22:23
The other day my heart rate reached almost 200 beats per minute, and theoretically, just theoretically, I burned almost 800 calories. I'm dying here, but 800 calories in an hour of tennis? This can't be true. So today I'm back with this. I've bought every smartwatch, tracker, everything on the market, and we'll find out which one is right. Come on. Dozens of universities have confirmed that smartwatch calculations can be very inaccurate, yet companies continue to add new health features. So are they just lying to us, or how do they actually count these calories and how accurate are they? That steak I eat after the match, did I really deserve it? Did I really burn it? Is my diet a lie?
I've been obsessed with this for the past few weeks, not only because I keep losing at tennis, but also because the health of at least 500 million people now depends on these devices. And not only did I find which one counted calories correctly, but in the process I discovered a much bigger problem in how we use smartwatches. So let me, break this down for you. Okay, let's see, let's see what they charged. 660, 6780 here. Fitbit-670. So, how much difference do we have? 200 calories between some of them. This is madness. Why are they, why are they so incredibly different? Here are my results. Some of them deviate by almost 20% from the average. But the main question is:
should we trust the average? In this video, we ran some very in-depth tests to find out, but first we need to understand what exactly we're trying to measure. So, in simple terms, a calorie is the amount of energy required to heat 1 g of water by 1 °C. Of course, this is a purely metric system. Usually, in food, we operate in kilocalories, which is essentially the same thing, but it is the amount of energy needed to heat 1 kg of water. So, even though we're dealing with kilocalories, food labels simply say " calories," which is a little confusing, but rest assured-they're talking about kilocalories. This is not very useful. How does this affect my belly fat or the effort I put into my workout? This is because every time you contract a muscle,
your body expends physical energy to make it happen. And calories are literally units of measurement for this energy. To find out the number of calories in a pizza, steak, or holiday dinner, you simply place the food in a thick-walled steel container called a " bomb." You fill this thing with oxygen, which is very important, and submerge the whole thing in water. So, you set fire to the contents of the " bomb" and measure how much the water around it has heated up. This will be your accurate and flawless calorie indicator. And how do you know if your body has burned those calories? process. You immerse your entire body in the water. Back in the 1890s, chemist Wilbur Olin Atwater created the first human-sized version of such a camera. They locked people in this
cell for several days, forced them to read, iron clothes, or pedal an exercise bike, and By measuring exactly how many degrees the water temperature rose, they could track the exact amount of heat released by the human body. These experiments have become the foundation of our understanding of metabolism, and that's great, but that's not what we're here for example, I wanted to find out which watch really shows the correct data. I was actually wondering if I was exercising enough to burn off that steak I ate, or that Milky Way for dessert. The Milky Way package lists 240 calories. So when Mars wrote that 240, they actually tested it, right? They did an experiment with a "bomb", burning it in water. Or not? Of course not. Atwater, a scientist who conducted
experiments with the camera, came up with a simpler way to estimate the calorie content of food. He found that 1 g of carbohydrates contains 4 calories, the same as 1 g of protein, and 1 g of fat contains 9 calories. That's all. But how accurate is this system? Plus, we no longer put people in water tanks. The Apple Watch doesn't have a thermometer, although some others do. So how do they know if I actually burned those calories? It seems like we are just making estimates based on other estimates. Today we know that the "4-4-9" method is already somewhat outdated. For example, this study found that for foods like nuts, the Atwater method overestimates calorie counts by 30%. Here's a good reason to eat Milky Way instead of Snickers
. Snickers is disgusting, by the way. The thing is, when you eat something like almonds, you have to work at it. You need to burn calories to digest it. But once the food has been processed at the factory, half the work is already done. This is a centuries- old guess. It has been criticized for decades, but it still remains the main way to label products. But let's get back to the watches. How accurate are their measurements if we are not immersed in water? Here's a breakdown of the sensors in each of these devices. Many of them don't even have thermometers. So how do we get these estimates? The Apple Watch, for example, will try
to estimate calories by taking your baseline readings. This is your age, your weight, your height. He will introduce them to this massive sensor data fusion system. Your heart rate, which is read by that green light, GPS data about speed and movement, an accelerometer to track arm swings-it all goes into the "black box" of the formula that tries to estimate the number of calories you've burned. We don't know what this formula is, but we do know that it doesn't see the real reactions that are happening in your cells. For example, a large-scale Stanford study tested seven of the best fitness trackers on the market.
