How accurate are Americans at estimating calories? | Hacker News Reader

How accurate are Americans at estimating calories?

survata.com

37 pointsckelly77 comments

Sure, these probably aren't necessary for people whose parents taught them proper portion size and diet control. But some of us grew up with parents who still fight eating disorders. Two tools were instrumental in teaching myself how much to eat:

1. Calorie counting app. I use MyFitness daily and record everything in it. The app gives me one number -- how many calories are remaining for today -- on which I can focus and remember. Other apps are probably as good, I've never tried. This really helped me learn how to control my portions and stay honest with myself.

2. Calorie counts on menus. When looking at fast food, I'll remember how many calories are remaining for today, and I'll often change my choices. I've chosen the 300-calorie cheeseburger over the 600-calorie double cheeseburger many times, mostly because the difference was listed on the menu.

One of my goals was to change how I thought about food -- instead of being a magic elixir, it became a game: I have 2000 calories to spend today, but if I spend 600 calories on this burger, I won't be able to afford the 400 calories of beer I want to drink tonight. This is because I've already budgeted 1600 calories across my other meals. I've lost 80 lbs this way and maintained this lower weight for months.

Congrats on meeting your goals!

That sounds like a great idea and I've been looking for something similar to try and lose weight. But if you eat out a lot and the calorie count is not written on the menue, how are you going to record it? For example, if you order chicken with rice, that may be a fairly healthy choice except if there's to much cream and butter in the sauce.

Eating healthy definitely is hard. Also can you link to the app? My google-fu is failing me.

Calorie estimates are never perfect, but a rough estimate is much better than nothing.

Looks like http://www.myfitnesspal.com/apps

I think the fact you're looking at fast food menus could be the problem. Cook food from scratch -- it's fun and healthy.

Don't assume as much, as recent as today: http://www.bbc.co.uk/news/health-20713985

A chef meal will be different from a home-cooked one. Chefs are known to go for things like lots of fat (which acts as a base for spice flavor), sugar, caramalizing (to improve taste) and other things that add taste but also calories.

My mom cooked from scratch. She fried chicken, put lots of butter in the mashed potatoes, put butter in vegetables, always had gravy, etc. But "it was ok because they're low-sodium low-cholesterol heart-healthy home-cooked recipes".

Just cooking from scratch by itself doesn't solve the problem. You have to ensure you're using healthy ingredients.

IMHO, calorie counting is not as important as carbs and fats counting on menus.

One of our greatest problems with dieting is that calories are not really equal and the effect food has on you really depends a lot on your particular metabolism. Maybe if you eat more than 2000 daily calories worth of stake and diary you get fat, but I can eat more than that without any effects whatsoever, while a single cookie ads 2 pounds on me (my body retains a lot of water after eating sugar).

And it's a problem because the common advice given to people doesn't work so well. Which is why dieting hasn't really caught on.

Personally I separate foods in dangerous (for me) and non-dangerous. I measure my weight every morning and I know precisely what types of food lead me to gain weight. For the non-dangerous kind I have no restrictions and eat as much as I please, as long as I don't combine it with the dangerous kind ... whenever I do that, I starve myself a little (for a day or so, eating some fruit and a yogurt), then it's back to normal.

Btw, I lost 44 pounds this year after cutting my sugar intake drastically + a dissociated diet that lasted for 4 months. Right now I'm happy that the diet is over and my weight has stabilized to the point that I eat whatever I like, except snacks (which I never liked anyway) and sugars. Of course I don't overeat, because I don't feel the need to do that anymore - somehow giving up on sugar has cut into my appetite.

Calories are the only determinant as to whether you gain or lose weight. It's thermodynamics. 3500 calories deficit or surplus is approximately equal to 1 lb of body weight lost or gained.

Your body burns X calories per day to feed itself. If you eat less than this number, you will lose weight. Period. If you eat more than this number, you will gain weight. Period. Your TDEE or Total Daily Energy Expenditure can be easily estimated if your are so inclined, but it isn't necessary. Want to lose weight? Start at 2,000 calories per day. Monitor your weight across 4 weeks. If you aren't losing at a rate you're happy with, drop it to 1,800 calories per day. Repeat.

If you want to lose 5 lbs, you need a net deficit over time of 3500 * 5 = 17,500 calories.

Good/bad foods play a different role in your overall health, for example by impacting your level of water retention or satiety.

I wish there were some way other than the impolite downvote to register just how miserable believing this broken logic made me for a decade.

Can you expand on that? Calories in/out has lost me a net 30 lbs.

