As far as I can tell, ketogenic diets solve the problem of fat loss. I know, anecdotes are not data, but it’s worked wonders for everyone I know who’s tried it (myself included).
Err on the side of posting solutions which may not be universal but are still likely to be helpful to many people.
This is the sole reason I’m posting this. Keto works for very many people. The short story of keto is that your brain can only eat certain kinds of chemicals. Glycogen from eating carbohydrates is one of them. Ketones generated from fat is another. Your body will preferentially use the first over the second, since turning fat into ketones is expensive. So if you eat few enough carbs (<30g per day is the figure I remember) and plenty enough fat (2:1 fat to protein is what I heard), your body will eventually start doing chemistry that turns dietary and body fat into ketones.
There’s some more practical advice about how to induce ketosis quickly (muscles store glycogen, so exercise helps) and how to make low-carb versions of foods you enjoy, but that’s pretty much the gist of it.
Ketogenic dieting has been very effective for me. But I’m not convinced that this story about the body learning to turn body fat into ketones is actually how it works. My sense is that a super low-carb diet may just a good way of keeping appetite down and maintaining a caloric deficit. At the very least, that seems to be part of why it works so well: high fat low carb foods tend to be much more satisfying per calorie than foods heavy in carbohydrates. E.g. A Starbucks blueberry muffin is 380 calories which is like eating ten strips of bacon or 5 hardboiled eggs or more celery than you could possibly eat in one sitting. Whether or not the chemistry stuff is actually true keto is a good way to feel satisfied on a lower number of calories.
I wonder if a diet that was actually optimized for high satisfaction/calorie would be a) different and b) more effective.
high fat low carb foods tend to be much more satisfying per calorie than foods heavy in carbohydrates
I think it depends on the person. A meal without enough carbs just doesn’t feel ‘complete’ to me, and it’d take lots of willpower for me to not eat anything else for a while (but I’m mostly thinking about stuff like pasta or rice or potatoes or bread or fruits, rather than muffins); OTOH I can go several days without much proteins before starting to crave for them. And eating ten strips of bacon without anything else in one sitting would feel very distasteful to me. But I know there are people for whom it’s the other way round.
Just doing intermittent fasting by itself didn’t do much for me, but doing intermittent fasting where I also measured and reduced calorie intake allowed me to lose 15 pounds (from 185 to 170). The LeanGains website was really helpful. It’s also worth noting that prior to starting this, I already had a relatively healthy diet and exercised regularly.
That being said, even the guy who made LeanGains doesn’t claim it will work for everyone. But for the first month I did it I screwed it up (by not also restricting calorie intake) and saw no change, then after fixing that I lost about a pound a week.
Have you tried eating less and exercising more? How long did you “diet”?
Also, how closely were you monitoring things? How many calories below maintenance were you consuming daily, on average [300-500 kcal’s generally touted for muscle preservation for those not on steroids by the internet, but that’s still pretty slow and not obvious weightloss against a backdrop of fluctuating water weight]? How long did it take you to enter ketosis if you were carb cycling (measured more definitively using something like ketostix and not my housemate-on-keto’s “I can just feel it!”)?
Ketosis sticks did not show my entering ketosis even with as close to zero carbs as I could get (admittedly counting things like 3g carbs in a serving of protein powder). I don’t recall how long I tried. Probably between 1 week and 2 weeks before giving up on almost-zero carb, then a month of very low carb before giving up entirely. Memory is fuzzy.
Upvoted for evidence. I’ve read your comments on how diets that ought to work don’t for you, and that it’s not as simple as calories-in-calories-out, but have been skeptical because my prior for “Eliezer is a metabolic mutant” vs. “Eliezer has the same trouble sticking to a diet (and being honest when they fail) that most people do” is low.
In local parlance, my assignment for “Eliezer is a mutant” and “net calories aren’t everything” have both risen.
I am not a metabolic mutant. There are plenty of people in the world who cannot seem to lose weight, and they aren’t all weak-willed scum, and it’s not because they just haven’t tried your favorite diet.
The world is full of metabolic diversity. The fortunate who do not appreciate this are the metabolically privileged. That they can lose weight with an effort causes them to be unfortunately deluded about what is going on.
I am not a metabolic mutant. There are plenty of people in the world who cannot seem to lose weight, and they aren’t all weak-willed scum, and it’s not because they just haven’t tried your favorite diet.
I’m not sure I deserved the heat here. I prescribed no particular diet and said nothing about weak willed scum. I’m of the tentative opinion that modern weight-control problems are just a case of human brains not being built for an environment of plenty. Even if it was simply that people on average can’t keep their hands out of the pastry box, that’s not a moral failing, just an outdated adaptation. It’s worth fixing ourselves because it’s now a maladaptation and evolution is too slow about fixing it.
The world is full of metabolic diversity. The fortunate who do not appreciate this are the metabolically privileged. That they can lose weight with an effort causes them to be unfortunately deluded about what is going on.
What is going on? Is there a thread around here that you think covers it in useful detail? It seems to me that there must be some lower bound on food intake beyond which one can’t help but lose weight—otherwise you could eat nothing and still not lose weight, which seems spectacularly unlikely to me barring the aforementioned metabolic mutation.
It’s not a rhetorical question; I have only mild difficulty controlling my weight, but my partner has a much harder time. (though both our weights respond to consistent food restriction) Useful information would be useful.
It seems to me that there must be some lower bound on food intake beyond which one can’t help but lose weight
It has been observed that no fat people emerged from Auschwitz. But if some people do not lose weight below that level of privation, they don’t stand much chance of doing so voluntarily.
Perhaps they all died because they couldn’t access their metabolic reserves? [/devilsadvocate]
All the worse for those with great difficulty losing weight. Apply enough determination and it might kill you!
People would probably object to using Nazi data, but it would be interesting if they ever kept logs of inmates’ body weight and non-deliberately-caused deaths.
I’m not sure I deserved the heat here. I prescribed no particular diet and said nothing about weak willed scum.
As you may have guessed, this isn’t the first time the subject has come up. Frustration builds.
I’m of the tentative opinion that modern weight-control problems are just a case of human brains not being built for an environment of plenty.
It is not just the brain but the entire human body that isn’t specialized for an environment of plenty. Lowered food intake changes metabolism and energy expenditure, it doesn’t just make you crave more food.
Even if it was simply that people on average can’t keep their hands out of the pastry box, that’s not a moral failing, just an outdated adaptation. It’s worth fixing ourselves because it’s now a maladaptation and evolution is too slow about fixing it.
That would indeed still be true.
It seems to me that there must be some lower bound on food intake beyond which one can’t help but lose weight—otherwise you could eat nothing and still not lose weight, which seems spectacularly unlikely to me barring the aforementioned metabolic mutation.
That seems technically inevitable. The question then becomes whether this happens before or after your body enters a coma. (Or, more practically, whether valuable muscle mass is lost before undesired fat and whether the effect on fatigue and energy levels is debilitating.)
A ridiculously charged topic, how could I miss it?
We’re probably among the last generations (as in so many things) that need to bother with the now counterproductive and out-of-place esterification making us fat. Just as we managed to separate the carrot from the stick (the stimulus we get from an action versus the original evolutionary incentivized purpose of that action) with sex/procreation, so will we eventually be able to indulge in feeling satiated without actually storing unwanted lipids.
If not for too strict pharmaceutical standards, some (more) drugs would probably already be available. Given the large impact of diabetes, CHD and other obesity related diseases, even severe side effects in animal trials could be outweighed by the benefits. If not for the fear of lawsuits and strict regulations that throw promising drugs out of the pipeline prematurely. It took a decade and untold needless deaths for gene therapy to recover from a few mishaps to where it can be pursued again.
Regarding losing weight, personally I like the volumetrics approach, it’s easily combinable with most diets:
Feeling satiated is—mostly—a combination of mechanoreceptors in the stomach being activated (which is how gastric lap band surgery works) and various hormones reacting to e.g. rising blood sugar (hunger-stimulating Ghrelin gets inhibited), presence of food in the intestinal tract (hunger-inhibiting PYY is released, NB: it’s released more effectively by high-protein intake which would help explain the effectiveness of some high-protein diets, such as variants of keto), and leptin (released by adipocytes, can be mostly ignored, since obese people apparently have high leptin levels and a corresponding high leptin resistence).
Now, there is of course a latency between food entering your stomach and the mechanoreceptors triggering and PYY being released.
Therefore, a sensible measure is the following: Drink a large-ish quantity of water (cold and tasty with lemon) before every meal (to pre-load the mechanoreceptors), eat slowly (so you don’t eat more when your satiation signal is already in the process of being triggered), add taste-neutral e.g. salad to your normal food (I just pack my sausage sandwiches with large quantities of unprepared salad = little extra effort, halves the amount of high-caloric food I require before feeling completely full, mixed in it doesn’t even taste any different*.) Don’t drink caloric beverages, they’re not as effective calorie-for-calorie in distending your stomach, since fluids aren’t bulky enough.
These changes do not inhibit my “food experience” in any meaningful way, yet nearly halved my daily caloric intake.