Even the most accurate of those tested was 27% wrong. The worst one overestimated calories burned by 93%, and that's assuming the other measurements are correct. Tattoos and dark skin can distort Moreover, this was measured under controlled conditions on a treadmill. Running is a sport for which there is a lot of data, but in tennis, with my dark skin, I am the worst test subject for it. But I have a better option for you. This is my good friend Diana, and she is definitely the strongest athlete I know. She finished 55th in the marathon at the 2025 World Championships in Athletics. She holds the Central American marathon record and agreed to conduct this test for us.
I usually wake up at 4:30 in the morning. I start running at about 5:30. In a day, I usually run, uh, Today I'll run at a steady easy pace and see what happens. So, Diana prefers the Garmin over the Apple Watch, and the Garmin theoretically has a few extra features for more accurate measurement. For example, they have HRV, which measures the intervals between your heart beats. So, if your heart rate is steady, it means you are at peak performance, at maximum effort, and if it is uneven, you are warming up or recovering. They acquired this company to gain access to this algorithm. So he should be better, right? To test the devices, I ran 5 km in about 21 minutes. Apple Watch, active calories: 307.
Fitbit: 219 calories. Garmin: 345 calories. Coros: 204 calories. These two devices should be closer together due to the location of the clock. Pixel Watch: 290 calories. Coros Pace Pro: 204 calories. Coros Apex: 205 calories. Oura: 157 calories. Those closer to 200 calories seem most accurate, since it was about a 20- minute run at a moderate pace, not at the limit of effort. So, numbers closer to 200 calories seem most accurate, to be honest. In Diana's measurements, the Garmin readings were the furthest from the average. And if you look at both tests, a pattern begins to emerge: Apple Watch and Garmin overestimate the results, while Coros and Oura Ring underestimate them relative to the average. But again, we don't know the correct number, and we can't submerge
both of us in water to measure the temperature change. So what alternative do we have for measuring calories burned? And that's the fundamental problem with every watch, isn't it? They use algorithms and tools to do human work, which is the opposite of what happens in a sales department. You are using humans to do the work that algorithms should be doing. Building lists, defining your target audience, writing emails that don't sound like another cold email spam, or following up after a no- response. I used to do a lot of this outbound stuff at Slidebean and I remember what it looked like. These were not calls, not demos, but everything that preceded this.
That's why Artisan created Ava to take all of that off your shoulders. Ava is the first autonomous AI- BDR. It searches over 250 million verified B2B contacts or takes them from your CRM and investigates each lead individually. She writes a personalized email based on this research. It processes responses, handles objections, and books meetings directly in your manager's calendar. And during the sales process, the built-in auto-dialer connects the representative with the customer as soon as they pick up the phone. So they spend time talking, not dialing. Calls and deals are left to your team-it's work that only humans should do. You may know Artisan from their " stop hiring people" billboards. They abandoned this slogan because the goal was never
to replace people. This is Ava doing the algorithmic, work part so your managers can be more human in the conversations that really matter. Avy is trusted by thousands of sales teams at companies like DoorDash and Grammarly, and you can hire her at artisan.co. Link in the description or scan this QR code. Okay, back to calories. Of course, our body is not just about heating water. Since we don't have a furnace in our stomachs, we burn food through a chemical reaction, which can be used to measure calories.
Simply put, everything we eat-proteins, carbohydrates- is broken down in our bodies into energy units called ATP. Now there will be some chemistry. Don't worry, just the basics needed to understand why these watches are so wrong. Take glucose, for example. All the sugar in my Milky Way bar will eventually turn into glucose. After receiving glucose, our body breaks it down again into two molecules called pyruvate. This splitting gives us two units of energy, that is, two units of ATP , plus a free hydrogen ion, which will soon be in our way. I use lightning to denote ATP. ATP is a molecule similar to these, but I won't go into that in this explanation.
