Calorie discrepancy

I am entering foods and to MFP and noticing a big discrepancy with the calories that the app gives me and the calories that I can find a calorie websites. I'm also noticing this on recipes that I enter with the website address. For instance I had half a teaspoon of cinnamon come up as 300 calories. Anyone else double checking their calories?
«1

Replies

  • kami3006
    kami3006 Posts: 4,978 Member
    edited January 2019
    There is an issue with the servers at the moment so that can cause some discrepancies.

    Having said that, you always need to double check entries in the database. MFP entries are user-entered and prone to errors. Also, nutrition formulas change making some entries outdated and formulas vary by region/country.
    Barcode entries should also be checked as scanning them pulls up user-entered entries and barcodes also get reused for newer products so they may not match what you're actually eating; this happens a lot with in-house made items from grocery stores.

    For whole foods, you can double check entries against/or use the USDA database: https://ndb.nal.usda.gov/ndb/search/list
  • KipDrordy
    KipDrordy Posts: 169 Member
    I look for the items with check marks. Even then, I sometimes double check when things look wrong or I see a list of the same food that varies widely, I look further into it.
  • kellyjellybellyjelly
    kellyjellybellyjelly Posts: 9,480 Member
    KipDrordy wrote: »
    I look for the items with check marks. Even then, I sometimes double check when things look wrong or I see a list of the same food that varies widely, I look further into it.

    I think the hardest thing with the database is when we do enter something wrong there's no way to delete it anymore. I accidentally messed up a few of the yogurts I entered that weren't in the database & thought I had double checked to make sure I entered it correctly & somehow put the grams & the calories as the same.
  • MichelleSilverleaf
    MichelleSilverleaf Posts: 2,028 Member
    I resorted to making a lot of my own entries. Just easier, I know it's accurate and I don't have to worry about someone else down the road changing them on me.
  • bluesheeponahill
    bluesheeponahill Posts: 169 Member
    if the cinnamon is coming up as 300 calories, then you've got the serving size wrong like you're saying one teaspoon, but its entering it as 1 cup? you definitely have to double check. when i enter recipes in, saying 1 cup of x isn't going to cut it. i have to edit the qty/or serving size because it cant always get it right. I do have to do some searching to make sure its the right one, because there will be about 50 different cinnamon entries there, just need to find the right one for your brand and amount entered.
  • lemurcat2
    lemurcat2 Posts: 7,899 Member
    if the cinnamon is coming up as 300 calories, then you've got the serving size wrong like you're saying one teaspoon, but its entering it as 1 cup? you definitely have to double check. when i enter recipes in, saying 1 cup of x isn't going to cut it. i have to edit the qty/or serving size because it cant always get it right. I do have to do some searching to make sure its the right one, because there will be about 50 different cinnamon entries there, just need to find the right one for your brand and amount entered.

    Some entries are genuinely way off and totally inexplicable.
  • JeromeBarry1
    JeromeBarry1 Posts: 10,182 Member
    For spices, I usually get good results searching for something of this format "spices, cinnamon, ground, usda"
  • Bcally64
    Bcally64 Posts: 3 Member
    I am trying to figure out why the unsweetened cold brew at Starbucks is 5 cal and when I scan the code on the Starbucks at home cold brew box, it says 270 cal. I have looked everywhere for the nutrition label but cannot find it. Thoughts? I am thinking of contacting Starbucks.
  • yirara
    yirara Posts: 9,329 Member
    I don't think it should happen, but I have a few recipes saved that I've not cooked in a while. Now there are really odd entries in there, like 160gr pasta for 2000kcal.
  • wilson10102018
    wilson10102018 Posts: 1,306 Member
    MFP should cleanse the database by deleting entries that are more than a year old. And, as said above, the verification means nada. All sort of nonsense in the database.
  • kimny72
    kimny72 Posts: 16,023 Member
    Bcally64 wrote: »
    I am trying to figure out why the unsweetened cold brew at Starbucks is 5 cal and when I scan the code on the Starbucks at home cold brew box, it says 270 cal. I have looked everywhere for the nutrition label but cannot find it. Thoughts? I am thinking of contacting Starbucks.