It seems irrational to give up on dieting merely because the process can be complicated. That’s not a good reason to give up on FAI, so why should it be for dieting? The impact in energy levels between being obese and normal is redunculous, even if losing weight is hard, it still promises the most bang-for-the-buck in increasing your productivity, your self-image and your quality of life. Sure some have it easier or harder than others. C’est la vie.
* The trick is that salad-diluted sausage still tastes just like sausage, and still causes the satiety inducing effects on the same order as double sausage.
Obesity is interesting because I regard it as a partially-solved problem. For example, dinitrophenol would solve much of it: it makes mitochondria less efficient and so effectively increases metabolism, but at the cost of emitting waste heat—which is potentially fatal and got it banned despite its apparent effectiveness. It could still be safely used; it’s 2013 so electronic thermometers are a dime a dozen. Take patients to a fat camp, dress them in clothes with thermometers constantly recording and now doses can be adjusted based on detailed data and the thermometers can warn the patient to jump into conveniently located ice baths. Voila. And I’m not clear on how dangerous it really is when not made illegally and used recklessly by young kids; Wikipedia cites a 1934 paper as estimating that there were ~100k users of DNP before it was banned, and those authors remark, after discussing the grand total of 4 deaths up to that point due to the drug’s use under medical supervision, that
When one considers that some one hundred thousand patients have been treated with this exceedingly potent therapeutic agent, it is a matter of some gratification to know that fatalities have not been more numerous. It might be added in this connection that fatalities from the fever of dinitrophenol can be largely prevented, in animals at least, by chilling the skin with ice packs and by giving oxygen inhalations.17
It would seem that like the War on Drugs in general, the attempt to reduce the harm from DNP has resulted in far more harm than was ever the case before.
It is frustrating to me that I want powerful cognitive enhancement, but instead of increasing metabolic efficiency, we’re decreasing it. Can’t we funnel those extra calories the the brain somehow?
I think most of the value in being thin is looking attractive and being able to be physically active. I think Eliezer doesn’t really need to be more attractive than he already is (4 girlfriends) and isn’t a huge fan of rock climbing or whatever. As far as I’ve heard, most of the health benefits of exercise can be gotten without needing to actually be thin.
I somehow doubt that all of this effect is due to thin people exercising more. ETA: looks like the ‘optimal’ BMI for women is larger than for men, BTW.
Considering the social import of being attractive and getting around to see people and the correlation between active social life and longevity I think it’s more than you might think, but I agree with you. On the other hand, the trade off is a lot more reasonable if you can be relatively healthy and happy while fat AND it’s particularly hard for you to lose weight.
Considering the social import of being attractive and getting around to see people and the correlation between active social life and longevity I think it’s more than you might think, but I agree with you.
Well, they did control for, among other things, marital status. (Also, I’d guess that the BMI that maximizes conventional attractiveness would be higher for men than for women, and wouldn’t depend much on smoking.)
On the other hand, the trade off is a lot more reasonable if you can be relatively healthy and happy while fat AND it’s particularly hard for you to lose weight.
Yes. ISTM that for certain people losing weight has become a lost purpose.
Remember that BMI is based on the total body mass, and that muscle is denser than fat. (OTOH, that study corrected for level of exercise, and it’s quite possible that the BMI that would maximize an average white American man’s attractiveness if he’s not allowed to vary his level of exercise would indeed be around 20.)
that curvy figures are considered (by many) to be attractive?
I am one of those “many”, too, but ISTM that in present-day Western cultures we’re a minority; “thin” seems to have become a compliment. (Last year, someone offered to set me up with her roommate who probably had BMI around 18, and when I told her that I didn’t fancy her, she retorted “but she’s so skinny!” as though it was a positive.)
I don’t want to speculate about EY other than saying my model of him didn’t expect to ever see “too complex” brought up as a reason not to try anything. Maybe there’s too much overlap with HPMOR:Harry. It is possible that there are individuals whose akrasia levels / mental energy and self-image are unaffected by being overweight compared to not being overweight. Just unlikely.
As far as I’ve heard, most of the health benefits of exercise can be gotten without needing to actually be thin.
That’s true, but doesn’t change that ceteris paribus given little exercise, you’ll still live longer not being obese. Note that I’ve not even mentioned exercise. It’s certainly better being overweight and exercising (while still being overweight), than being overweight and not exercising.
It seems irrational to give up on dieting merely because the process can be complicated.
Expected value calculation > your ‘seeming’.
That’s not a good reason to give up on FAI, so why should it be for dieting?
Because creating an FAI has (strictly!) greater expected benefits than one person successfully losing weight? To the extent that the rhetorical question is ridiculous.
it still promises the most bang-for-the-buck in increasing your productivity, your self-image and your quality of life.
I’d wager you’ve never been overweight. Strap a few dozen lbs around your waist or don them as a vest, see what it does to your daily routine. We’ve done that once in some class or other, and I’ve bordered the 30 myself as well, from time to time. You’re affected constantly, we’re as of yet embodied agents, not free floating minds. What’s the likelier explanation for a lack of action, expected value calculations or—here it comes - ‘akrasia’.
Because creating an FAI has (strictly!) greater expected benefits than one person successfully losing weight?
Depends from whose point of view. E.g. passing away in the knowledge that you’ve contributed to the eventual creation of FAI (which gives you fuzzies, or at least utilons) can be outweighed by living decades with more mental energy (which also contributes to your development efforts) and a better self-image.
That aside, that a task is a complicated problem/puzzle to be solved can be an incentive to solve it in and of itself, especially for certain kinds of people.
This seems false.
Assuming the increase in productivity, self image and quality of life (consider the metabolic syndrome, preventing decades of injecting insuline can have quite the impact on your QALY) to be fixed/constant for an individual, “true” or “false” does depend on how easy/hard it would be for that individual to efficiently attain and keep a lower BMI. For metabolically priviledged people, or just those with an easy to fix problem such as hypothyreodism, the statement is probably true. For someone who for whatever reason cannot lose any weight whatever he tries (within his motivational reach given his current energy levels … there’s a catch-22 present), it would be false.
Eliezer has mentioned many of the things he has tried to lose weight (including ketogenic diets and even clenbuterol). I’ve tried all those he has mentioned. The difference is for me they work (I call it ‘cutting’ and can merrily play around with my body composition all sorts of ways). But if, like Eliezer, I had expended huge amounts of effort and my body did not respond significantly then I would update my expectations.
What’s the likelier explanation for a lack of action, expected value calculations or—here it comes - ‘akrasia’.
Expected value calculations. Unless you are making accusations of lies—outright fabrication of self reports.
Depends from whose point of view. E.g. passing away in the knowledge that you’ve contributed to the eventual creation of FAI (which gives you fuzzies, or at least utilons) can be outweighed by living decades with more mental energy (which also contributes to your development efforts) and a better self-image.
For the purpose of declaring an accusation of irrationality false the relevant point of view is Eliezer’s. If Eliezer had someone else’s values then it would make sense to evaluate the rationality of a given choice for him according to those other values.
For metabolically priviledged people, or just those with an easy to fix problem such as hypothyreodism, the statement is probably true.
Yes (or at least it would be up there on the list). It just isn’t true in this case.
Depends from whose point of view. E.g. passing away in the knowledge that you’ve contributed to the eventual creation of FAI (which gives you fuzzies, or at least utilons) can be outweighed by living decades with more mental energy (which also contributes to your development efforts) and a better self-image.
Pretty sure most people involved in FAI efforts are fatoring in more than warm fuzzies in their EU calculations.
A ridiculously charged topic, how could I miss it?
Well, I missed it; I kind of wish the LW inbox included replies-to-replies. In any case I find myself thinking this shouldn’t be a charged topic, even though it clearly is. As a culture we’re still hung up on the whole self-indulgence-as-moral-failure absurdity.
Just as we managed to separate the carrot from the stick (the stimulus we get from an action versus the original evolutionary incentivized purpose of that action) with sex/procreation, so will we eventually be able to indulge in feeling satiated without actually storing unwanted lipids.
A thousand times yes. Sex without babies is a nearly-solved problem, lacking only a male equivalent of the IUD. Weight control should be just as solved, and no, self-control in food intake is not a solution even if it does work. Like condoms, it’s a badly suboptimal necessary evil. I should be able to eat whatever I feel like while maintaining whatever weight I damn well please.
That this has failed to happen despite the massive amount of money thrown at weight-control products says something, but I’m not sure what. Has anyone made a serious, prolonged attempt not at “getting people to ‘stay good’ around food” but at “decoupling food from fat entirely”? If not, why not?
That this has failed to happen despite the massive amount of money thrown at weight-control products says something, but I’m not sure what. Has anyone made a serious, prolonged attempt not at “getting people to ‘stay good’ around food” but at “decoupling food from fat entirely”? If not, why not?
May I direct you to the subsequent paragraph in the grandparent? :)
Sort-of-valid point. But all drugs I’ve seen or heard of have been essentally appetite suppressants plus varying levels of bullshit. I wouldn’t expect metabolic decouplers to be disproportionately cut out by regulation; if there was substantial research in that area, I would expect some of them to be on the market.