Glucose and pyruvate contain much more energy, and this process is not the best way to obtain it. The best way is to add some oxygen, if available. Fortunately, we constantly breathe oxygen. If you have oxygen, your cells can convert each pyruvate into another molecule called acetyl-CoA. And each of them, if everything works, will produce about 15 units of ATP. So this whole reaction will produce about 30 units of ATP, so 30 plus two, and since we used some oxygen, we'll have some CO2 produced. Six units of CO2 emissions are actually not that different from your car. Theoretically, if these calculations are correct, if we somehow measured how much O2 we inhaled and how much CO2 we exhaled, we could measure calories fairly accurately, right? OK. So
let's do it. I don't know if I can reach that speed, but.come on. So, Diana passed the test first. We measured the actual number of liters of inhaled and exhaled CO2 and O2. And then, using Weir's equation, we take the actual numbers and calculate that during this exercise, Diane burned about 140 calories, which is much, much less than any of these devices were showing, man. Garmin was 63% wrong. But then it was my turn, and we discovered another problem. During training, there are two key points related to burning calories, called aerobic and anaerobic. When your body performs aerobic exercise, it uses oxygen to efficiently produce energy, as we saw in the previous process. You are breathing normally, the standard energy splitting process
is working normally. But as you train more intensely, there comes a point when the aerobic system simply can't keep up. This is because your muscles need so much energy and so quickly that the body is forced to switch to anaerobic exercise to produce ATP faster. And that brings with it a lot of bad news. Anaerobic means that some of the energy is produced without air, right? You don't use oxygen for another process because your cells can't keep up. So your body bypasses this process and just starts breaking down glucose into pyruvate to make two ATP molecules, which isn't much, right? So you break down more glucose molecules, which produces additional pyruvate, and so on. But it also produces a lot of excess hydrogen ions. And your body, your blood
will begin to fill with them. So you get energy, but you also get hydrogen. The problem with hydrogen is that it makes your blood acidic. Your body will try to counteract this, and for example, pyruvate will combine with hydrogen to create this molecule called lactate, but it's not enough. You may have heard the term "lactic acid" and how it is what causes muscle pain after a workout. But in reality this is a myth. Lactate is actually trying to get rid of the excess hydrogen ions that build up in your blood as a result of this process. So, lactate is actually good, but not good enough. So some of these ions will remain in your blood, and you will have to find a
way to get them out. Not to mention that you're acting very inefficiently, aren't you? You break down glucose and only get two ATP molecules per glucose, when you could get up to 32. There's another problem with the anaerobic process: while you can break down glucose to make ATP without oxygen, you can't do that with fat, which is also in a Milky Way bar. Under favorable conditions, you could break down a fat molecule into eight acetyl-CoA. This is a large amount of energy. However, you need oxygen. Is n't that right? So you could split the fat into eight parts. And if you were to break that down to ATP, you would get about 120 ATP molecules. Breaking down fat is
very beneficial. Plus carbon dioxide, of course. But this is only possible in the presence of oxygen. You cannot break down fat during anaerobic exercise. So, when the doctor increased the speed on the treadmill, my body started to give up. I went from aerobic to anaerobic at about 9 km/h. This is not very high endurance for a 37- year-old man. For comparison: Diana, who is also 37 but is a professional athlete, had to reach 16 km/h on the track before she went into an anaerobic state. Important note: most smartwatches have a system for measuring these indicators. They are usually called training zones. The problem is that these zones are based solely on your heart rate. But it's not your heart rate that determines whether you're going into
an anaerobic state. The pulse is just a symptom of what is happening in your body. What exactly made me go into an anaerobic state is visible in this diagram. The blue dots here indicate the volume of exhaled oxygen (O2 ), and the red dots indicate the volume of exhaled carbon dioxide (CO2). Notice: at first I exhale more oxygen because I am training efficiently. I burn fuel aerobically, but after a certain point, things change and I start to exhale more carbon dioxide. The moment when they switch places, when there is more carbon dioxide, is the moment of transition to the anaerobic state. My body produces so much carbon dioxide that I have to desperately blow it out to deal with it. So my breathing becomes faster
because I need to get rid of the CO2 in my blood. Now the excess CO2 and all the hydrogen ions have entered the blood through this anaerobic energy equation. My blood is becoming acidic, which could kill me. My body knows this. Pyruvate is converted to lactate. This is just an attempt to help a little, but it's not enough. I try to add some bicarbonate and the body does it. He adds bicarbonate to reduce the acidity with hydrogen, but this also creates excess CO2, which I have to get rid of with my exhale.
