Inaccurate Caloric Count? How does myfitnesspal calculate calories? Am I doing my math wrong?

2»

Replies

  • azuki84
    azuki84 Posts: 212 Member
    So many factors:
    Industry variance: Nutritional facts are FDA approved, but the law allows a margin of error up to 20 percent.
    Human error: Database input by humans. No need to explain this one unless you really don't know this
    Conclusion: Do the best you can, and while you weigh yourself for data, look at YOUR BODY COMP OVER 1-2 WEEKS!!!
  • claireychn074
    claireychn074 Posts: 1,324 Member
    I pay for MFP and expect it contain an accurate database of food counts. I am so disappointed that I can't rely on it to be accurate. It is supposed to make tracking your food intake easy, it doesn't. Instead I have to look up all the data elsewhere and then enter my own. I could do that in a spreadsheet for free. Is there a way to get the company to fix this??? How do I contact them?
    Aside from what everyone else has said, it’s worth noting that the macro content of food is not actually consistent. Salmon has widely varying amounts of fat depending on whether it’s wild caught or farmed, the type of salmon and where it’s from. There is no way to be accurate, so even the food labels on the packets are an average. Watermelon will vary hugely in calories (as a percentage rather than total as it’s low in cals generally) depending on how dense the flesh is i.e. how much water it received when growing. Apples have hugely different amounts of sugar between varieties, and from year to year (depending on sunlight, water, nutrients and pests). Food is not static so actually there can never be an accurate reading from day to day.

    As others have said, pick an average from the entries or pick entries which closely fit any packet info and use those. It’s as accurate as you can be - without exactly preparing double the food intake, burning one half in a lab to measure the gasses given off and calculate the calorific value!
  • gpanda103
    gpanda103 Posts: 189 Member
    I pay for MFP and expect it contain an accurate database of food counts. I am so disappointed that I can't rely on it to be accurate. It is supposed to make tracking your food intake easy, it doesn't. Instead I have to look up all the data elsewhere and then enter my own. I could do that in a spreadsheet for free. Is there a way to get the company to fix this??? How do I contact them?

    The way I do this is I compare the results with what the usda food database has listed. If you are eating blueberries for example, the most accurate result is gonna come from the search “blueberries usda”
  • AnnPT77
    AnnPT77 Posts: 31,966 Member
    edited May 2022
    gpanda103 wrote: »
    I pay for MFP and expect it contain an accurate database of food counts. I am so disappointed that I can't rely on it to be accurate. It is supposed to make tracking your food intake easy, it doesn't. Instead I have to look up all the data elsewhere and then enter my own. I could do that in a spreadsheet for free. Is there a way to get the company to fix this??? How do I contact them?

    The way I do this is I compare the results with what the usda food database has listed. If you are eating blueberries for example, the most accurate result is gonna come from the search “blueberries usda”

    No, that one will be user entered, and subject to the same vagaries as any other user-entered item in the database. (I admit that "USDA" in the label implies that the person looked there, didn't just make it up. But we're still trusting their accuracy and typing skills.)

    The actual entry that was loaded from the USDA database at MFP start-up would be "Blueberries - raw", with a default serving size of 1 Cup. That will not show up anywhere near the top if you search "blueberries USDA".

    This one has a less bureaucratic name** than most (which is unhelpful), but I guessed it would be this one because of the 1 Cup default quantity for a thing it's inaccurate to measure in cups (not a guarantee of USDA source). Then, when I click the serving drop-down, there are quantities of different types: Counts ("50 berries", volume measures (cups, milliliters), weights (grams, pounds)). Sometimes there will be inch sizes, too, but not for blueberries. The different types of serving sizes is a sure sign. (Even some of those show a dramatically inaccurate calorie level for one or two of the calories, due to some long-ago database updates, I'm told.)

    ** Example of more typical name "Tomatoes, red, ripe, raw, year round average".

    /pedantic
  • cwolfman13
    cwolfman13 Posts: 41,876 Member
    edited May 2022
    I pay for MFP and expect it contain an accurate database of food counts. I am so disappointed that I can't rely on it to be accurate. It is supposed to make tracking your food intake easy, it doesn't. Instead I have to look up all the data elsewhere and then enter my own. I could do that in a spreadsheet for free. Is there a way to get the company to fix this??? How do I contact them?

    You can also do MFP for free and it's not materially different than the paid version. I've been using this site for 10 years and never paid a dime. The database is user based crowdsourced. It's the same database in the free vs "premium" version. I've never entered my own stuff, but I have had to look at different entries to find the correct one that matches the nutritional label I'm looking at. I used USDA when looking up whole foods. Part of the problem is user error...the other part of the problem is different labeling in different parts of the world for the same product. Another issue is that people assume this to be some kind of exact science...and it's not. The macro calorie distribution isn't exact...it's an estimate. Foods aren't exactly XXX calories and blah, blah, blah macros...the same size ribeye steak is going to vary in fat content and protein content for example. Never mind that nutritional labels and restaurants are allowed error margins of up to 20%.
  • AnnPT77
    AnnPT77 Posts: 31,966 Member
    I pay for MFP and expect it contain an accurate database of food counts. I am so disappointed that I can't rely on it to be accurate. It is supposed to make tracking your food intake easy, it doesn't. Instead I have to look up all the data elsewhere and then enter my own. I could do that in a spreadsheet for free. Is there a way to get the company to fix this??? How do I contact them?

    You file a trouble report through the Help function, or post (or support an existing post) here:

    https://community.myfitnesspal.com/en/categories/feature-suggestions-and-ideas

    I hope you realized that basic MFP is free? There are some good features in the for-pay premium version, but they're not essential to most people for weight management.

    Speaking for myself hereafter, not directly to your question:

    I have pretty good spreadsheeting skills, plus a long career history as a system designer/developer, and there's no way I could set up a spreadsheet that would be as easy or quick as using MFP, even in the long run, let alone the short run. After a short time getting used to the app, I spend maybe 10 minutes a day logging food with adequate accuracy to lose weight (obese to healthy weight) and maintain (for 6+ years since loss). YMMV, of course.

    I think what you're looking at here is a design tradeoff. MFP has a huge database, and with a little manageable checking, the accurate entries are in there. I rarely have to add a food; they're already there. That's because the database is crowd-sourced: The first person who wants to log a food adds it, then others can use their work.

    Once I find a good entry, as long as I use it semi-regularly, it stays in my recent/frequent list and comes up first when I search/add foods. For sure, finding the accurate entry almost always takes less time than entering it from scratch, so this is a win, for me.

    There are other apps that quality control their database accuracy more tightly, but they will tend not to be as comprehensive - not as many pre-existing foods. (I believe Cronometer is one of those, from what others have said.)

    So, like I said: A design tradeoff between a crowd-sourced but very extensive database that has errors, versus a more reliably/consistently accurate database that isn't as comprehensive.

    I don't think you'll convince MFP to change that philosophy. I, for one, wouldn't want them to change it. I only speak for myself though.