Corrected. Although, it’s somewhat hard replacing carbohydrates without also increasing the overall protein intake. There are probably variants of keto that fit, and variants that don’t. There’s apparently no catchy phrase for the overall “high protein diet” reference class.
Yeah, most people who do keto or other low-carb variants tend to eat a high-protein diet. But for fat-loss something like twice as much weight of dietary fat as protein is recommended along with sub-30 grams of non-fiber carbohydrates. And the medical keto diet, iirc, was something like 4:1 fat:protein. It’s a lot easier to hit the protein macros than the fat macros.
As you may have guessed, this isn’t the first time the subject has come up. Frustration builds.
Fair. I’d seen some of the previous conversations or I wouldn’t have responded as I did; but I’m guessing I missed occasions on which Eliezer demonstrated that his body chemistry was provably not doing what theory would predict it should (with regard to ketosis), which I find much more convincing than “I tried X, Y, and Z and none of them worked.”
If such occasions exist, then I just plain missed them and I’ve stepped on toes unnecessarily, and I apologize.
That seems technically inevitable. The question then becomes whether this happens before or after your body enters a coma. (Or, more practically, whether valuable muscle mass is lost before undesired fat and whether the effect on fatigue and energy levels is debilitating.)
I’m going to guess “before muscle loss or coma”, on the grounds that fat is supposedly for long-term energy storage, and I would expect “break down muscle and/or go into coma in preference to using stored energy” to be maladaptive even in the EEA. I have no controlled study to support that guess, however.
I could see “after excessive fatigue,” I suppose; that might be physically necessary (fat burning is slow, IIRC) but not so maladaptive as to defeat the purpose of storing fat.
it’s not because they just haven’t tried your favorite diet.
This seems somewhat unfair. There are a handful of diets that work on a broad variety of people, such that the prior any one will work for a particular person is higher than a novel diet like Shangri-La. And so unless you’ve tried slow carb/ketogenic, intermittent fasting, 30g of protein for breakfast, and ECA stacks and none of them worked, it seems like you’re updating too far in the direction of “all diets don’t work for me” from the evidence that “diet X didn’t work for me.”
(The only one of those I’ve tried is IF. It worked for me.)
The world is full of metabolic diversity. The fortunate who do not appreciate this are the metabolically privileged. That they can lose weight with an effort causes them to be unfortunately deluded about what is going on.
What’s the difference between this and the claim that you are a metabolic mutant?
It might be worth going to a sleep doctor; sleep apnea can really fuck up your metabolism, not to mention causing unbelievable akrasia. I would say sleep tests are a GOOD THING, something everyone should do. I had sleep apnea for years. It was like some eldritch monster was sucking away my willpower and I wasn’t even aware. Within a few months of getting my mouth guard, which keeps my tongue from blocking my airway while in REM, I lost thirty pounds and gained an enormous well of mental stamina. A small minority of the “metabolically challenged” may just have undiagnosed sleep problems.
Well, there’s your problem right there—it sounds like you didn’t really especially “do” a ketogenic diet, if you never once reached ketosis (as I understand, you typically need to deplete glycogen stores before entering ketosis, and so you might have 3,000~6,000+ kcal to burn through first, and then there’s an interim period of glycogen depletion before it actually begins. Which is where the magnitude of your caloric deficit becomes relevant; was it large enough to get through your stored glycogen in the 1-2 weeks you ate no-carb? I think recently consumed food is generally metabolized first, before dipping into glycogen stores, though that’s likely a huge oversimplification), any more than someone eating an egg for breakfast can be said to have “tried” keto.
And even with weight as your metric you should have seen some noticeable reduction after a month and a half of “dieting” on a >300 kcal deficit if you were consistent in your measurement conditions (eg, every day after waking/bathroom and before breakfast). Maybe you just overestimated how many calories you needed for maintenance each day? Which is quite common among “dieters”. Like I mentioned, the usual advice for muscle preservation is to eat 300-500 kcal below maintenance each day, and if after a month you fail to note any weight loss or note weight gain to reduce your consumption by an additional 300 kcal, rinse and repeat. It might take several months to note any reduction in weight if you had a shoddy initial estimate (or if your metabolism is exceptionally sensitive to intake, though that can only account for so much. Eventually as you progressively reduce consumption [or increase activity!] you will lose weight).
Haha, and many people report a “mental fog” when first trying ketogenic diets. Maybe that’s why you can’t remember :D
disclaimer: this isn’t my field of study and it’s been several years since I tried keto and my research then was cursory, at best, which is why I used less-than-confident language in the above. In the three months that I ran a cyclic ketogenic diet I think it took 2-3 weeks to make the ketostik turn purple (or whatever) the first time, then a weekish after that, ulitmately down to 2-4 days after each weekly recarb.
edit: also, I’ve heard it might be useful to think of diets less in the “dieting” sense (a temporary change in eating behavior that will end once you reach a predefined goal) and more in a “biological” sense (the typical assortment of food that an organism habitually consumes). Less temporary, more permanent or semi-permanent lifestyle change.
After numerous previous failures, if it’s that complicated I’m not going to bother. Complicated things seem even less likely to work than simple things, and simple things almost never work in the first place.
In my experience, no matter what you try, there’s always an excuse when it doesn’t work. Then when it still doesn’t work there’s something else you’re not doing exactly right that they forgot to mention earlier. Oddly enough, when something does work for someone, nobody bothers to check to see if they were doing everything exactly right by way of confirming that all these extra frills are actually required as opposed to just being excuses that are only invoked when it doesn’t work because, in reality, metabolisms are different.
Do you have any data on your eating and exercising habits? I’d love to know what or how your eating now. If you’re eating the standard american diet, then you’re definitely doing something wrong, and there are a few simple things you could do to at least eliminate deleterious factors. Do you drink Coca Cola everyday for example? What’s your baseline diet? Before ever trying a new one, find out what’s wrong with the baseline and tweak it.
[300-500 kcal’s generally touted for muscle preservation for those not on steroids by the internet, but that’s still pretty slow and not obvious weightloss against a backdrop of fluctuating water weight]
It is obvious if you weigh yourself every day for a couple months or longer and you know how to do stats.
(FWIW, my weight since 12 February fits to a straight line a + bx where a = (93.74 ± 0.19) kg, b = (−0.018 ± 0.007) kg/day, and x is the time elapsed since 12 February; the RMS of residuals is 0.68 kg. Approximating the posterior pdf of b as a Gaussian, which ought to be close enough given 46 degrees of freedom, I’m 99.42% sure that b < 0.)
Haha, well yeah. Though you should hardly need stats if you’re recording over a period of months (“golly, I wonder if my 40 lb weight change these past 6 months is just me being dehydrated right now? Maybe I should wait till after I drink my morning 4 gallons just to be sure”). I meant it more on time scales of “between 1 week and 2 weeks”, or for where weight loss was very minor due to a tiny caloric deficit.
With more precise measurement (eg, via bodpod) of body composition you would better be able to track smaller changes, too.
I suspect you’re basically correct, but I would not take the stats results at face value. There are many possible problems resulting from the physical and electrical properties of the scale you’re using, that I would not expect to be well behaved in a stats sense. In particular: quantization errors, non-linearity / non-monotonicity of the scale A/D converter (depends strongly on type of A/D used), temperature dependence of both the scale strain gauges and A/D, etc.
The general rule here is that trying to get too many more bits of precision out of a measuring device than it is intended to provide is tricky.
You could calibrate the scale in a number of ways; easiest would probably be to check that it gives consistent readings over time for a fixed weight that’s not too small compared to you. You could simply weigh the fixed weight, or you could weigh you and (you + weight).
You’re right, any time-varying systematic error (due to temperature, ageing of the scale, etc.) would screw up the analysis. (Quantization errors shouldn’t matter that much so long as they’re much smaller than day-to-day fluctuations.)
Worked for me, though I have been plateaued for the last year a few pounds and several waste inches above where I’d ideally like to be. Haven’t figured out how to bust through that yet.
I strongly suspect that when we do figure this out, ketogenesis will be a large part of any eventual solution, but I don’t think we have all the answers yet.
I do know folks for whom low carb diets failed, but in all cases I’m personally familiar with that’s because they couldn’t maintain them. (which is a strike against low-carb diets, of course; an unmaintainable diet isn’t useful) I know of one case of someone who took off 150+ pounds using low-carb and then put 50 or so pounds back on while continuing low carb. Still he never came close to getting back to where he was pre-low-carb. I’m curious if anyone has simply failed to lose weight while maintaining a low carb diet. I think Atkins talked about this possibility in the last edition of his diet book he authored. I’ll have to look up the citation, but there are some medical conditions that can cause weight gain/prevent weight loss that show up in a few percent of patients.
Err, have you been lowering calorie intake relative to your activity and changing metabolic rate? Lighter bodies require less energy both to maintain and move around. If you haven’t been adjusting your “dieting” diet it’s no wonder you plateaued, because where initially you ran a caloric deficit you’re now much closer to equilibrium.