I'm working at my limit, I'm hyperventilating, and of course, I'm exhausted. So that also explains why I struggle in tennis: if I go into an anaerobic state too quickly, I gasp for air, my decision-making deteriorates, and I start doing stupid things on the court or forgetting to set my calorie counters to see which watch is working properly. I mean just a few, but yes. The ones I've tested show overestimated calorie results by at least 22%. Now that the test is underway, my body is ready to "shut down." There are a lot of H+ ions in my blood. I threw out the bicarbonate to try to clean them, but it created CO2 and I don't have time to get it out. So before I reach a truly dangerous level, my brain will stop the process. I have to
admit it because the doctor stopped the test. Watches have something very useful for this: after all, training is about developing an aerobic base, that is, working at a certain level of effort or tracking your heart rate to stay in that zone. So, with these test results, I set up my Apple Watch. Almost every smartwatch allows you to set heart rate zones. Again, these will be different for everyone. These are just my figures. But, theoretically, I'm just training-jogging, swimming, and rowing in the second zone, which, to be honest, is a bit boring because it's a low-intensity activity. You don't get very tired, or I reach 135 beats per minute very quickly. So, I'm training in a certain zone, and theoretically my body should become more efficient at producing aerobic
energy, allowing me to do more exercise without going into an anaerobic state for a few weeks. Just don't think I've forgotten why you came here. This video is not clickbait. We'll still find out which watch counts calories correctly. We have a portable VO2 mask that I can wear during the game. We have all the watches. This thing will measure my CO2 and O2 levels, so we can get more accurate calorie data to compare with the watch. By the way, now every Thursday after each video we do a live broadcast. So, if you want to ask additional questions, this will be your opportunity. Let's see how it goes. You can breathe in this thing. Everything is fine. Just can't see the volume, dude. It's kind of like here. It's not that I don't know. Do you
see anything in my video? Hmm. At least this watch shows 1160 calories, and this one shows about 600. The variance is huge, but we can do the math to find out which number is correct. And here are the results. 767 on Apple Watch. 769 on Fitbit. 1250 on Garmin. 739 on the Google Pixel Watch, 662 on the Coros, and 654 on the Oura ring. But which one of them is right? So, the official calorie figure from VO2 Master, based on oxygen consumed and CO2 exhaled, is 662.1 calories. This is exactly what Coros showed. It's crazy that in all the tests Garmin was so far from the truth. This test had almost double the actual number of calories. Most other watches mostly overstated the readings by 10-15%. Oura Ring is consistently underselling. But anyway,
given the "black box" of their formulas and what they're selling, I don't know what you think-should they be more accurate? I looked at the Coros website to find out their secret, but to be honest I didn't find any secret. Their support page doesn't mention any better algorithms than other watches. Some studies show that Coros sometimes underestimates calorie burn. So there is a factor of certain conditions, sports settings, my skin color to determine the pulse. So if anyone from Coros is watching this, first of all, congratulations.
Contact us and tell us how you did it. But more importantly for you, first understand that it's not just about calories. Your physical fitness, the number of calories burned-there's another level here related to aerobic and anaerobic exercise. So, use your smartwatch to understand these zones. Keep in mind that calorie readings will be off by 10-20% unless you use a Garmin, in which case they will be twice as inaccurate. So, divide in half. This Garmin error reminded me of the VAR system, which has another very critical error. We covered this in a video a few weeks ago. You should watch it. See you in the next video.
