Comment by ninkendo | Hacker News Reader

Comment by ninkendo | Hacker News Reader

Parent Full thread

It sounds like any other of these diets that tell you when and how to eat but not how much. I’m pretty sure I could gain weight when doing everything you describe, if I eat really terribly during the days I eat “plenty of carbs”.

The key to CICO is that you’re actually measuring not just the calories you eat, but the weight you gain/lose, and are continually calibrating. If my diet jacks up my leptin levels such that I don’t lose weight, that will show up on the scale, and I have to adjust my calories down to compensate. With this system I am guaranteed to lose weight so long as I don’t cheat. This is trivially true: the calibration could technically push you all the way to zero calories per day (which will never happen in practice, obviously.)

Case in point, I lost 50 lbs with this approach, eating McDonald’s nearly every single day. I kept it off for roughly 8 years. (I since became a dad and have let myself go tremendously, gaining ~30lbs back, but I will cross that bridge soon.)

>If my diet jacks up my leptin levels such that I don’t lose weight, that will show up on the scale, and I have to adjust my calories down to compensate.

>With this system I am guaranteed to lose weight so long as I don’t cheat.

Unfortunately, that's just not true. Here's some great quotes from the book "The Obesity Code" (which I've just recently found and find an excellent book):

>The abrupt increase in obesity began exactly with the officially sanctioned move toward a low-fat, high-carbohydrate diet.

>A 30 percent reduction in caloric intake results in a 30 percent decrease in caloric expenditure.

>Total energy expenditure is the sum of basal metabolic rate, thermogenic effect of food, nonexercise activity thermogenesis, excess post-exercise oxygen consumption and exercise.

>Since hormones control both Calories In and Calories Out, obesity is a hormonal, not a caloric, disorder.

>If hormones regulate fat growth, then obesity is a hormonal, not a caloric disorder.

>Sugar will increase the blood glucose level and provoke an insulin response from the pancreas. Olive oil will not...[cause] significant increase in blood glucose or insulin. The two different foods evoke vastly different metabolic and hormonal responses."

All of this is to say that it's not enough to simply adjust calories. Yes, calories matter to a point, and everyone knows that starvation (e.g., 0 calories for an extended period of time) is effective in the long-term, but starvation will also harm your body, and no reasonably-minded person can suggest it as a viable approach to weight loss.

The best available scientific evidence shows that continuing to decrease caloric intake can make weight loss more difficult for an obese person.

whoosh

That was the sound of you zooming right past my argument and pretending I was saying something I wasn’t.

I’m not saying CICO is a good idea. I’m not saying it’s healthy. I’m not saying it’s easy. I’m not saying it will have a high chance of success vis a vis people actually being able to stick with it. I’m merely saying IF you do it right, THEN you will lose weight. This is trivially and tautologically true (to the point of CICO being useless to most people!), but it’s extremely tiring having to explain it over and over that it still works despite all of this.

To put it another way: there are two operative definitions of “works” for a diet:

1. Will it work if you do it exactly as directed?

2. Is a typical dieter likely to have success by attempting this diet?

CICO fails miserably at 2. It passes with flying colors at 1. Its best not to conflate the two definitions in the middle of an argument because you end up talking right past one another.

This is something I've said repeatedly in other places and this is a certain segment of people who can't understand it.

CICO is, especially at the population level, terrible dieting advice. We have decades of experience and oodles of research showing that, as advice it does not work.

However, it is, as you point out, basically tautologically true/correct. For CICO to _not_ be correct would essentially be breaking physics and generating mass/energy from nothing.

Yes, your body can change it's energy expenditure in response to dieting, but this is accounted for in the CO part of CICO. If you consume fewer calories than your body expends (in all the myriad of ways that your body can/does expend calories), then you _must_ lose weight.

The fact that CICO is true does not make it good advice but the fact that it's bad advice doesn't make it false.

I think of it this way: CICO is an axiom we have to assume in order to figure out why a diet isn't working.

E.g. if your input is 1200 calories and you work out, but you aren't losing weight? Maybe your NEAT is way down (or you have a thyroid issue). You are super active but can't lose weight? Maybe you aren't tracking food calories accurately, or you need to eat more satiating foods.

CICO isn't a strategy, it's part of the landscape and your diet is a strategy to navigate that landscape.

But no matter what your particular strategy, it is always going to involve eating less hyper-pallitable foods and getting more activity. That may not be all, but you cannot go from obesity to a healthy weight without those things.

CICO makes me pay attention.

Pay attention to what I eat and how often I exercise.