Also, I’ve intentionally gained weight on cyclical low-carb diets (<5 g carbs each weekday, 1200-1500g carb-up on weekends). It’s because I ate a lot.
As far as I can tell, ketogenic diets solve the problem of fat loss. I know, anecdotes are not data, but it’s worked wonders for everyone I know who’s tried it (myself included).
If I may ask, what was your starting weight; what is your current weight; and how long have you been at your current weight?
when I started keeping track 479 days ago I was 270, right now I am at 228. This isn’t pure keto, because at first I was doing slow carbs, and then when I switched to no carbs I kept the tim feriss style cheat day. I’ve also cheated on various trips and so on I’ve taken, but in general I have 6 days on 1 day off diet. I’ve gone down something like 4 pants sizes. Right now I’m trying to transition to fewer cheat days or none at all because I feel like I’m plateauing, but it’s rough.
A little? I tend to feel really heavy and bleh at the end of cheat nights and kinda weird the next day but it’s not really what I would call “keto flu”
lost some progress, up to 238.6 as of this morning, which is down from getting back up over 240. I stopped tracking regularly for a while, and now I am again and I don’t know why but tracking on a daily basis seems to have some sort of causal effect making me lose weight again.
I haven’t been keeping the closest track, so this isn’t even a good anecdote. I was a 165 lb and 5′10 man going out of high school seven years ago, and went up to 220 as of a year ago. It wasn’t all fat (I’m significantly stronger now than I was then), but I did go from wearing 34 waist jeans to a 38.
So I started at 220 and a 38 waist, and I’ve gone down to 180 and I think size 34 jeans (I haven’t gotten around yet to buying more jeans, but my size 36′s need a belt now). Exercise involved hasn’t been too extensive—mostly just walking and housework. I’m still work-in-progress, but I haven’t had a history of my weight yo-yoing on diets, and I intend to stay on keto for the foreseeable future.
I haven’t been keeping the closest track, so this isn’t even a good anecdote. I was a 165 lb and 5′10 man going out of high school seven years ago, and went up to 220 as of a year ago. It wasn’t all fat (I’m significantly stronger now than I was then), but I did go from wearing 34 waist jeans to a 38.
So I started at 220 and a 38 waist, and I’ve gone down to 180 and I think size 34 jeans (I haven’t gotten around yet to buying more jeans, but my size 36′s need a belt now). Exercise involved hasn’t been too extensive—mostly just walking and housework. I’m still work-in-progress, but I haven’t had a history of my weight yo-yoing on diets, and I intend to stay on keto for the foreseeable future.
Thank you for your response. Will you agree to update this response in a year or two?
Great, thanks. I have been doing informal research on diet and weight loss for a while now and I think there is value in following up with people a year or two down the road.
Update: I’m at pretty much the same place now as I was then. Dropped the keto diet since I was happy with where I was. Still fairly active but not hardcore about it.
The calories-in calories-out model is attractive, but it doesn’t appear to be all that accurate, or at least it’s incomplete. The body responds differently to different foods. They might have different effects on various hormones (e.g. the ones that regulate hunger), and they might be broken down and redistributed in different ways. In one study (Kekwick and Pawan), three groups of people were put on 1,000 calorie diets of 90% fat resp. 90% protein resp. 90% carbs. The first group lost 0.9 lbs / day, the second group lost 0.6 lbs / day, and the third group gained 0.24 lbs / day. (I don’t know to what extent the study controlled for exercise but I think it’s safe to assume that the difference in the amount of exercise that each group did wasn’t large enough to explain these results.) As Tim Ferriss puts it in The 4-Hour Body:
The creator of the “calorie” as we know it, 19th-century chemist Wilbur Olin Atwater, did not have the technology that we have today. He incinerated foods. Incineration does not equal human digestion; eating a fireplace log will not store the same number of calories as burning one will produce. Tummies have trouble with bark, as they do with many things.
In the context of solving the specific problem of fat loss, one goal is not to lose muscle, and what you’re eating should affect how easy it is to target fat and retain muscle as well.
And, of course, diets don’t work if they can’t be maintained. The fact that different foods affect the hormones that regulate hunger differently (some foods even make you hungrier!) means that different diets, even with the same number of calories, require different amounts of willpower to maintain.
I’m heartened that your comment is so well-liked. I made the same point a year or two ago and got back a bunch of nonsense about how the second law of thermodynamics cannot be violated.
It’s interesting that the second law of thermodynamics is what people use here. As long as your metric is losing weight, the relevant physical law is conservation of mass, and starting from conservation of mass helps clarify the issue enormously, I think. (Apparently the mechanism by which burning calories actually causes you to lose weight is exhalation of carbon dioxide. I feel like I knew this once, but forgot and only very recently relearned it.)
Carbon Dioxide and water are the two main byproducts of fat metabolism, which made me really happy to learn because it basically killed my worries about focusing on bowel movements. You literally can sweat the fat away!
Indeed not from burning calories, sorry. However, it still has to be subtracted from the intake, just like the amount you exhale, so reducing absorption is just as important as increasing burning.
Actually, I could not find any data online on the food energy utilization in the small intestine in humans and on the factors affecting it. Apparently the obvious ways to prevent absorption, like laxatives, don’t really work in the long term.
I don’t think so, because calories are a unit of energy, so a simple calories-in calories-out model would necessarily model energy balance as energy intake (through food) and energy expenditure (through body maintenance and activity). Your thermodynamic/energy balance is what would ultimately determine either the anabolism or catabolism of different tissues (a more complex calories-in calories-out model, rather than the simpler one mentioned, would have the greatest explanatory power, I imagine. Metabolic rate is under hormonal control, hormones interact in complex ways, and intake of different foods and different activity patterns can alter hormone expression. A calorie-budget model incorporating varying hormone expression/sensitivity, genetics [for both little things like minute differences in receptor molecules and bigger ones like hyperthyroidism], and different metabolic pathways used [as dictated largely by the previous two], would be pretty accurate, I reckon).
Weight can also be stored as different things, which is why it’s not the best proxy for the success or failure of a “diet”. Different substances (like fat or muscle or glycogen or water) have different energy densities, so not all changes in weight signify the same thing (presumably, a “dieter” wants to lose fat. Drinking a gallon of water might cause his weight to rise beyond where it was a month ago, but that does not mean that his “diet” has been shot, or that he has gained fat). A study that looks only at weight change and not change in body composition under different conditions would enormously simplify what that weight represents—in the study Qiaochu_Yuan mentions, 90% fat diets might have gained fat and lost muscle, water, and glycogen, resulting in a net weight loss, where the 90% carbs group gained muscle, glycogen, and water, and lost fat, resulting in net weight gain. The second case is obviously preferable to “dieters” than the first.
(what I suspect happened in the study is that the first two groups were depleted of glycogen, an energy source the body tends to use before it starts catabolizing fat or muscle deposits. Glycogen also tends to increase water retention, further inflating the weight of high-carb dieters. If they put all three groups on a 0-carb diet for a week to deplete glycogen stores and then dehydrated them, I’d suspect their net weight loss to be much more similar (changes in muscle and fat would vary [again, because of hormonal interactions and different energy densities], but bodies seem to need a lot less protein for maintenance than often suggested so unless the non-protein groups were consuming mostly carbs/fats for their 10% I doubt there’d be a huge difference)(also, the fact that obese individuals might gain muscle/fat on a 1000 kcal diet is obviously silly. If they were using the [again, stupidly oversimplified] BMI to define obesity [where obesity occurs at BMIs>30], a 5′10″ man would have to weigh >209lbs, giving him a BMR at minimum of roughly between 1500 and 2000 calories, which is a good bit more than 1000kcal, so he’d still probably lose weight [but ignoring changes in BMR from hormones and such] completely sedentary)
There’s also calorie type influencing activity level (eg, People starting ketogenic diets often report lethargy that eventually levels off, but whether that’s a true return to past alertness or a subjective change in perceived vigor I’m not sure).
I don’t understand what you’re disagreeing with. Is it “as long as your metric is losing weight, the relevant physical law is conservation of mass”? Because that seems obviously true to me. What you seem to be arguing is that your metric shouldn’t be losing weight, which is reasonable, but you’re not disagreeing with me.
Huh, I think I read your comment too quickly and missed the “as long as...: qualifier and then started replying and went off on a tangent and forgot what the original comment was. Hah. My bad. Also didn’t notice your name, hence my reference to you in the 3rd person.
Yeah, weight’s not the best metric to use without taking into account body composition.
Oh wait, I think I figured it out. I’d combined your post and paper-machine’s in my head, so I thought the simple calories-out-calories-in model in the highest level post being the thing referred to by discussion of conservation laws.
I’m heartened that your comment is so well-liked. I made the same point a year or two ago and got back a bunch of nonsense about how the second law of thermodynamics cannot be violated.
I noticed the same difference in response myself and was similarly pleasantly surprised.
1,000 calorie diets …
third group gained 0.24 lbs / day
I noticed I was confused. This doesn’t seem consistent with the results of the Minnesota Starvation/Semistarvation Study. I went to Wikipedia.
