hundred times (probably literally if secondhand “Jim says X about nutrition” counts), but unless https://www.lesswrong.com/tag/nutrition is seriously failing me, you’ve never written about why you believe any of that. Which seems especially bad to me in a domain where most information is untrustworthy, and we all know that it’s untrustworthy but most of us don’t know how to figure what is true. I feel like I’m just supposed to believe you because you’ve said it so many times with so much conviction for so many years. Can you write a post or something?
I agree that I really need to write a post on this (and a half-dozen other nutrition-related things that I’ve become known for saying in person and in comments, but haven’t written up properly).
For this specific point—about 2000 calories per day being a shortfall—it’s fairly straightforward. There’s a scientific-consensus answer to how many calories people typically need, the Harris-Benedict formula. (This has a few variations with slightly different constants, all of which trace back to linear-regressions on measured energy expenditure.) I typed in my own parameters, and those of a few other people, and observed that, within the demographics of people I know, ie mostly youngish male and with a height reflective of decent childhood nutrition, it was consistently well above 2000. I also traced the history of where the 2000 number used on nutrition facts panels came from, and found that it was never intended as any sort of recommendation, and seemed to have been misinterpreted as one by historical accident. I try to always include a link to the formula or a calculator for it, when I talk about calorie intakes, so that people can get the real number for themselves.
Thx for the comment.
Out of interest, I checked wikipedia, which gives three different equations (Harris-Benedict, Mifflin St Jeor, and Katch-McArdle), saying “Historically, the most notable formula was the Harris–Benedict equation”.
I just put in my details on them, and my daily caloric burn varied over 1000 calories (one was 500 less than the other which was 500 less than the other). One of them was pretty close to 2000. /shrug
Isn’t the link to the Harris-Benedict formula sufficient evidence? Like, that’s the actual formula nutritionists use. I don’t know any specific reason for using the 2000 calories per day target other than that it’s a nice round number.
I mean, not really? Everyone has told me that ‘nutritionists do/believe it’ is not a good reason to do/believe something. I’m also not saying that I stand behind 2000 calories per day; I’m saying that Jim says a lot of things about nutrition all the time and I want to know why.
Are you sure about the Harris-Benedict formula? It seems like Mifflin-St Jeor is the most reliable. Nonetheless, I’m curious if you have any recommended articles/books on diet and mental health?
https://en.wikipedia.org/wiki/Harris%E2%80%93Benedict_equation lists both, and seems to consider the Mifflin-St. Jeor version a revision that sort of uses the same name. The calculator I linked uses constants from Roza-Shizgal, which is technically neither of these. All of the sets of constants were produced by taking linear regressions on measured energy expenditure among different groups of people. I don’t have much in the way of opinions about which of these sets of constants is best; the differences between them are small compared to the difference between “2000 is what it says on the nutrition facts panel” vs actually-using-any-sort-of-formula, and compared to the error in the activity-factor constant and the measurement error in a non-Soylent diet.