I rarely actually count, but I do lower calories when I look at a label and say OMG. Like, who knew a shake was 2000 calories? Hmm, fries only 500. I just switch and cut my calories by 1500. I’m so good, I deserve these fries! Yum.

Well if CICO is not a good idea, not really healthy not really applicable to most people, then why bring it up when we are talking about solution to obesity ?

The main problem with CICO is that it is confusing the description of a process with an intervention designed to modify said process. Both things a very different.

> I’m merely saying IF you do it right, THEN you will lose weight... > that it still works despite all of this...

That's the first part, the "works" include the IF... You mixing efficacy and effectiveness is here.

> 1. Will it work if you do it exactly as directed?

Try to apply CICO to water and hydration level : You can modulate your hydration level by driking more or less. BUT, you can go above a certain level or below a certain level independent of how much you drink.

The problem with CICO's is you fail to realize that CI and CO are not the control variables and you might be getting the causality wrong.

Much in the same way that you can change the temperature of a room by changing the value on a thermometer... You can't change the body set point and fat disposition by "just" changing CI/CO.

> Well if CICO is not a good idea, not really healthy not really applicable to most people, then why bring it up when we are talking about solution to obesity ?

You may want to read my first post again. I spend nearly the entire post agreeing about the problems with CICO. But in my last paragraph I explain that I still like it and that it works for me, and explain why. YMMV.

> That's the first part, the "works" include the IF... You mixing efficacy and effectiveness is here.

Right, that’s called propositional logic. It’s perfectly valid to make “CICO works” predicated upon whether you follow it correctly. This isn’t as trivial it sounds: there are diets for which this is not true!. I can totally stick to a low carb diet to the letter and still gain weight, because if you don’t include a calorie limit, you can still go way over your calorie expenditure even with low carbs. As trivial as my “definition 1” is for “works”, it’s still a bar that many diets don’t pass.

Well the point i am trying to make is that your explanation on why it work for you is probably wrong, and even more the advice you generate from your theory can be dangerous and counter productive.

> Right, that’s called propositional logic. It’s perfectly valid to make “CICO works” predicated upon whether you follow it correctly.

Well no. And that really the core of the issue. I see that you didn't answer to the example i gave you.

But if we want to be pedantic :

1 - There is no definition of "works" in propositional logic.

2 - Here what i understand from your logic deductions :

  A - CICO can be rephrase as : For a given individual , for every time duration t : delta_weight(t) = CI(t) - CO(t)


  B - What you are saying is  : for a given individual , for every time duration t : delta_weight(t) = CI(t) - CO(t), THEREFORE, for every delta_weight, there is a time duration T, and CI(t) and CO(t) such that delta_weight = CI(t) and CO(t);

And my point is that B is false (empirically), and does NOT follow from A (logically). And to repeat myself you can try this game with room temperature and thermometer readings.

Your understanding of CICO (or at least the point I am making about it) is wrong in your “B” paragraph.

My argument boils down to the following:

For every individual, at any given time, there is a metabolic rate M for which, if their calorie intake was larger than that, they’d gain weight, and if their calorie intake was smaller, they’d lose weight. Crucially, M changes over time.

But going CICO properly means using the scale to measure M, every day, by diffing your expected value of your weight loss given how much you’re eaten, with the actual value on the scale. You now have an updated value for M to use in calorie counting.

Since M is determined by empirical observations of your weight loss/gain in response to the calories you have consumed, you will by definition lose weight if you consume less than M calories. This is the argument that CICO must work by definition. It’s a logical contradiction for it not to work.

> Your understanding of CICO (or at least the point I am making about it) is wrong in your “B” paragraph.

Just to clarify B is not cico, it's the inference that you seems to make from CICO.

> For every individual, at any given time, there is a metabolic rate M for which, if their calorie intake was larger than that, they’d gain weight, and if their calorie intake was smaller, they’d lose weight. Crucially, M changes over time

This is "not" the definition of CICO, and again does not follow from CICO. CICO is fundamentally a thermodynamic statement about the observed state of a system. You are making predictive statement about a possible future state of the system.

Again, if you want to believe that statement that's fine. The point i am making is just it's does not follow from cico, no have i seen anything empirically that validate that statement.

> Since M is determined by empirical observations of your weight loss/gain in response to the calories you have consumed, you will by definition lose weight if you consume less than M calories.

sure this derived from your "definition" of CICO... but your definition is not thermodynamically derived...

> This is the argument that CICO must work by definition

the premise (the definition of the argument) is flawed.

> It’s a logical contradiction for it not to work.

Not really, CICO not working (as in this case) can be just the result of sound logic based on bad premises.

Reply on news.ycombinator.com