Kekwick and Pawan, 1956 conducted a study of subjects consuming 1000-calorie diets, some 90% protein, some 90% fat, and some 90% carbohydrates. Those on the high fat diet lost the most, the high protein dieters lost somewhat less, and the high carbohydrate dieters actually gained weight on average.
Kekwick and Pawan noted irregularities in their study (patients not fully complying with the parameters of the study). The validity of their conclusions has been questioned, and follow-up studies over a longer duration concluded that these temporary differences were due chiefly to changes in water balance (citation)
My prior consider it quite ludicrous that you can gain weight eating at a 50% deficit, no matter what your macros. The criticisms seem reasonable enough to explain the effect.
Note that the link in the citation claimed that when told to cut out carbs and eat as much protein and fat as they liked, “In all subjects, there was a reduction in calories ranging from 13% to 55% during the time they were consuming the low-carbohydrate diet.”
So long as your diet isn’t nearly that lopsided, IME (YMMV) the calories-in calories-out is a more decent first-order approximation than many people realize. See also The Hacker’s Diet. Second-order effects exist, but they’re second-order effects.
Places way too much focus on losing weight. See parent; losing weight by losing muscle mass isn’t desirable.
Second-order effects exist, but they’re second-order effects.
Your claim here hinges on the presumption that CI and CO are the only first-order effects, which is almost certainly false. Age, body fat proportion, maximal oxygen uptake, etc., are plausible candidates that I’ve seen in mathematical weight models.
Age, body fat proportion, maximal oxygen uptake...
In my experience, these tend to be taken into effect when calculating the “calories out” part of the equation. By what mechanism were you thinking that these mattered, that’s not “calories out”?
Sure, but as you probably know, calories eaten are stored into multiple deposits, depending on various internal and external factors. If the availability of those deposits varies, it might not be so easy to deplete most of the calory intake, short of completely starving oneself.
As far as I can tell, ketogenic diets solve the problem of fat loss. I know, anecdotes are not data, but it’s worked wonders for everyone I know who’s tried it (myself included).
This is the sole reason I’m posting this. Keto works for very many people. The short story of keto is that your brain can only eat certain kinds of chemicals. Glycogen from eating carbohydrates is one of them. Ketones generated from fat is another. Your body will preferentially use the first over the second, since turning fat into ketones is expensive. So if you eat few enough carbs (<30g per day is the figure I remember) and plenty enough fat (2:1 fat to protein is what I heard), your body will eventually start doing chemistry that turns dietary and body fat into ketones.
There’s some more practical advice about how to induce ketosis quickly (muscles store glycogen, so exercise helps) and how to make low-carb versions of foods you enjoy, but that’s pretty much the gist of it.
Ketogenic dieting has been very effective for me. But I’m not convinced that this story about the body learning to turn body fat into ketones is actually how it works. My sense is that a super low-carb diet may just a good way of keeping appetite down and maintaining a caloric deficit. At the very least, that seems to be part of why it works so well: high fat low carb foods tend to be much more satisfying per calorie than foods heavy in carbohydrates. E.g. A Starbucks blueberry muffin is 380 calories which is like eating ten strips of bacon or 5 hardboiled eggs or more celery than you could possibly eat in one sitting. Whether or not the chemistry stuff is actually true keto is a good way to feel satisfied on a lower number of calories.
I wonder if a diet that was actually optimized for high satisfaction/calorie would be a) different and b) more effective.
I think it depends on the person. A meal without enough carbs just doesn’t feel ‘complete’ to me, and it’d take lots of willpower for me to not eat anything else for a while (but I’m mostly thinking about stuff like pasta or rice or potatoes or bread or fruits, rather than muffins); OTOH I can go several days without much proteins before starting to crave for them. And eating ten strips of bacon without anything else in one sitting would feel very distasteful to me. But I know there are people for whom it’s the other way round.
Hasn’t worked for me.
Sigh… me neither. Have you tried intermittent fasting? I’m just adapting to it these weeks. Will see if it works.
Just doing intermittent fasting by itself didn’t do much for me, but doing intermittent fasting where I also measured and reduced calorie intake allowed me to lose 15 pounds (from 185 to 170). The LeanGains website was really helpful. It’s also worth noting that prior to starting this, I already had a relatively healthy diet and exercised regularly.
That being said, even the guy who made LeanGains doesn’t claim it will work for everyone. But for the first month I did it I screwed it up (by not also restricting calorie intake) and saw no change, then after fixing that I lost about a pound a week.
Have you tried eating less and exercising more? How long did you “diet”?
Also, how closely were you monitoring things? How many calories below maintenance were you consuming daily, on average [300-500 kcal’s generally touted for muscle preservation for those not on steroids by the internet, but that’s still pretty slow and not obvious weightloss against a backdrop of fluctuating water weight]? How long did it take you to enter ketosis if you were carb cycling (measured more definitively using something like ketostix and not my housemate-on-keto’s “I can just feel it!”)?
Ketosis sticks did not show my entering ketosis even with as close to zero carbs as I could get (admittedly counting things like 3g carbs in a serving of protein powder). I don’t recall how long I tried. Probably between 1 week and 2 weeks before giving up on almost-zero carb, then a month of very low carb before giving up entirely. Memory is fuzzy.
Upvoted for evidence. I’ve read your comments on how diets that ought to work don’t for you, and that it’s not as simple as calories-in-calories-out, but have been skeptical because my prior for “Eliezer is a metabolic mutant” vs. “Eliezer has the same trouble sticking to a diet (and being honest when they fail) that most people do” is low.
In local parlance, my assignment for “Eliezer is a mutant” and “net calories aren’t everything” have both risen.
I am not a metabolic mutant. There are plenty of people in the world who cannot seem to lose weight, and they aren’t all weak-willed scum, and it’s not because they just haven’t tried your favorite diet.
The world is full of metabolic diversity. The fortunate who do not appreciate this are the metabolically privileged. That they can lose weight with an effort causes them to be unfortunately deluded about what is going on.
I’m not sure I deserved the heat here. I prescribed no particular diet and said nothing about weak willed scum. I’m of the tentative opinion that modern weight-control problems are just a case of human brains not being built for an environment of plenty. Even if it was simply that people on average can’t keep their hands out of the pastry box, that’s not a moral failing, just an outdated adaptation. It’s worth fixing ourselves because it’s now a maladaptation and evolution is too slow about fixing it.
What is going on? Is there a thread around here that you think covers it in useful detail? It seems to me that there must be some lower bound on food intake beyond which one can’t help but lose weight—otherwise you could eat nothing and still not lose weight, which seems spectacularly unlikely to me barring the aforementioned metabolic mutation.
It’s not a rhetorical question; I have only mild difficulty controlling my weight, but my partner has a much harder time. (though both our weights respond to consistent food restriction) Useful information would be useful.
It has been observed that no fat people emerged from Auschwitz. But if some people do not lose weight below that level of privation, they don’t stand much chance of doing so voluntarily.
Perhaps they all died because they couldn’t access their metabolic reserves? [/devilsadvocate]
All the worse for those with great difficulty losing weight. Apply enough determination and it might kill you!
People would probably object to using Nazi data, but it would be interesting if they ever kept logs of inmates’ body weight and non-deliberately-caused deaths.
As you may have guessed, this isn’t the first time the subject has come up. Frustration builds.
It is not just the brain but the entire human body that isn’t specialized for an environment of plenty. Lowered food intake changes metabolism and energy expenditure, it doesn’t just make you crave more food.
That would indeed still be true.
That seems technically inevitable. The question then becomes whether this happens before or after your body enters a coma. (Or, more practically, whether valuable muscle mass is lost before undesired fat and whether the effect on fatigue and energy levels is debilitating.)
A ridiculously charged topic, how could I miss it?
We’re probably among the last generations (as in so many things) that need to bother with the now counterproductive and out-of-place esterification making us fat. Just as we managed to separate the carrot from the stick (the stimulus we get from an action versus the original evolutionary incentivized purpose of that action) with sex/procreation, so will we eventually be able to indulge in feeling satiated without actually storing unwanted lipids.
If not for too strict pharmaceutical standards, some (more) drugs would probably already be available. Given the large impact of diabetes, CHD and other obesity related diseases, even severe side effects in animal trials could be outweighed by the benefits. If not for the fear of lawsuits and strict regulations that throw promising drugs out of the pipeline prematurely. It took a decade and untold needless deaths for gene therapy to recover from a few mishaps to where it can be pursued again.
Regarding losing weight, personally I like the volumetrics approach, it’s easily combinable with most diets:
Feeling satiated is—mostly—a combination of mechanoreceptors in the stomach being activated (which is how gastric lap band surgery works) and various hormones reacting to e.g. rising blood sugar (hunger-stimulating Ghrelin gets inhibited), presence of food in the intestinal tract (hunger-inhibiting PYY is released, NB: it’s released more effectively by high-protein intake which would help explain the effectiveness of some high-protein diets, such as variants of keto), and leptin (released by adipocytes, can be mostly ignored, since obese people apparently have high leptin levels and a corresponding high leptin resistence).