    Scanning the code does not take the info from the manufacturer. It simply searches the MFP database for a similar entry, just like you searching by typing would. You are just as likely to get a bad entry from scanning. You need to search for an accurate entry with info that matches Starbucks nutrition info or create your own entry.
  • garystrickland357
    garystrickland357 Posts: 598 Member
    Yes, double check. Use a healthy dose of common sense as well. Realize that logging calories is not an exact exercise as well - there is error. That doesn't mean it isn't a useful tool. Log as accurately as possible and then monitor your weight for trends over time.

    Yesterday is a good example. We had our church annual meeting. There was a pot luck luncheon. I logged my food choices as accurately as possible - but I'm totally aware that I could be off by +- 500 calories.

  • AnnPT77
    AnnPT77 Posts: 31,717 Member
    MFP should cleanse the database by deleting entries that are more than a year old. And, as said above, the verification means nada. All sort of nonsense in the database.
    MFP should cleanse the database by deleting entries that are more than a year old. And, as said above, the verification means nada. All sort of nonsense in the database.

    Sigh. Are you now, or have you ever been, a data quality analyst or data analyst?

    If they wanted to improve the database quality, there are statistical methods they could use that would be more sensible. (They'd still be a lot of work. Yes, more work than offloading old dates, but dropping old entries would simply shift the work to us users to add back entries that were formerly present, and reasonably accurate on average . . . and the add-backs would be likely to be just as error-ridden as the entries that were cut, because humans.)

    The database accuracy is mostly not terrible for simple foods (ingredients). It's crazy-variable if we're picking other people's home-made food entries (and always will be, because their recipe is not our recipe: We have no concept what's in their recipe, and whether they truly estimated it systematically, or guessed).

    The database accuracy is an obstacle for new users, who need to check entries at first. Once one's recent/frequent foods list is populated, most of us need to check only occasional items. With experience, we also develop a sense of how many calories look realistic or unrealistic for foods similar to others we eat. It becomes manageable, even easy.

    It's regrettable that the problem affects new users the most/worst, I admit.
  • lemurcat2
    lemurcat2 Posts: 7,899 Member
    AnnPT77 wrote: »
    MFP should cleanse the database by deleting entries that are more than a year old. And, as said above, the verification means nada. All sort of nonsense in the database.
    MFP should cleanse the database by deleting entries that are more than a year old. And, as said above, the verification means nada. All sort of nonsense in the database.

    Sigh. Are you now, or have you ever been, a data quality analyst or data analyst?

    If they wanted to improve the database quality, there are statistical methods they could use that would be more sensible. (They'd still be a lot of work. Yes, more work than offloading old dates, but dropping old entries would simply shift the work to us users to add back entries that were formerly present, and reasonably accurate on average . . . and the add-backs would be likely to be just as error-ridden as the entries that were cut, because humans.)

    The database accuracy is mostly not terrible for simple foods (ingredients). It's crazy-variable if we're picking other people's home-made food entries (and always will be, because their recipe is not our recipe: We have no concept what's in their recipe, and whether they truly estimated it systematically, or guessed).

    The database accuracy is an obstacle for new users, who need to check entries at first. Once one's recent/frequent foods list is populated, most of us need to check only occasional items. With experience, we also develop a sense of how many calories look realistic or unrealistic for foods similar to others we eat. It becomes manageable, even easy.

    It's regrettable that the problem affects new users the most/worst, I admit.

    There's definitely lots of nonsense in the database, but a lot of issues are thing like people picking "homemade lasagna, 1 serving" when how on earth could you not realize that there's too much variety in how people make the dish, not to mention what they think a serving size is?

    Also, some of the perceived inaccuracies are things like different versions in different countries or some countries having labels that are net carbs or serving sizes having changed.

    And some are of course just bizarre.

    Learning to find the MFP-inputted USDA database items is the best tip I can give anyone. The sad thing is it used to be easy to tell which one was them from the list that came up, and they got rid of the old system for the terrible verification system.
  • AnnPT77
    AnnPT77 Posts: 31,717 Member
    lemurcat2 wrote: »
    AnnPT77 wrote: »
    MFP should cleanse the database by deleting entries that are more than a year old. And, as said above, the verification means nada. All sort of nonsense in the database.
    MFP should cleanse the database by deleting entries that are more than a year old. And, as said above, the verification means nada. All sort of nonsense in the database.

