Are You Really Losing Weight? | Hacker News Reader

Are You Really Losing Weight?

jasmcole.com

45 pointsalexmolas58 comments

This feels like the stack-exchange math-nerd's fitness version of "I wanted to be more productive so I spent 1000 hours writing this weird TODO app."

All that typing and mental effort must have burned some calories.

This is important because most people “losing weight” have made one change to their diet and exercise program while simultaneously eating more or otherwise cheating. However it feels like they are more healthy, and the trend line seems to agree. (Often they are much healthier, just not changing weight but redistributing it, but measuring that is far harder).

People eventually get jaded of the false progress cycle and give up on weight loss because it feels impossible. An app implementing this system would be of great help.

> People eventually get jaded of the false progress cycle

Most people just don't want to do the work. Changing a lifelong eating habit is hard. Moving from a sedentary lifestyle to a more active lifestyle is hard. There are ups and downs and overall the progression is definitely more of a journey than just an overnight thing.

We live in a therapeutic culture which just wants to address things with a pill or with a surgery. Actually admitting that your lifestyle sucks and that it must change is not something we a society have embraced much in the last few decades.

Weight is a poor metric to track in isolation, prone to error. Much more robust to also track waist circumference.

There is actually a neat formula called the navy formula to estimate body fat and lean mass from four variables for men and five for women. For men: waist circumference, weight, neck circumference, height. Women should also add hip circumference.

There are calculators online, but for anyone interested I expressed the formula into this Shortcut for Apple devices. You enter your height (do this once, in the blank text field above height)

Then, measure your neck circumference and remember it. This won’t change much. I measure mine maybe once every 1-2 months or if my other measures changed significantly.

Then, whenever you would normally weigh yourself, measure your waist too. Then enter weight + waist + neck circumference. The Shortcut will then calculate the following and log to Apple Health:

* Weight

* Waist circumference

* BMI

* Estimated lean mass

* Estimated body fat

The last two obviously have an error, but it will be a fairly consistent error so the trend over time will be clear.

For instance recently I gained a bit of weight. But my waist circumference went down a little. So my weight increase is actually muscle, and I’m going in the right direction.

Shortcut is here: https://www.icloud.com/shortcuts/1be74cce1ac24c03821feb7e36d...

Don’t have a woman’s version unfortunately, made this for my own use. If there’s interest I could probably create a version that includes hip circumference

Does this assume specific units for the measurements of weight and circumference? If yes, what are those and how would one change the formulas for other units of measurement?

Oh right, this one is inches and pounds. You would change it by going into the shortcut and entering different conversion figures based on the formula in metric instead

Navy Formula calls me an athlete and is SORELY mistaken. BMI accurately calls me overweight, but is probably mistaken about how much.

It’s more about direction than absolute level. It has my body fat far too low I think but I can see it change with my waist/weight over time

BMI being accurate is really a broken clock being right twice a day sort of thing. If you have any appreciable muscle mass, or do any sort of training beyond just running (or run a lot) BMI is wrong almost all the time.

I prefer waist circumference, but BMI is not far off for most people. I’m quite muscular, only times I ever hit “slightly overweight” on BMI I also had a small gut

I checked the the BMI of elite hockey players ranged from 25.5 to 27. And that’s for rather muscular people, also tall, at the far edge of the “high bmi, not overweight” spectrum.

Virtually guaranteed if you have a BMI 28-29+ as a male you also have excess waist circumference. The idea that BMI is comically wrong is…well, wrong.

The premise of this makes no sense. Who is actually doing this process?

Quite surprised at how people are reading a little mathematical exercise as anything to do with either weight loss or what people are doing in reality.

> throw away all of the results which showed an increase in weight

Huh?

^ core assumption of this article is that people weigh themselves multiple times a day, and only keep the minimum weight. Maybe this is the case in the first few days, but I’d be surprised to hear most people weigh themselves _many_ times a day.

I have been doing this in order to learn how much weight I immediately gain after a meal to actually learn how what I eat results in weight changes. Like eating 6 "meatless" chicken nuggets vs a slice or two of pizza and adjust accordingly to meet my daily target weight for the end of the night right before bed for healthy weight loss on a weekly basis. Portion sizes in the USA are so far different from reality that what you think is "one serving" is actually 2-3+ for a given meal, and I have had to completely rethink what a "healthy" portion size actually looks like for breakfast, lunch, and dinner.

Pretty sure the dominant factor here is salt content and thus water retention.

You could also frame it as a protein vs carbs argument and which one fills you up more with less overall calories after a given meal.

But fiber and water fill you up too.

I have tried focusing on fiber in the past and while it does fill me up, without protein that feeling of being full doesn’t last very long. Have added 2-3 cups of tea every day and it helps me relax before a meditation session if nothing else.

Sheer water content in the food.

Why would you do thst instead of weighing the food directly which would be far more accurate?