Now, there is of course a latency between food entering your stomach and the mechanoreceptors triggering and PYY being released.
Therefore, a sensible measure is the following: Drink a large-ish quantity of water (cold and tasty with lemon) before every meal (to pre-load the mechanoreceptors), eat slowly (so you don’t eat more when your satiation signal is already in the process of being triggered), add taste-neutral e.g. salad to your normal food (I just pack my sausage sandwiches with large quantities of unprepared salad = little extra effort, halves the amount of high-caloric food I require before feeling completely full, mixed in it doesn’t even taste any different*.) Don’t drink caloric beverages, they’re not as effective calorie-for-calorie in distending your stomach, since fluids aren’t bulky enough.
These changes do not inhibit my “food experience” in any meaningful way, yet nearly halved my daily caloric intake.
It seems irrational to give up on dieting merely because the process can be complicated. That’s not a good reason to give up on FAI, so why should it be for dieting? The impact in energy levels between being obese and normal is redunculous, even if losing weight is hard, it still promises the most bang-for-the-buck in increasing your productivity, your self-image and your quality of life. Sure some have it easier or harder than others. C’est la vie.
* The trick is that salad-diluted sausage still tastes just like sausage, and still causes the satiety inducing effects on the same order as double sausage.
Obesity is interesting because I regard it as a partially-solved problem. For example, dinitrophenol would solve much of it: it makes mitochondria less efficient and so effectively increases metabolism, but at the cost of emitting waste heat—which is potentially fatal and got it banned despite its apparent effectiveness. It could still be safely used; it’s 2013 so electronic thermometers are a dime a dozen. Take patients to a fat camp, dress them in clothes with thermometers constantly recording and now doses can be adjusted based on detailed data and the thermometers can warn the patient to jump into conveniently located ice baths. Voila. And I’m not clear on how dangerous it really is when not made illegally and used recklessly by young kids; Wikipedia cites a 1934 paper as estimating that there were ~100k users of DNP before it was banned, and those authors remark, after discussing the grand total of 4 deaths up to that point due to the drug’s use under medical supervision, that
It would seem that like the War on Drugs in general, the attempt to reduce the harm from DNP has resulted in far more harm than was ever the case before.
It is frustrating to me that I want powerful cognitive enhancement, but instead of increasing metabolic efficiency, we’re decreasing it. Can’t we funnel those extra calories the the brain somehow?
I think most of the value in being thin is looking attractive and being able to be physically active. I think Eliezer doesn’t really need to be more attractive than he already is (4 girlfriends) and isn’t a huge fan of rock climbing or whatever. As far as I’ve heard, most of the health benefits of exercise can be gotten without needing to actually be thin.
I somehow doubt that all of this effect is due to thin people exercising more. ETA: looks like the ‘optimal’ BMI for women is larger than for men, BTW.
Considering the social import of being attractive and getting around to see people and the correlation between active social life and longevity I think it’s more than you might think, but I agree with you. On the other hand, the trade off is a lot more reasonable if you can be relatively healthy and happy while fat AND it’s particularly hard for you to lose weight.
Well, they did control for, among other things, marital status. (Also, I’d guess that the BMI that maximizes conventional attractiveness would be higher for men than for women, and wouldn’t depend much on smoking.)
Yes. ISTM that for certain people losing weight has become a lost purpose.
Higher for men? Despite the fact that women have higher BMI on average and that curvy figures are considered (by many) to be attractive?
Remember that BMI is based on the total body mass, and that muscle is denser than fat. (OTOH, that study corrected for level of exercise, and it’s quite possible that the BMI that would maximize an average white American man’s attractiveness if he’s not allowed to vary his level of exercise would indeed be around 20.)
I am one of those “many”, too, but ISTM that in present-day Western cultures we’re a minority; “thin” seems to have become a compliment. (Last year, someone offered to set me up with her roommate who probably had BMI around 18, and when I told her that I didn’t fancy her, she retorted “but she’s so skinny!” as though it was a positive.)
Ah, good points.
I don’t want to speculate about EY other than saying my model of him didn’t expect to ever see “too complex” brought up as a reason not to try anything. Maybe there’s too much overlap with HPMOR:Harry. It is possible that there are individuals whose akrasia levels / mental energy and self-image are unaffected by being overweight compared to not being overweight. Just unlikely.
That’s true, but doesn’t change that ceteris paribus given little exercise, you’ll still live longer not being obese. Note that I’ve not even mentioned exercise. It’s certainly better being overweight and exercising (while still being overweight), than being overweight and not exercising.
You’re reading “too complex” as “difficult” but eliezer means it as “something with inherently low priors for working”.
Pretty sure he meant “since this has inherently low priors for working, it’s too difficult to be worth it” ie expected utility is too low.
Expected value calculation > your ‘seeming’.
Because creating an FAI has (strictly!) greater expected benefits than one person successfully losing weight? To the extent that the rhetorical question is ridiculous.
This seems false.
I’d wager you’ve never been overweight. Strap a few dozen lbs around your waist or don them as a vest, see what it does to your daily routine. We’ve done that once in some class or other, and I’ve bordered the 30 myself as well, from time to time. You’re affected constantly, we’re as of yet embodied agents, not free floating minds. What’s the likelier explanation for a lack of action, expected value calculations or—here it comes - ‘akrasia’.
Depends from whose point of view. E.g. passing away in the knowledge that you’ve contributed to the eventual creation of FAI (which gives you fuzzies, or at least utilons) can be outweighed by living decades with more mental energy (which also contributes to your development efforts) and a better self-image.
That aside, that a task is a complicated problem/puzzle to be solved can be an incentive to solve it in and of itself, especially for certain kinds of people.
Assuming the increase in productivity, self image and quality of life (consider the metabolic syndrome, preventing decades of injecting insuline can have quite the impact on your QALY) to be fixed/constant for an individual, “true” or “false” does depend on how easy/hard it would be for that individual to efficiently attain and keep a lower BMI. For metabolically priviledged people, or just those with an easy to fix problem such as hypothyreodism, the statement is probably true. For someone who for whatever reason cannot lose any weight whatever he tries (within his motivational reach given his current energy levels … there’s a catch-22 present), it would be false.
Eliezer has mentioned many of the things he has tried to lose weight (including ketogenic diets and even clenbuterol). I’ve tried all those he has mentioned. The difference is for me they work (I call it ‘cutting’ and can merrily play around with my body composition all sorts of ways). But if, like Eliezer, I had expended huge amounts of effort and my body did not respond significantly then I would update my expectations.
Things that are expected to fail have low expected value. Sometimes you need to shut up and multiply instead of shut up and do the impossible.
Expected value calculations. Unless you are making accusations of lies—outright fabrication of self reports.
For the purpose of declaring an accusation of irrationality false the relevant point of view is Eliezer’s. If Eliezer had someone else’s values then it would make sense to evaluate the rationality of a given choice for him according to those other values.
Yes (or at least it would be up there on the list). It just isn’t true in this case.
Pretty sure most people involved in FAI efforts are fatoring in more than warm fuzzies in their EU calculations.
Well, I missed it; I kind of wish the LW inbox included replies-to-replies. In any case I find myself thinking this shouldn’t be a charged topic, even though it clearly is. As a culture we’re still hung up on the whole self-indulgence-as-moral-failure absurdity.
A thousand times yes. Sex without babies is a nearly-solved problem, lacking only a male equivalent of the IUD. Weight control should be just as solved, and no, self-control in food intake is not a solution even if it does work. Like condoms, it’s a badly suboptimal necessary evil. I should be able to eat whatever I feel like while maintaining whatever weight I damn well please.
That this has failed to happen despite the massive amount of money thrown at weight-control products says something, but I’m not sure what. Has anyone made a serious, prolonged attempt not at “getting people to ‘stay good’ around food” but at “decoupling food from fat entirely”? If not, why not?
May I direct you to the subsequent paragraph in the grandparent? :)
Sort-of-valid point. But all drugs I’ve seen or heard of have been essentally appetite suppressants plus varying levels of bullshit. I wouldn’t expect metabolic decouplers to be disproportionately cut out by regulation; if there was substantial research in that area, I would expect some of them to be on the market.
(it is possible some are that I’m unaware of.)
Keto isn’t a particularly high-protein diet. It’s generally a high dietary fat diet. I mean, it can be high protein, but the general idea is high fat.
Corrected. Although, it’s somewhat hard replacing carbohydrates without also increasing the overall protein intake. There are probably variants of keto that fit, and variants that don’t. There’s apparently no catchy phrase for the overall “high protein diet” reference class.
Yeah, most people who do keto or other low-carb variants tend to eat a high-protein diet. But for fat-loss something like twice as much weight of dietary fat as protein is recommended along with sub-30 grams of non-fiber carbohydrates. And the medical keto diet, iirc, was something like 4:1 fat:protein. It’s a lot easier to hit the protein macros than the fat macros.
Fair. I’d seen some of the previous conversations or I wouldn’t have responded as I did; but I’m guessing I missed occasions on which Eliezer demonstrated that his body chemistry was provably not doing what theory would predict it should (with regard to ketosis), which I find much more convincing than “I tried X, Y, and Z and none of them worked.”