Agreed -- the method I explained was just calories in/out. The problem I had was that I didn't have enough data. The app and calorie counts on menus give me the data.

I lost 44 lbs on a disassociated diet, which is now over and my weight is stable by abstaining from sugar, as otherwise I only try to eat healthy, but I do not count calories or whatever.

Congratulations on your 30 lbs loss. Unfortunately not many people can abstain from eating. For some people it's a source of depression.

What works for you doesn't necessarily work for somebody else - and have you noticed how some people can eat whatever they want without getting fat? Freaking bastards.

>I lost 44 lbs on a disassociated diet, which is now over and my weight is stable by abstaining from sugar, as otherwise I only try to eat healthy, but I do not count calories or whatever.

It doesn't matter if you count it or not if you lost weight you were taking in less energy than you were expending.

>What works for you doesn't necessarily work for somebody else - and have you noticed how some people can eat whatever they want without getting fat? Freaking bastards.

Those people are a myth and it's been proven, they just consider "eating a lot" to be less than the amount needed to get them to gain weight. I mean I could eat 2000 calories of chocolate everyday and aside from the massive problems related to not getting enough vitamins/macros I'd still be losing weight.

Anyway congrats on the weight lost, we're all gunna make it bro.

Really? Then couldn't one subsist by just eating one or two bars of chocolate per day and some vitamin pills? Or maybe a liter of coke per day would work just as well? It would be a very cost efficient way of eating as both candy and vitamins are cheap around here (well, not vitamins but a bottle of pills last a long time).

I've always assumed that that wouldn't work which is why I'm skeptical to your "it's all about the calories" argument.

Well chocolate doesn't actually have that much to it (a regular chocolate bar is only like 250 calories) so it'd have to be more than two bars a day.

The main problem (as I'm told) with eating nothing but pure calorie sources supplemented with...uh, supplements is that the cost of the supplements is very high. Remember you need protein, carbs(I guess these wouldn't be in short supply on the candy bar diet) and fats plus micronutrients too or your body will start to break and those supplements aren't cheap.

I do know people that subsist mostly on protein shakes, rice and various supplement pills when they are doing hard cuts for body building competitions and while it's stressful on your body you aren't going to die from it.

Sure. Here's the classic example: http://www.cnn.com/2010/HEALTH/11/08/twinkie.diet.professor/...

Thanks! It's nice to be proven wrong sometimes.

The human body is not a bomb calorimeter. Simply cutting calories in (and running to try to increase calories out) reduced my energy and crushed my mood. My body compensated for lower input with lower output.

Cutting out sugar and grains, on the other hand, improved all of those things, and dropped 50 pounds off me despite actually cutting out the running as well. I never had to count anything; without the broken satiety mechanism, I'm not going to accidentally eat 20 oz of steak like you might accidentally eat a half a pound of spaghetti.

This is the point where you say "ok but that's just a trick to reduce your calories in," but that's entirely the point: calories in and out is descriptive, not prescriptive. Thermodynamics says that if I am losing weight I am consuming less energy than I am expending; it does not say that if I consume less energy that my expenditures will remain constant and fat will disappear.

The major catch there is that, if some common hypotheses hold, then where a calorie comes from has a direct affect on how your body burns it. The example commonly used is to say you need and consume 3500 calories to support your current level of activity. If you consume 600 that your body will only store, then your body will behave as though you still need an additional 600 to maintain your energy level. Likely leading to you being lethargic until you get additional food. (Which, as stipulated, you don't actually want/need to consume.)

Not really. Obviously, preservation of energy must hold, but body weight isn't just a function of calorie intake because metabolism is regulated by hormones, which affect the rate of calorie burn and fat tissue formation. Human metabolism is far from being completely understood.

Then why do type 1 diabetics only count carbs? A diabetic can eat as much protein, fat and fiber as they want, but have to match insulin levels to their carbohydrate consumption (except fiber).

I'm highly simplifying, but carbohydrates are converted to blood sugar which insulin converts to energy at a cellular level. Insulin is needed so your cells can feed off the blood sugar. The blood sugar that isn't used is stored as fat. Carbs are quickly and easily converted to fat. Other forms of food can eventually get converted to blood sugar, but not as easily and quickly as carbs. You absorb protein, minerals, vitamins, fibers, etc. very differently than pure Carbs. So a Carb rich diet will make you fat. Diabetics are only concerned with Carbs because they have the highest impact on their blood sugar level.