    Sigh. Are you now, or have you ever been, a data quality analyst or data analyst?

    If they wanted to improve the database quality, there are statistical methods they could use that would be more sensible. (They'd still be a lot of work. Yes, more work than offloading old dates, but dropping old entries would simply shift the work to us users to add back entries that were formerly present, and reasonably accurate on average . . . and the add-backs would be likely to be just as error-ridden as the entries that were cut, because humans.)

    The database accuracy is mostly not terrible for simple foods (ingredients). It's crazy-variable if we're picking other people's home-made food entries (and always will be, because their recipe is not our recipe: We have no concept what's in their recipe, and whether they truly estimated it systematically, or guessed).

    The database accuracy is an obstacle for new users, who need to check entries at first. Once one's recent/frequent foods list is populated, most of us need to check only occasional items. With experience, we also develop a sense of how many calories look realistic or unrealistic for foods similar to others we eat. It becomes manageable, even easy.

    It's regrettable that the problem affects new users the most/worst, I admit.

    There's definitely lots of nonsense in the database, but a lot of issues are thing like people picking "homemade lasagna, 1 serving" when how on earth could you not realize that there's too much variety in how people make the dish, not to mention what they think a serving size is?

    Also, some of the perceived inaccuracies are things like different versions in different countries or some countries having labels that are net carbs or serving sizes having changed.

    And some are of course just bizarre.

    Learning to find the MFP-inputted USDA database items is the best tip I can give anyone. The sad thing is it used to be easy to tell which one was them from the list that came up, and they got rid of the old system for the terrible verification system.

    This is going to sound crazy to some people (probably not to you, lemurcat2 ;) ), but one hint I use to find them is to notice database entries whose default quantity is 1 Cup, but for which that's a completely nutty way to measure. Click the serving-size drop-down; if there's a big laundry list of different serving sizes that can't be calculated from one another (like 1 Cup, 2" sized fruit, 100g, 1" slice, etc., all on the same list), I think it's probably one of the old USDA entries, and probably pretty good.

    A few of those have a messed-up entry for Tablespoons, with the others OK. The ones that are messed up are some of the real MFP howlers, but just for one or two of the serving size choices. Go figure! The 100g entries in those are pretty much all sound, in my experience.

    Example (if you look it up, you'll find 23 different serving size options, including 100g):

    7tef9ohs9zbs.png
  • wilson10102018
    wilson10102018 Posts: 1,306 Member
    I should have been more specific. The database should be cleansed by our selections. Items which have not been added to food diaries for a long time should be deleted. That method of user selection would be the best method for upgrading the database. We are the experts and if we aren't selecting an item from the MFP database it is because it is wrong or irrelevant.
  • cmriverside
    cmriverside Posts: 33,905 Member
    I should have been more specific. The database should be cleansed by our selections. Items which have not been added to food diaries for a long time should be deleted. That method of user selection would be the best method for upgrading the database. We are the experts and if we aren't selecting an item from the MFP database it is because it is wrong or irrelevant.

    "We" are the experts? You joined in September. I joined in August of 2007.

    Early on every food that was added to the database automatically became shared with the world. There used to be no option to NOT share it. That was how the huge database of foods was sourced, mostly. Like Ann says, there are ways to find the good ones.

    If they deleted those that are more than a year since being used, I'd probably lose a lot of MY FOODS and that would not make me happy.

    Just learn to use it as is. It's not that hard.
  • wilson10102018
    wilson10102018 Posts: 1,306 Member
    edited January 2019
    I should have been more specific. The database should be cleansed by our selections. Items which have not been added to food diaries for a long time should be deleted. That method of user selection would be the best method for upgrading the database. We are the experts and if we aren't selecting an item from the MFP database it is because it is wrong or irrelevant.

    "We" are the experts? You joined in September. I joined in August of 2007.

    Early on every food that was added to the database automatically became shared with the world. There used to be no option to NOT share it. That was how the huge database of foods was sourced, mostly. Like Ann says, there are ways to find the good ones.

    If they deleted those that are more than a year since being used, I'd probably lose a lot of MY FOODS and that would not make me happy.

    Just learn to use it as is. It's not that hard.

    Every one has to be crossed checked. Half of them are garbage.