when I renovate my bathroom I'm making a perfectly flat surface just for the scales.
Bathroom floors are not flat and when you put scales on them the torquing of the scales messes with its mesurements.
I can move my "smart scales" around the room and get a different result each time.
A running average is useless for something like a weight-losing diet (or for a weight-increasing diet). It is good only for verifying that your weight is stable.
When you are serious about losing weight, you must measure your weight every day at the same time and in the same relationship vs. eating/drinking and eliminating that.
You must lose weight neither too slow, nor too fast. A rate of around 100 to 150 grams per day is good.
Especially at the beginning of losing weight, accurate measurements of the weight are essential for adjusting the diet, i.e. when there is no weight loss one must eat less, and when the weight loss is too great, one can eat more.
A running average could not guide efficiently your decisions for the current day, whether to eat more or less. Most likely it would result in big oscillations around the ideal weight losing rate.
> A running average is useless for something like a weight-losing diet
I disagree, it worked fine for me, validating my ongoing calorie limits and estimates across several months and dozens of pounds.
> A running average could not guide efficiently your decisions for the current day, whether to eat more or less.
That's just using the wrong tool for the wrong job. You don't don't starve yourself today because you were constipated yesterday. (Or if you do, it's for issues of immediate discomfort rather than a dieting goal.)
Instead, you eat a consistent number of estimated calories, adjusting it to fit the observed trend in weight loss.
Paying too much attention to daily data points is also a psychological danger, people will put too much emphasis on noisy/singular data point and then use that as a rationale to give up or cheat.
Can you really measure your weight that accurately? What about your body's hydration level, or whether you've just taken a dump or not? Those would probably skew your measurement by half a pound.
Like I have said, you have to take the measurement every day at the same time and in the same order with respect to eating, drinking and dumps.
And yes, you can measure that accurately.
I had been obese for many years and I had many failed attempts to lose weight. This has changed only after I bought accurate digital scales and I have started to measure carefully, every day at the same time and in the same conditions.
During a day, your weight will vary a lot, perhaps by more than two pounds between the maximum and the minimum. Nevertheless, if you take care to measure at the same point every day, the accuracy is good enough, the uncertainty is less than 1/10 of a pound, if you are careful.
When I was losing weight, I could see very clearly the effect of eating one spoon more or one spoon less per day of certain foods, so I was eventually able to lose one third of my initial body weight, in about ten months.
This might be more true for a male than a female. Buildup of water in the body changes a lot and has nothing to do with what you eat. So measure your weight is very dangerous for your motivation if you eat almost nothing and still gain weight.
The buildup of water may change a lot, but it is not random.
When you eat similar things each day and you do similar amounts of physical activity, the buildup of water will also be mostly the same.
That is why during a weight decreasing diet or weight increasing diet you must observe a mostly constant diet.
If you eat wildly different kinds of food each day, with very variable amounts of carbohydrates and salt, then yes, you will not be able to monitor accurately your weight, due to the variable hydration of the body.
Also one must be aware that while gaining weight appears to happen instantly after a day with copious meals, frequently due to increased amounts of water retained in the body, at the beginning of a weight decreasing diet there can be a significant delay, e.g. of a week or even more, between decreasing the daily food intake and the beginning of weight decreasing.
Only after the weight has begun to decrease steadily (when the body has already decreased the metabolic rate to a minimum, in order to minimize its food necessities), then you can see the direct influence of greater or smaller daily intakes of food on the rate by which the weight decreases.
It's not random but it is also much larger variables than just what you eat. The body is very complicated. Women has the added fun of a monthly period and when that one stops, they have the even worse menopause. Not to mention a lot of difference depending on medical problems that anyone can have. It's great that it works for you but it does not work for a lot of people.
It's very frustrating that after eating exactly the same things and amounts for years, still the scale is jumping around just because of day of month.
That leaves you with a problem when it breaks, or the maker goes out of business so all the smart features (or in some cases the basic ones too) stop working, or the maker decides it isn't taking enough money from you so starts hawking ads before it'll show you any or your information, … You'll probably have a remodelling job to do to make any replacement device fit, rather than just placing the new device on the same flat surface.
I can move my "smart scales" around the room and get a different result each time.