If such occasions exist, then I just plain missed them and I’ve stepped on toes unnecessarily, and I apologize.
I’m going to guess “before muscle loss or coma”, on the grounds that fat is supposedly for long-term energy storage, and I would expect “break down muscle and/or go into coma in preference to using stored energy” to be maladaptive even in the EEA. I have no controlled study to support that guess, however.
I could see “after excessive fatigue,” I suppose; that might be physically necessary (fat burning is slow, IIRC) but not so maladaptive as to defeat the purpose of storing fat.
This seems somewhat unfair. There are a handful of diets that work on a broad variety of people, such that the prior any one will work for a particular person is higher than a novel diet like Shangri-La. And so unless you’ve tried slow carb/ketogenic, intermittent fasting, 30g of protein for breakfast, and ECA stacks and none of them worked, it seems like you’re updating too far in the direction of “all diets don’t work for me” from the evidence that “diet X didn’t work for me.”
(The only one of those I’ve tried is IF. It worked for me.)
What’s the difference between this and the claim that you are a metabolic mutant?
It might be worth going to a sleep doctor; sleep apnea can really fuck up your metabolism, not to mention causing unbelievable akrasia. I would say sleep tests are a GOOD THING, something everyone should do. I had sleep apnea for years. It was like some eldritch monster was sucking away my willpower and I wasn’t even aware. Within a few months of getting my mouth guard, which keeps my tongue from blocking my airway while in REM, I lost thirty pounds and gained an enormous well of mental stamina. A small minority of the “metabolically challenged” may just have undiagnosed sleep problems.
Since it was cheaper than a sleep study, I bought a self-adjusting CPAP on Craigslist and just tried it. Nothing miraculous occurred.
A history of diets—please note the repetition of ideas and lack of effectiveness for people in general.
Well, there’s your problem right there—it sounds like you didn’t really especially “do” a ketogenic diet, if you never once reached ketosis (as I understand, you typically need to deplete glycogen stores before entering ketosis, and so you might have 3,000~6,000+ kcal to burn through first, and then there’s an interim period of glycogen depletion before it actually begins. Which is where the magnitude of your caloric deficit becomes relevant; was it large enough to get through your stored glycogen in the 1-2 weeks you ate no-carb? I think recently consumed food is generally metabolized first, before dipping into glycogen stores, though that’s likely a huge oversimplification), any more than someone eating an egg for breakfast can be said to have “tried” keto.
And even with weight as your metric you should have seen some noticeable reduction after a month and a half of “dieting” on a >300 kcal deficit if you were consistent in your measurement conditions (eg, every day after waking/bathroom and before breakfast). Maybe you just overestimated how many calories you needed for maintenance each day? Which is quite common among “dieters”. Like I mentioned, the usual advice for muscle preservation is to eat 300-500 kcal below maintenance each day, and if after a month you fail to note any weight loss or note weight gain to reduce your consumption by an additional 300 kcal, rinse and repeat. It might take several months to note any reduction in weight if you had a shoddy initial estimate (or if your metabolism is exceptionally sensitive to intake, though that can only account for so much. Eventually as you progressively reduce consumption [or increase activity!] you will lose weight).
Haha, and many people report a “mental fog” when first trying ketogenic diets. Maybe that’s why you can’t remember :D
disclaimer: this isn’t my field of study and it’s been several years since I tried keto and my research then was cursory, at best, which is why I used less-than-confident language in the above. In the three months that I ran a cyclic ketogenic diet I think it took 2-3 weeks to make the ketostik turn purple (or whatever) the first time, then a weekish after that, ulitmately down to 2-4 days after each weekly recarb.
edit: also, I’ve heard it might be useful to think of diets less in the “dieting” sense (a temporary change in eating behavior that will end once you reach a predefined goal) and more in a “biological” sense (the typical assortment of food that an organism habitually consumes). Less temporary, more permanent or semi-permanent lifestyle change.
After numerous previous failures, if it’s that complicated I’m not going to bother. Complicated things seem even less likely to work than simple things, and simple things almost never work in the first place.
In my experience, no matter what you try, there’s always an excuse when it doesn’t work. Then when it still doesn’t work there’s something else you’re not doing exactly right that they forgot to mention earlier. Oddly enough, when something does work for someone, nobody bothers to check to see if they were doing everything exactly right by way of confirming that all these extra frills are actually required as opposed to just being excuses that are only invoked when it doesn’t work because, in reality, metabolisms are different.
Anyway, not interested. Thanks for trying.
Do you have any data on your eating and exercising habits? I’d love to know what or how your eating now. If you’re eating the standard american diet, then you’re definitely doing something wrong, and there are a few simple things you could do to at least eliminate deleterious factors. Do you drink Coca Cola everyday for example? What’s your baseline diet? Before ever trying a new one, find out what’s wrong with the baseline and tweak it.
It is obvious if you weigh yourself every day for a couple months or longer and you know how to do stats.
(FWIW, my weight since 12 February fits to a straight line a + bx where a = (93.74 ± 0.19) kg, b = (−0.018 ± 0.007) kg/day, and x is the time elapsed since 12 February; the RMS of residuals is 0.68 kg. Approximating the posterior pdf of b as a Gaussian, which ought to be close enough given 46 degrees of freedom, I’m 99.42% sure that b < 0.)
Haha, well yeah. Though you should hardly need stats if you’re recording over a period of months (“golly, I wonder if my 40 lb weight change these past 6 months is just me being dehydrated right now? Maybe I should wait till after I drink my morning 4 gallons just to be sure”). I meant it more on time scales of “between 1 week and 2 weeks”, or for where weight loss was very minor due to a tiny caloric deficit.
With more precise measurement (eg, via bodpod) of body composition you would better be able to track smaller changes, too.
Excellent point.
I suspect you’re basically correct, but I would not take the stats results at face value. There are many possible problems resulting from the physical and electrical properties of the scale you’re using, that I would not expect to be well behaved in a stats sense. In particular: quantization errors, non-linearity / non-monotonicity of the scale A/D converter (depends strongly on type of A/D used), temperature dependence of both the scale strain gauges and A/D, etc.
The general rule here is that trying to get too many more bits of precision out of a measuring device than it is intended to provide is tricky.
You could calibrate the scale in a number of ways; easiest would probably be to check that it gives consistent readings over time for a fixed weight that’s not too small compared to you. You could simply weigh the fixed weight, or you could weigh you and (you + weight).
You’re right, any time-varying systematic error (due to temperature, ageing of the scale, etc.) would screw up the analysis. (Quantization errors shouldn’t matter that much so long as they’re much smaller than day-to-day fluctuations.)
Ketogenic diets didn’t work for me. In fact, they were one of the worst diets I’ve ever tried.
http://paleohacks.com/questions/25449/why-do-people-consider-ketosis-stressful-to-the-body
Worked for me, though I have been plateaued for the last year a few pounds and several waste inches above where I’d ideally like to be. Haven’t figured out how to bust through that yet.
I strongly suspect that when we do figure this out, ketogenesis will be a large part of any eventual solution, but I don’t think we have all the answers yet.
I do know folks for whom low carb diets failed, but in all cases I’m personally familiar with that’s because they couldn’t maintain them. (which is a strike against low-carb diets, of course; an unmaintainable diet isn’t useful) I know of one case of someone who took off 150+ pounds using low-carb and then put 50 or so pounds back on while continuing low carb. Still he never came close to getting back to where he was pre-low-carb. I’m curious if anyone has simply failed to lose weight while maintaining a low carb diet. I think Atkins talked about this possibility in the last edition of his diet book he authored. I’ll have to look up the citation, but there are some medical conditions that can cause weight gain/prevent weight loss that show up in a few percent of patients.
Err, have you been lowering calorie intake relative to your activity and changing metabolic rate? Lighter bodies require less energy both to maintain and move around. If you haven’t been adjusting your “dieting” diet it’s no wonder you plateaued, because where initially you ran a caloric deficit you’re now much closer to equilibrium.
Also, I’ve intentionally gained weight on cyclical low-carb diets (<5 g carbs each weekday, 1200-1500g carb-up on weekends). It’s because I ate a lot.
What do you think of the research described at https://slimemoldtimemold.com/tag/a-chemical-hunger/ regarding lithium?
If I may ask, what was your starting weight; what is your current weight; and how long have you been at your current weight?
when I started keeping track 479 days ago I was 270, right now I am at 228. This isn’t pure keto, because at first I was doing slow carbs, and then when I switched to no carbs I kept the tim feriss style cheat day. I’ve also cheated on various trips and so on I’ve taken, but in general I have 6 days on 1 day off diet. I’ve gone down something like 4 pants sizes. Right now I’m trying to transition to fewer cheat days or none at all because I feel like I’m plateauing, but it’s rough.
Do you get keto flu type hangovers after your cheat days?
