tag:blogger.com,1999:blog-2800316528417070757.post6063293126364055581..comments2023-04-28T22:14:09.751+01:00Comments on Random Thoughts: Shangri La Diet: It Works!John Lawrence Aspdenhttp://www.blogger.com/profile/02587130870181071109noreply@blogger.comBlogger4125tag:blogger.com,1999:blog-2800316528417070757.post-72401353960114147342012-08-13T16:38:03.975+01:002012-08-13T16:38:03.975+01:00Glen, I hate to sound like a fan-boy but your expe...Glen, I hate to sound like a fan-boy but your experiences sound like what Seth Roberts' theory predicts!<br /><br />My friend Ruth told me the 'any diet works for a bit' idea, so I've added into my second go at this experiment:<br /><br />http://johnlawrenceaspden.blogspot.co.uk/2012/08/shangri-la-diet-ii-another-go-with-more.html<br /><br />The weird thing is that you found the oil becoming nauseating. Normally you'd find an unpleasant stimulus losing its power with practice, I think, as long as it doesn't do harm.<br /><br />I've always quite liked Olive Oil. I've been known to drink the real thing out of the bottle. Drinking the tasteless stuff is a bit weird (I literally can't tell I've drunk it except that my teeth feel slippery), but it's not nauseating. <br /><br />In fact I'm starting to look forward to it, which is vexing because it probably means that I'm starting to make the flavour-calorie association that will wreck the scheme.<br /><br />I'm currently seeking even more flavourless oils. <br /><br />Did you try the sugar-water version? I'm a bit reluctant to try that since I don't think it's a good idea to screw with your glucose metabolism in that way.<br /><br />I know about regression. One reason I tried this experiment is because the conventional advice and Seth Roberts advice are directly opposed. Trying S-L should make you *heavier*, especially if you're not really trying to lose weight. I'm actively trying not to try!<br /><br />If Seth's theory is correct, then there are actually loads of diet hacks that could work for you even if you can't stand the oil.<br /><br />If it's not, then it should fail for me. But your experience seems to confirm his theory rather than refute it.<br /><br />John Lawrence Aspdenhttps://www.blogger.com/profile/02587130870181071109noreply@blogger.comtag:blogger.com,1999:blog-2800316528417070757.post-92159154362888321252012-08-13T16:17:18.616+01:002012-08-13T16:17:18.616+01:00Your graph is most impressive! I get the impressio...Your graph is most impressive! I get the impression that 2008 was a conventional diet and rebound. Can't have been much fun.<br /><br />But it looks like Shangri-La really worked for you. Did you stop at 88kg because you're happy with that, or is that where the Shangri-La effect bottomed out?<br /><br />Don't take the Bayes stuff too seriously. <br /><br />I'm no statistician and I'm treating it like a set of independent trials (As if you were trying to work out what dice your friend is rolling if he tells you only the results). I imagine that there are considerably better ways to do this.<br /><br />On the other hand, I also took statistics in college (I'm a maths grad), and I found the frequentist approach so baffling that I dropped the subject. <br /><br />Bayes seems transparently clear (a branch of probability theory rather than a separate ad-hoc thing).<br /><br />My probability-related intuitions are all 'parallel universe'-type foolishness, but despite the foolishness, they don't tend to lead me to make mistakes when solving real problems.<br /><br />I've written a number of Bayes-related posts, trying to capture the intuition without using scary symbols. Here are a couple:<br /><br />http://johnlawrenceaspden.blogspot.co.uk/2010/12/captain-kirk-mr-spock-and-reverend.html<br /><br />http://johnlawrenceaspden.blogspot.co.uk/2011/01/many-classical-worlds.html<br /><br />and here was an attempt to figure out whether England or Australia were better at cricket during the last Ashes series, using the simplest possible models and the five win/lose/draw results:<br /><br />http://johnlawrenceaspden.blogspot.co.uk/2010/12/bayes-and-ashes.html<br /><br />For a sane (and detailed) approach, try David Mackay's beautiful Information Theory, Inference, and Learning Algorithms: http://www.inference.phy.cam.ac.uk/mackay/itila/book.html<br /><br />or ET Jaynes combative and fun: 'Probability Theory: The Logic of Science' http://omega.albany.edu:8008/JaynesBookJohn Lawrence Aspdenhttps://www.blogger.com/profile/02587130870181071109noreply@blogger.comtag:blogger.com,1999:blog-2800316528417070757.post-54096524680928216872012-08-13T15:35:37.447+01:002012-08-13T15:35:37.447+01:00"All diets work in the short run" seems ..."All diets work in the short run" seems like a good rule of thumb. Especially diets that require you to eat something nauseating (like oil) and/or boring (like sugar water). People lose weight in the first month of, say, an "eat lots of celery" or "eat lots of carrots" diet too. You force yourself to do something weird with your diet and it's disorienting. Eventually your body either adapts to the new diet or you run out of willpower and reject it.<br /><br />My experience was that I lost about 10-15 pounds when I tried Shangri-la, but couldn't force myself to stay on it (the oil became more nauseating and the appetite suppression effect tapered off) and I gained all the weight back within ~6 months.<br /><br />Another thing to keep in mind is regression toward the mean - people tend to start and stick to bizarre diets when they are *unusually heavy* and then credit the diet for a return to their own "normal".Glenhttps://www.blogger.com/profile/14661650090485723755noreply@blogger.comtag:blogger.com,1999:blog-2800316528417070757.post-55798612953923632152012-08-13T14:02:20.256+01:002012-08-13T14:02:20.256+01:00When I took statistics in college (and again in gr...When I took statistics in college (and again in graduate school), I found the section on Bayes to be confusing. I'm still confused.<br /><br />In any case, here is my experience with the Shangri-La diet:<br /><br /><a href="http://www.astrocyte-design.com/shangri-la-diet/index.html" rel="nofollow">www.astrocyte-design.com/shangri-la-diet</a><br />Alex Chernavskyhttp://www.astrocyte-design.com/noreply@blogger.com