Further, America has proved this. Few cultures consume as much carbohydrate rich food as America, and as a result, they have the biggest issue with diabetes, obesity, etc. Carbohydrates are the cheapest food to buy (fast food and junk food are very heavy in carbohydrates), so a disproportionate number of low income people have obesity problems

Dude, its the carbs.

Diabetics have to manage insulin levels. They manage their carb intake because of this. It has nothing to do with managing their weight.

With respect to obesity you might consider what you just said: carbs are abundant and cheap. Therefore, people tend to eat a lot of them. Significantly more than they should. Carbs also do not tend to induce a level of lasting satiety that protein or fat does (note, 1g carb = 4kcal, 1g protein = 4kcal, 1g fat = 9kcal). So not only do people tend to eat a lot of carbs but they also tend to eat them more often because they don't feel full for very long.

Simple carbs will be converted to sugar quicker than complex carbs, and quicker than protein or fat.

But if you eat calories - in any form - that you don't need then you're going to store those calories.

And you're incorrect when you say that diabetics are only concerned with carbs. Diabetics cannot eat as much fat as they like, and are warned to eat a "healthy diet".

http://xkcd.com/793/

Shouldn't you expect to lose that water weight at some point though? Longer term weight gain should be calories in - calories out. What you eat might affect how full you feel (which could cause you to overeat in some circumstances), and macros matter if you're trying to build/maintain muscle.

My understanding is that peoples' weight problems are that they underestimate the calories of what they eat (related to specific foods or portions or forgetting details like the butter used for cooking) or overestimate that caloric effects of exercise.

Overeating may be the big problem, but abstinence from overeating is not really feasible for most people (it's depressing, really) and the truth is there are better ways of tackling overeating at the core. For me the interesting question is "why" we overeat.

For instance by eating healthier foods, instead of sugars and snacks filled with fats and refined carbs, you tend to feel less hungry. The theory behind it is proven actually, because sugars and refined carbs burn too fast, leaving you either with an empty stomach that wants more or with fat deposits from all the food that didn't burn at the same rate.

Water weight may make it harder to track your "true" weight, but that shouldn't affect your eating decisions.

More specifically, your water retention will fluctuate up and down within a narrow band depending on your diet, but you will never put on 40 pounds of water weight.

It's similar to limiting your how much water you drink so that you don't have all the liquid in your belly making you heavier.

Water retention is also a warning sign that something's wrong (like a hormone imbalance, one of the biggest causes for water retention) and you should probably take it easy. Fat people retain a lot of water. When people start a high-fat diet the first 5 pounds or so are from all the water loss (or even more).

> It's similar to limiting your how much water you drink

Not at all. In fact when dieting it's good to drink a lot of water, precisely to stop your body from retaining it. Odd, isn't it?

aside from hormone imbalances, why do you care at all about water retention?

Because hormone imbalance is the primary reason for why some people are fat and some aren't, being the reason for why type 2 diabetics can die of hunger while still being overweight.

Do some reading on the subject dude.

You should be able to get a relatively accurate number by always measuring your weight first thing in the morning, before drinking anything. You have to be careful not to get fooled, this is some kind of "best possible scenario" because your body loses a lot of water through perspiration during the night, but it'll give you a somewhat solid baseline.

IMHO fats is too generic.

Do you mean unsaturated, polyunsaturated or saturated fats?

Margarine or butter?

The answer is just as significant for any diet as the carbohydrates quantity.

How did you determine your target caloric intake value? I recently graduated and had a moment of self-realization as to how poor my diet was the last semester off campus and without a meal plan. I'm determined to do something about it, but health and fitness have never been my strong suit.

Use a calorie calculator such as http://www.freedieting.com/tools/calorie_calculator.htm or http://www.1percentedge.com/ifcalc/ and then adjust based on if you want to gain or lose weight.

Also, check out the reddit/r/fitness FAQ: http://www.reddit.com/help/faqs/Fitness for answers to other questions you're likely to have next.

Sigh. The OK Cupid "Let's look at the data" posts were so great that it's depressing to see other companies try to get attention by half-assing their way through a little research and coming up with a surprising conclusion.

Protip: When the conclusions of your amateur study contradict a peer-reviewed study done by professionals, your first thought should not be: gosh, those eggheads were wrong! Instead it's time to re-check your study design.

Recently the places I've lived have all had calorie counts on menus. It's very helpful. That way, when presented with a choice between a 400 calorie meal for $10 and a 1200 calorie meal for $10, I know which one will give me more calorie for my buck.