Not a bad idea, going to order myself a food scale on the next pay day. Still figuring things out, I just know what has been working for me so far.

Weight after eating is not a good predictor for how a food will affect body composition. Examples of two extremes: you could eat a big salad that's relatively heavy (water + fiber) but that weight won't stick around more than a day. Or you could have a bunch of cookies that are relatively light weight, but easily tip your energy balance and end up increasing body fat.

I measure my weight throughout the day to see how what I do activity wise and eating wise results in changes. Weight after eating is just one indicator.

I think this is incorrect.

As stated in the article:

“One were to weigh themselves every day and throw away all of the results which showed an increase in weight.”

Imagine I weigh 190 today, and I measure 180 tomorrow, 185 the day after, and 170 the day after that. The graph of values would be 190->180->170, ignoring the day when we increased from 180 back up to 185

Ah, yeah I guess the article could be doing on a day-to-day basis instead of multiple times a day.

Maybe the more misleading part of the article is that the number of data points is relatively constant. I would think there's some rough constant on frequency of data points (ie, 1/d), and as you get closer to a true minimum, you have much fewer data points. Having fewer data points as time goes on makes the inaccuracy more obvious, and thus this article less interesting.

Def an interesting read tho.

I have weighed myself many times per day but I kept the times consistent to see how it fluctuated. It was decently interesting, essentially always the first weighing in the morning was the lightest when fasted and dehydrated

I have body image issues and I weigh myself multiple times a day; and potentially this has an effect on how and what I do and put in my body. Probably bc the outdated BMI charts say I’m overweight at 155lbs and 5’6” height.

But I also track every medication, beverage, food consumption, junk I put in my body.

For me, more data provides more insight and control over my body. Maybe I’m just nuts?

Weighing yourself over the course of weeks and months will reveal trends, but over the course of a single day, body weight fluctuates wildly, sometimes by several pounds. This depends on food intake, water intake, bowel movements, sweating, choice of clothing, etc.

So it's hard to imagine how weighing yourself multiple times per day can provide any insight whatsoever. It sounds more likely to hurt, and not help, body image issues.

Oh it certainly leads to healthier eating choices and/or unhealthier meal skipping but I do try to be completely aware of my surroundings including clothes, bms, etc.

I mean it can be the decision maker between a coke and a water for me. Regardless if the data is not sane.

Probably I should just only drink water and only eat whole grains, veggies, and fruits. But who doesn’t love a coke?

No, your real weight does not fluctuate around a mean with a standard deviation. For one thing, your weight today has a great deal to do with your weight yesterday. This alone means you shouldn’t expect the kind of graph in TFA if your weight is in fact stable.

Check out trendweight. With a smart scale it's been awesome. It tries to show you a moving average as your true weight and not emphasize the actual day to day weight.

http://trendweight.com

What actually is the standard deviation of the weight of a normal healthy adult human (not actively trying to lose weight)? I kind of assumed it was fairly low.

Not sure, but mine is changing by ~ 6% every year for the past 10 years, by seasons (yearly cycle). I started to consider this "normal".

Things like water retention and how much food is in your gut can cause variance throughout the day. If you weigh yourself at a consistent time of day and consistent with bowel movements it should be pretty steady.

Add to that: if you do it in the morning try to have consistent sleep cycle, and make a "once and for all" decision for stuff like should you pee before, should you drink/take a shower before. Lots of things can influence your weight. My scale will also give a few different results for reasons I ignore, try to weigh yourself a few time to see if that's the case and choose what to take (always the highest/lowest? Average? Median?)

Isn't that exactly why most modern scales are calculating your body fat?

They do what now? How does that work?

Yes. They measure electric resistance of your body to determine fat vs muscle ratio.

I am skeptical anyone needs this. If you have love handles that can be pinched, you already know where you are.

But being able to measure changes in the pinchable love handles is not so simple. After all, getting regular feedback on whether a change is making a difference can be an incentive to maintain that change.

Using a cloth measuring tape and a mirror for reference gives plenty of feedback. Bf% from a scale is going to be so inconsistent I could only think it would be useful over a large timeframe to notice trends, which at that point your jeans are too big.

Are you basing this on actual experience or just your belief?

Most modern scales use Bioelectrical impedance analysis [1] to calculate your body fat percentage.

[1] https://en.m.wikipedia.org/wiki/Bioelectrical_impedance_anal...

It’s so inaccurate and fluctuates wildly if you are hydrated.

One the other hand if one does it in the same conditions it allows to track relative changes which is good enough for casual monitoring.

I have a Tanita scale that does the body fat thing and more. It's so inaccurate that measuring twice back to back can yield vastly different results.

I doubt Most modern scales use Bioelectrical impedance analysis.

Do these scales exits? Yes.

Can normal people get them? Yes.

Do most scales do this? No, most scales are $20 scales bought at Target.

Reply on news.ycombinator.com