Subscribe
Share
Share
Embed
Micro wisdom delivered to your ears every morning in voice notes ranging from 3 to 15 minutes long. Wisdom on how to live a healthier and more fulfilling life. Every podcast will ground you in the present moment to ensure you know what's important, the here and now.
Hello everyone, welcome back to the podcast. So today's topic, I'm going cover a new study, well relatively new, 2014 in September. It covers what's better to track, is there a food diary, just writing out the food you've had, sometimes with serving sizes, the best of your ability, or taking photos of foods and getting that analysed by artificial intelligence. So which one is better? The reason I want to bring this up is because the app uses all of these methods, you can do both methods, whatever's better for you is usually the best one to do.
Speaker 1:So for example a lot of people will say in the comments of some of the ads how is that possible, you need to be bang on and you'll have some 19 year old PT like you have to weigh everything to get results. It's like buddy calm down, no you don't. Okay, you don't have to be eating meal prep every day, you're 19 years old so you know you're not going to understand. Someone who's 35 with kids working long hours and has limited time and needs to cook for kids blah blah blah. So it's very different scenarios for different people so you can't take social media comments that seriously.
Speaker 1:But what is more accurate and what's important? So we often underestimate our food intake by around 30% to 50%. So say now you think you consumed 2,000 calories, you probably consumed 3,000 calories, okay, if you were on track. You consumed 3,000 calories, you're probably consuming more close to 4,500 calories. So we're really bad at this, this has been tested multiple times and even dietitians are not great at this and they work with food every day, so we're not great at tracking food with just our eyes and like estimating through that, we have to actually find a way to note it down.
Speaker 1:So one classic example is this, people will do a photo, will be two tables tablespoons of peanut butter, could easily be misjudiced as one tablespoon and that's like 100 calories difference. Some foods can be like that, so it's important that some of those foods are measured and you understand, so like peanut butter if you do that make sure you weigh it out and you understand the weight you're using. Oils are dense in calories so you want be careful with these things, but yeah there's different ways you can log your food, you've got a simple written food diary by hand, you've got taken forward as what you eat which could be sent to a coach or AI tracking app, like a bar pal, there's a tracking hand portions method where you use like a fist size for your carbs and a thumb size for your fats and a small hand size for your protein, stuff like that. There's weighing and measuring food and then recording in a food diary and there's more options. So the best method actually is what works best for you, so what you can do consistently, that's the key thing.
Speaker 1:A consistent food log no matter if it's a 100% accurate or 75% accurate, 83% accurate is the one that's going get you results over time, not the one you can do perfectly inconsistently. So if you track only 40% of your days a year with weight, you're missing out 60% of the other days, you've got no data on them, you don't know where you are, that's not going to be as good as being able to track 90% of your days with a method that might be slightly less accurate but gets you consistency. The reason for this is food awareness is important, like being aware of the foods you've eaten and actually being able to make decisions and look at your behaviour that way. Motivation is another one, being able to like, you're tracking, you're doing something, you're keeping an eye on it, that typically leads to results as well. And just general habit building, so if you're building tracking or food awareness into your day to day, that food awareness itself is going to change your behavior.
Speaker 1:We see this in multiple studies and there's a great book called Tiny Habits by BJ Fogg that you can listen to. So this study, 26 people, nine males, 17 females visited a lab to learn how to use the food tracking methods. Food diary meant writing descriptions of everything they ate using household measurements like cups or tablespoons and stuff like that, and then you've got food photography, just taking a foodo, that's name of food photo, Jesus, I've just found a hack, taking photos of everything they ate in a standardized style, so like on a white plate from a certain distance which then AI was analysing. That's the thing with this one, they made sure the photos were bang on for the AI sometimes in the wild on these AI tracking, photo tracking is not going to be as good. So yeah, they basically tested them, they wanted to see how reliable and accurate each food method was and they wanted to ensure the practicality of each method, so having them do real life scenarios as well.
Speaker 1:So they did make sure they knew the actual quantities, the calories and macros of the foods that were given them so they could test it out to the actual, so like a retest, test retest analysis, it's a good thing to do. Both methods actually underestimated calorie value of the provided food items by 11 to 24%. The variation and accuracy depended on the food item method used, so in terms of methods on average food diaries were off by 13% for calories and photo tracking is off by an average 18% for calories. So food diary recordings of the provided food items were more accurate than food photos of these items, by more accurate I mean closer to the actual calorie and macronutrient values of the foods. Then food diaries writing down actually provided more reliable results than the photo tracking, but the photo tracking was also relatively reliable in terms of redoing the photos and getting similar results, But fats was not that reliable because you can't see fats in foods most of the time, so you can't see oils or fats you cook with or mix into foods, so it's making it hard to detect by the phototracking.