Yes, some of us use it to get more calories, not less. It's just far too helpful to know exactly how many calories you're getting. It's not just for people dieting or wanting to eat less.

Yup, I'm in this boat! I regularly use the calorie count to maximize my Cal/$. I have a hard time keeping weight on.

>Recently the places I've lived have all had calorie counts on menus.

The Affordable Care Act required displaying calorie counts across the entire United States. http://www.fda.gov/food/labelingnutrition/ucm217762.htm

Wasn't a coincidence you saw those counts in all the recent places you've lived (assuming you live in the US).

Living in NYC I've definitely appreciated seeing the number of calories on the menu; I'd like to see it everywhere and not just chains.

I'm much more likely to order something with a lower amount of calories. If there are three choices: 400, 600, and 800 calories, I'll pick the middle and feel better about the choice. I now know that a large combo meal at McDonalds is easily 3/4 of what my daily intake would be.

Secondly, it shows up the hidden calories. How many people realize that the 20oz of soda with their meal adds an addition 200-250 calories? Or that an venti eggnog latte at Starbucks is a scary 610 calories.

I think restaurants are a big culprit. You have no idea how good or bad the food you are ordering is.

One concern may be how easy would it be for a small restaurant to do this? Large chains with highly standardised processes probably have this information to hand. But for small restaurants, even if they analysed every meal before putting it on the menu, how do you know the chef wasn't a bit heavy handed with the cream that day.

The differences may be small enough to be negligible, but it could still be costly for small restaurants that are already operating under tight margins.

Calorie counts on menus is a good start, but it's not enough. It doesn't mean anything to the average American consumer that there are 1800 calories in that Starbuck's Mochaccino he just bought. Contextualize it with an Ad Council campaign to tell people that sedentary middle-aged people (for example) need to consume roughly 10 calories daily per pound of weight they want to maintain. You weigh 180? And you just bought that 1800-calorie candy coffee? Congratulations! you've achieved your maintenance goal for today. Any further calories you consume will be added to your waistline at 3500 calories per pound.

Then remind everyone that a sedentary lifestyle is a choice, and not a very wise one at that. Get your cardio on, and you'll burn through those excess pounds. Maybe even enable yourself to have something besides a Mochaccino each day. Get some resistance training going, and you'll need to ramp your daily intake up just to maintain all that lean muscle mass.

Of course, your mileage may vary; insult a doctor before beginning any weight-loss regimen; and if it hurts when you do that, don't do that.

Exactly. Few people understand what their calorie budget is and have no way to relate their intact decisions to that budget. Losing weight is simple: eat less than what your body needs to maintain itself. In doing so your body will pull from the energy reserves that it has (fat and muscle) to make up the difference.

Knowing what to do and doing it are very different things. It took a major macronutrient shift after at least a year of trying to diet and failing before I finally got on my way to losing weight.

I knew I was done for the day, but boredom, stress, and hunger made it only a matter of time before I would crack. A week of eating well, then cracking and eating anything I could get my hands on, then repeat.

Calories help, but especially for high-fat, low-carb diets they are pretty useless (see http://news.ycombinator.com/item?id=4892592 )

Interesting to see some evidence for my anecdotal experience that people severely overestimate McDonald's calorie counts (not to mention unhealthiness).

People look at me like I'm a crazy person when I tell them you could eat 4 cheeseburgers a day and still lose weight (provided you don't each much else, granted, but still).

I wonder if McDonald's ever ran any research on whether trying to reanchor their food as healthy would boost or hurt their sales.

Two points:

1) The minute your food touches a deep fryer, I think all bets on accurate estimates are off. My guess is that people don't know how much worse fried chicken sandwiches are for you compared to grilled chicken. A big part of fast food is deep fried french fries, which are mostly sold in "large" sizes due to value meals. There is also the soda element. I don't think most people realize how many calories are in the soda they get at fast food places, again, most likely in a "large" size.

2) A cheeseburger from McDonald's isn't that bad. But sandwiches like burgers have lots of ways to start pilling up the calories. Add an extra slice of cheese, toss on some mayonaise, maybe a little bacon. Pretty soon, you're hitting or going past the 1k calorie level.

So yes, a cheeseburger from McDonald's isn't that a bad (300 calories). A Angus Bacon Burger with large fries and a coke is a pretty bad meal (1600 calories).

Having a calorie count on the menu is helpful. The state of California started requiring this a few years ago and it has influenced what I order.