A little? I tend to feel really heavy and bleh at the end of cheat nights and kinda weird the next day but it’s not really what I would call “keto flu”
Thank you for sharing. Are you willing to follow up in a year or two to let me know where you are?
if you want to ask again in a year sure. Email me or something in a year. Drethelin@gmail.com
follow up plz?
lost some progress, up to 238.6 as of this morning, which is down from getting back up over 240. I stopped tracking regularly for a while, and now I am again and I don’t know why but tracking on a daily basis seems to have some sort of causal effect making me lose weight again.
I haven’t been keeping the closest track, so this isn’t even a good anecdote. I was a 165 lb and 5′10 man going out of high school seven years ago, and went up to 220 as of a year ago. It wasn’t all fat (I’m significantly stronger now than I was then), but I did go from wearing 34 waist jeans to a 38.
So I started at 220 and a 38 waist, and I’ve gone down to 180 and I think size 34 jeans (I haven’t gotten around yet to buying more jeans, but my size 36′s need a belt now). Exercise involved hasn’t been too extensive—mostly just walking and housework. I’m still work-in-progress, but I haven’t had a history of my weight yo-yoing on diets, and I intend to stay on keto for the foreseeable future.
Thank you for your response. Will you agree to update this response in a year or two?
Yeah, I can do that.
Great, thanks. I have been doing informal research on diet and weight loss for a while now and I think there is value in following up with people a year or two down the road.
Update: I’m at pretty much the same place now as I was then. Dropped the keto diet since I was happy with where I was. Still fairly active but not hardcore about it.
thank you
Surely the more obvious solution is “eat less calories than you use.”
The calories-in calories-out model is attractive, but it doesn’t appear to be all that accurate, or at least it’s incomplete. The body responds differently to different foods. They might have different effects on various hormones (e.g. the ones that regulate hunger), and they might be broken down and redistributed in different ways. In one study (Kekwick and Pawan), three groups of people were put on 1,000 calorie diets of 90% fat resp. 90% protein resp. 90% carbs. The first group lost 0.9 lbs / day, the second group lost 0.6 lbs / day, and the third group gained 0.24 lbs / day. (I don’t know to what extent the study controlled for exercise but I think it’s safe to assume that the difference in the amount of exercise that each group did wasn’t large enough to explain these results.) As Tim Ferriss puts it in The 4-Hour Body:
In the context of solving the specific problem of fat loss, one goal is not to lose muscle, and what you’re eating should affect how easy it is to target fat and retain muscle as well.
And, of course, diets don’t work if they can’t be maintained. The fact that different foods affect the hormones that regulate hunger differently (some foods even make you hungrier!) means that different diets, even with the same number of calories, require different amounts of willpower to maintain.
I’m heartened that your comment is so well-liked. I made the same point a year or two ago and got back a bunch of nonsense about how the second law of thermodynamics cannot be violated.
It’s interesting that the second law of thermodynamics is what people use here. As long as your metric is losing weight, the relevant physical law is conservation of mass, and starting from conservation of mass helps clarify the issue enormously, I think. (Apparently the mechanism by which burning calories actually causes you to lose weight is exhalation of carbon dioxide. I feel like I knew this once, but forgot and only very recently relearned it.)
Probably a brain fart for the first law?
Carbon Dioxide and water are the two main byproducts of fat metabolism, which made me really happy to learn because it basically killed my worries about focusing on bowel movements. You literally can sweat the fat away!
Probably not as much as excreting all the unprocessed food from the other end.
That’s not true. Most of the material in bowel movements was never in your fat to begin with.
Indeed not from burning calories, sorry. However, it still has to be subtracted from the intake, just like the amount you exhale, so reducing absorption is just as important as increasing burning.
Is this actually true? Do you have a citation for this?
Actually, I could not find any data online on the food energy utilization in the small intestine in humans and on the factors affecting it. Apparently the obvious ways to prevent absorption, like laxatives, don’t really work in the long term.
I don’t think so, because calories are a unit of energy, so a simple calories-in calories-out model would necessarily model energy balance as energy intake (through food) and energy expenditure (through body maintenance and activity). Your thermodynamic/energy balance is what would ultimately determine either the anabolism or catabolism of different tissues (a more complex calories-in calories-out model, rather than the simpler one mentioned, would have the greatest explanatory power, I imagine. Metabolic rate is under hormonal control, hormones interact in complex ways, and intake of different foods and different activity patterns can alter hormone expression. A calorie-budget model incorporating varying hormone expression/sensitivity, genetics [for both little things like minute differences in receptor molecules and bigger ones like hyperthyroidism], and different metabolic pathways used [as dictated largely by the previous two], would be pretty accurate, I reckon).
Weight can also be stored as different things, which is why it’s not the best proxy for the success or failure of a “diet”. Different substances (like fat or muscle or glycogen or water) have different energy densities, so not all changes in weight signify the same thing (presumably, a “dieter” wants to lose fat. Drinking a gallon of water might cause his weight to rise beyond where it was a month ago, but that does not mean that his “diet” has been shot, or that he has gained fat). A study that looks only at weight change and not change in body composition under different conditions would enormously simplify what that weight represents—in the study Qiaochu_Yuan mentions, 90% fat diets might have gained fat and lost muscle, water, and glycogen, resulting in a net weight loss, where the 90% carbs group gained muscle, glycogen, and water, and lost fat, resulting in net weight gain. The second case is obviously preferable to “dieters” than the first.
(what I suspect happened in the study is that the first two groups were depleted of glycogen, an energy source the body tends to use before it starts catabolizing fat or muscle deposits. Glycogen also tends to increase water retention, further inflating the weight of high-carb dieters. If they put all three groups on a 0-carb diet for a week to deplete glycogen stores and then dehydrated them, I’d suspect their net weight loss to be much more similar (changes in muscle and fat would vary [again, because of hormonal interactions and different energy densities], but bodies seem to need a lot less protein for maintenance than often suggested so unless the non-protein groups were consuming mostly carbs/fats for their 10% I doubt there’d be a huge difference)(also, the fact that obese individuals might gain muscle/fat on a 1000 kcal diet is obviously silly. If they were using the [again, stupidly oversimplified] BMI to define obesity [where obesity occurs at BMIs>30], a 5′10″ man would have to weigh >209lbs, giving him a BMR at minimum of roughly between 1500 and 2000 calories, which is a good bit more than 1000kcal, so he’d still probably lose weight [but ignoring changes in BMR from hormones and such] completely sedentary)
There’s also calorie type influencing activity level (eg, People starting ketogenic diets often report lethargy that eventually levels off, but whether that’s a true return to past alertness or a subjective change in perceived vigor I’m not sure).
I don’t understand what you’re disagreeing with. Is it “as long as your metric is losing weight, the relevant physical law is conservation of mass”? Because that seems obviously true to me. What you seem to be arguing is that your metric shouldn’t be losing weight, which is reasonable, but you’re not disagreeing with me.
Huh, I think I read your comment too quickly and missed the “as long as...: qualifier and then started replying and went off on a tangent and forgot what the original comment was. Hah. My bad. Also didn’t notice your name, hence my reference to you in the 3rd person.
Yeah, weight’s not the best metric to use without taking into account body composition.
Oh wait, I think I figured it out. I’d combined your post and paper-machine’s in my head, so I thought the simple calories-out-calories-in model in the highest level post being the thing referred to by discussion of conservation laws.
I don’t understand what the first sentence is disagreeing with.
I noticed the same difference in response myself and was similarly pleasantly surprised.
Due to the evolution of general sensibility or the Karma power-law?
No idea. Probably just random chance. The parent of Qiaochu’s comment is now at zero, so “evolution of general sensibility” is slightly less likely.
I noticed I was confused. This doesn’t seem consistent with the results of the Minnesota Starvation/Semistarvation Study. I went to Wikipedia.
My prior consider it quite ludicrous that you can gain weight eating at a 50% deficit, no matter what your macros. The criticisms seem reasonable enough to explain the effect.
Note that the link in the citation claimed that when told to cut out carbs and eat as much protein and fat as they liked, “In all subjects, there was a reduction in calories ranging from 13% to 55% during the time they were consuming the low-carbohydrate diet.”
Thanks for looking this up! Regrettably, I did not notice that I was confused.
So long as your diet isn’t nearly that lopsided, IME (YMMV) the calories-in calories-out is a more decent first-order approximation than many people realize. See also The Hacker’s Diet. Second-order effects exist, but they’re second-order effects.
That’s basically the point.
Places way too much focus on losing weight. See parent; losing weight by losing muscle mass isn’t desirable.
Your claim here hinges on the presumption that CI and CO are the only first-order effects, which is almost certainly false. Age, body fat proportion, maximal oxygen uptake, etc., are plausible candidates that I’ve seen in mathematical weight models.
In my experience, these tend to be taken into effect when calculating the “calories out” part of the equation. By what mechanism were you thinking that these mattered, that’s not “calories out”?
Calorie use is not constant, and varies with calorie intake and type, among other things.
Sure, but as you probably know, calories eaten are stored into multiple deposits, depending on various internal and external factors. If the availability of those deposits varies, it might not be so easy to deplete most of the calory intake, short of completely starving oneself.
Plus you also have to deal with massive willpower problems.