Speaker 1:So that's one thing that phototracking doesn't do as well as being able to actually write down well, there's olive oil here, we cook this with olive oil. So they both did underestimate but fairly fairly close 11%, sorry 11%, 13% to eighteen percent. This is why when we talk about tracking with our and this is why we did the tracking accuracy spectrum, so the tracking accuracy spectrum puts photo tracking as the least accurate because it can't see the fats, it has some variance, depends on the quality of the photo, blah blah blah. Describing the foods you've eaten is more accurate as proven in this study as well, describing with not just servings but like, well describing without servings and describing with servings and then describing with weight, you know, you've got those three tiers. So ideally you want a mix of all of them, right, you don't want be trying to do one method of tracking, you want to track with the photos if you need to do it that time, track with just writing down the food if you don't know any other information, tracking down with serving estimates like it was a tablespoon of olive oil or a drizzle of olive oil or it was a thumb sized portion of butter, you can use whatever you want, you can explain it and it's going be more accurate and then you've got weight is the most accurate.
Speaker 1:But the key thing is using all of the tools to be able to track every day easily. So all these studies are looking at one method versus another and I think what we've done is we've looked at all the methods and gone all of them have their place, let's put them into an app and make it as easy as possible to track and make it easy to focus on calories and protein. So this would really motivate you that the fact and also with our app we train our AI models to overestimate so we know that most things underestimate so if we were to have trained our AI and our app on this study we probably would have seen it actually closer or overestimate slightly because we train the AI on that. For example we tell the AI, and we can show this information in a group like how it thinks about stuff, Say, you you said I had scrambled egg on toast with an apple. So the AI will start thinking and you can actually see it's thinking and the thinking goes I guess because it's based on the training we've given it, right?
Speaker 1:So the thinking will go, okay scrambled egg on toast, so scrambled egg on toast probably gonna be two eggs, because then two eggs is gonna be covering toast. Toast, we're going to go with one piece of toast which is cut in half. We're going to assume butter's been used in the scrambled egg and assume some milk because it goes for the overestimation. So we're going to assume there's butter on the bread as well, so factor these things in. We've changed this because a lot of people might not add those little things in but it's better that they're in anyway.
Speaker 1:But if you did say I have scrambled egg and tossed with butter it's not going to double count butter. So when you don't say it's going to assume it's probably a bit used, if you do say it's going to take what you say. And it's important we do that because we've realised that people might have cooked stuff with things, people might have done other things, and like say you said I had something from a restaurant, there's like a restaurant calorie boost on top because restaurants typically use more fats and stuff for tastes so there's more calories added to them. So we're doing things like that now where we're trying to bridge the gap between what you're saying and then what actually is probably used and making sure we're overestimating more than underestimating because you're going to get likely better results that way than underestimating. So the good news is you need to use the method that you prefer.
Speaker 1:Parrot is the only app on the market that does all of them in one easy place. You don't have to be super precise with tracking to lose weight, it really doesn't matter that much, it's about the consistency and you might think no way, because you've been drilled into your head that you have to be turbo specific with your weight tracking and stuff like that because it comes from bodybuilding. To get to 5% body fat, of course you do, you've to be militant. To get onto a healthy body weight you don't have to be like that, that's the fact. People don't like listening to that, people in the fitness industry don't like it because their entire culture is relied on this super precise stuff and meal preps and stuff, but the truth is I've seen thousands of people lose weight by not being super 100% accurate, by just being consistent with it, and you'll get results.
Speaker 1:So make sure you use all the tools, Just to let you know, you can take a photo of a plate and it'll track it for you. Keep learning, keep using the tools, keep being flexible as well. Some people get really frustrated by saying stuff like, oh look, I put in a Luxispo bottle and the AI says 140 calories but it's actually 138 calories. It's like don't worry about the two calories, the AI is going to show you a slightly rounder number, don't worry about exacts, don't get stressed out about them, don't think I was pointless doing it, that's not the point, you're missing the point if you're thinking this way. You're going to fall down the trap of when you can't track then a weekend at all or you feel like it's not been as perfect as you think you need to be, you stop entirely and you don't get results.
Speaker 1:So make sure you don't fall down that trap. You can just edit it, change it to 139 is no problem, take you two seconds. You know you can do these things easily, can adjust them or whatever, you don't have to like go to the group and message and say oh this is too by the time you said that and in the group you could have just changed it yourself. There's no perfection in tracking. Food barcodes get updated, foods get updated, new formulas get pushed, all this stuff always happens.
Speaker 1:Some factories make the same brands but then they mix up on the barcodes and you might scan something and it says it's like one brand but actually another brand but it's the same actual food. You get a lot of Tesco strawberries, comes up as little strawberries or whatever. It's probably done in the same place. That's why there's mixed up air and stuff like that. So don't worry too much about the small things because it's going to lead to your downfall basically and you're going to get demotivated.
Speaker 1:So that's it for today, hope that was useful, use the tracking methods, let me know your favorite method and I'll speak to you tomorrow.