As an example, before the calorie count I would occasionally go for a tuna sandwich at Subway. I knew that the sandwich wasn't a healthy choice, but after seeing 1,000+ calories next to it on the menu I stopped ordering it. When you see items on the menu in the 1,000+ range and others half of that for the same portion it should influence your chooses.

"Estimates for fast food calories were significantly misguided – but in both directions! Apparently we have been misled by those oh-so-catchy jingles."

So, the short answer is: Yes.

See that gigantic logo at the top of the page that says "Survata"—it should be a link to your main web site.

In fact, such a link does not exist anywhere on your blog. You need to add one, or a few if possible.

Jesus. You'd think this was rocket surgery or something.

This article is rife with boloney statistical analysis and hand-waving, comes to conflicting conclusions at different points in the article, and really doesn't say anything.

Hi, OP here. We love feedback on our posts, so happy to hear your thoughts on conflicting conclusions - either here or at contact@survata.com. Thanks.

I'm glad that you collected this kind of statistical data - it's important, because caloric intake can really be misjudged (even if you are using a calculator).

I'd probably ask that you provide the data, so a better statistical analysis can be done on it than what was provided, and also that, in the future, you present statistics more scientifically. As others have noted, there seems to be some discrepancies with your reported calorie counts and what companies have reported - you should also address those.

Thanks. We did include both the raw data and calorie count references[1] in the footnotes in the original article. We also encourage you to play around with the data in Statwing [2]. Appreciate the feedback and curious what else you can find in the data!

[1] http://bit.ly/U6JQ5c [2] http://bit.ly/SMPeMZ

What is the purpose of calorie counting for humans exactly? Most people don't get fat/slim/strong/weak because they ate too much/too little [calories]. It's what you eat.

You eat X calories, you use Y calories. If X > Y most of the difference is stored (as fat), if Y > X your body burn the energy stored (fat, at some point, but it depends).

It's simple math and it's scientifically proven. 2000 kcal are 2000kcal, it doesn't matter if they come from a single burger or 10kg of lettuce.

It's not what you eat, it is how much you eat.

To be fair a lot of people are REALLY bad a measuring and estimating both how much they eat and how much they exercise.

True, but having calorie counters on the menu helps. People doesn't realize a frappuccino has 600 cals

I'd say it does matter, 10 kg of lettuce will take forever to digest and will require way more energy. Individual metabolism quirks will also affect how much of it comes out of the other end relatively intact.

It doesn't matter, you are just burning more calories in the digestion process but it's still an in vs out operation.

We are not perfect machines, but we are very efficient.

Calories do matter. It just so happens that when we make bad choices of food we take in much larger amounts of calories. Sure you can say certain macronutrients are better for building muscle/losing fat but in the end a caloric deficit will cause you to lose weight. I'm eating much more volume of food right now but still losing weight. It's hard to pack on calories eating just chicken, brown rice, and veggies. I could get 1 big mac and take in more calories than my entire day of chicken/rice.

I can see your point, but I hope you are eating more than 550 calories a day (amount of calories in a Big Mac in the US).

I should have said "big mac meal" which is more like 1350 calories, but still not too much for me to eat in 1 day. I can lose weight at < 2500 calories.

Some people can eat anything and naturally regulate their caloric intake.

Other people can eat whole foods (meat, vegetables, fruits, etc.) and keep a stable weight.

Others have an appetite that isn't well linked to their caloric intake and try to use external markers to guide what they should eat.

As long as it works, go for it.

A good resource for understanding calories and their effect on weight:

http://www.reddit.com/help/faqs/Fitness

1) estimate calories 2) add them up to too much 3) eat anyway and enjoy 4) drive back and resume sedentary lifestyle

What is really underestimated and forgotten are the calories of the daily casually snacks and drinks.

This would be insanely helpful for those of us who are cutting/bulking and want to keep track of exact amounts. Even better would be listing macro-nutrients along with the calorie count.

But at that point it becomes a UX issue. There's only so much space on a printed menu. Calories are an easy to print, single data point.

QR codes?

"For nutritional information check our website or scan this [qr code]"

Maybe .. but I can't imagine anyone sitting there designing a menu to actually say to themselves, "you know what would look awesome repeated on every line and column of this menu? 32 QR codes!"

Well I was thinking ONE QR on the back of the menu that takes you to a search or something. Obviously not for everyone but I know that I'd appreciate it.

What "we need" is oversight by another federal government agency that can issue us smart cards that are mandated by grocery stores and will help us in deciding what we can buy and how much.

I hear Michael Bloomberg has already invested in a startup.

Reply on news.ycombinator.com