There are various ways to do this. There are devices that you wear all day and night to track calorie burns. The problem is wearing the device 24 hours a day is a bit tedious. Not to mention that I'd probably obsess about every single calorie and spend way too much time checking calories during the day. I think for some people this may be a good way to go, particularly if the person doesn't really have a workout/exercise program to go with a weight loss program.
What I really wanted to do was to try and calculate how many calories I burned on a bike ride or run, and get reasonably accurate figures. I checked out many calculators available on the web. Most are based on your weight, the time you spend on a particular activity and sometimes even your average heart rate during that activity. Because of the general nature of these calculations, they are really only a guess as to what you might have burned during that activity. Most of the time, I discovered that the web based calorie burn calculators would overstate calorie burns. For example, when I did the Tour of Palm Springs Century, these calculators figured that I burned off well over 10,000 calories (It should have been closer to 3,500 for me)! If I had based my caloric limits on this, I'd be gaining weight instead of losing it.
Some of the burn calculators take into account your VO2 max (see below or click on the link) if you know it. A calorie burn is then calculated based on your average heart rate. These calculators are much more accurate than the general calculators that only calculate on weight, age and time of activity. Oftentimes, the calculated calorie burns with a known VO2 max calculator will come in at half of the ones using more general criteria.
To get decent accuracy in calculating calorie burns, you need at least a heart rate monitor (HRM) during a workout activity. Even more accurate would be a power meter, but they're expensive and probably not worth it if all you're doing is figuring out calorie burns. Over the last couple of years, I've used various Garmin GPS units, more particularly the Edge 500 and Forerunner 210, Polar units (FT-4 and RS300) and the new Motorola MotoActv. These units have the ability to constantly calculate calories during a workout based on your heart rate, instead of relying on an average heart rate. The HRM is key though; these units will generally indicate burns 2x higher without than with a HRM.
The first thing you'll notice when you set up one of these units is that you don't burn as many calories as you think. For an average person, any of these units are fairly accurate, and my research indicates accuracy to 10% to 20% is common, as long as you use a heart rate monitor. Comparing calorie burns with a web calculator that takes known VO2 max into account, on the average, these units can return a calorie burn about 60% less.
VO2 max is a measure of your body's ability to transport and use oxygen during a workout. In other words, it's a measure of how physically fit you are. Calorie calculations in the aforementioned units are usually based, among other things, on VO2 max, your age, Body Mass Index (BMI) and your heart rate during exercise. All of the HRMs I tried would calculate and "estimated" VO2 max based on age, weight and heart rate parameters in order to arrive at a calorie calculation.
You can stop reading here, if you are more or less average. If you're more like me and your heart rates tend to be slow, read on...
The problem for me was that because my heart rate tends to be slow, these units would calculate really low calorie burns. (Relying on low calorie burns can be dangerous... You'll end up taking in far fewer calories than you require for a healthy weight loss)
The HRMs I tried determine your calorie burns by estimating your VO2 max and calculating a calorie burn based on that VO2 max, your heart rate, Body Mass Index (BMI), age and gender among other things. The Garmin units in particular use an algorithm developed by a Finnish company called Firstbeat that also uses Activity Class into account (particular only to Garmin HRM units). The Firstbeat algorithm predicts your VO2 max according to a formula (Jackson, et. al.). For someone like me, 57 years old with a relatively low heart rate the calorie calculation comes in way too low:
VO2max = 56.363 + 1.921xActivity_Class - 0.381xAge - 0.754xBMI + 10.987xG
Activity class: 1-10 (although 7, 8, 9 & 10 are supposed to be equivalent; however it does seem to change the results)
BMI=weight in kilograms divided by the square of your height in meters.
G=1 for males and 0 for females.
In my case, my predicted VO2 max was coming in at around 29, using my correct age (57), BMI, etc. My VO2 max is supposed to be around 47.3. To "reverse engineer" the formula so that it calculates your correct VO2 max (and therefore calories) you can fiddle with activity class (raise it), weight (increase it, but not recommended because the weight is used in other calculations), BMI (decrease it by increasing your height). Over the last year or so, I've had to keep increasing my height (now at 7'6") in order to keep up as my fitness level increased.
So how do you figure out your VO2 max? The simplest and most accurate way is to get tested at a sports clinic that specializes in fitness testing. These tests cost about $150 to $350 on the average, and you'll need to get re-tested as your fitness level changes. You can also estimate your VO2 max using various methods by timing runs, rides, swims, etc. A search on the web for "estimate VO2 max" will provide a lot of options without spending the $$ for a fitness test. Not as accurate as a true fitness test, but close enough for my purposes.
My Garmin units still calculate calories on the low side, but within what I feel are acceptable limits. I didn't have do to much "reverse engineering" with my Polar units, although the FT-4 read a bit high and the RS300x was close to the Garmin reading after fudging the Garmins. The MotoActv seems to be the most accurate on runs and other activities, but the calorie burns in the cycling mode are ridiculously high, up to 4x higher than the Garmins. On a particular bike route that I use often, the MotoActv will say I burned 2,700+ calories in the cycling mode; in the "other outdoor" mode, on the same route, I'll only burn about 800+, which is about the same as the Garmins. I still haven't figured it out the MotoActv. I've sent data to Motorola and haven't yet received a solution from them. My riding buddies don't seem to have the same problem with the cycling mode on their MotoActvs.
Update: I discovered quite by accident that the Garmin Speed and Cadence sensor on my bike accounted for the crazy calorie numbers reported by my MotoActv. During a recent ride, the little magnet slid away from the sensor and stopped recording cadence and wheel speed. Suddenly the crazy high calorie burn numbers on the MotoActv dropped dramatically and my Garmin Edge 500's seemed to come up just a bit (although I have no empirical evidence to support the Garmin's increased calorie burn reporting). The bottom line is that the MotoActv and the Garmin, when used side-by-side during the same activity now report virtually the same calorie burns.
Was it worth all the effort to get calorie calculations that are at best only 80% to 90% accurate? Close enough... The proof is in the weight loss.
(Again, calories calculations only relate to weight loss. You still need to pay attention to the food you eat, e.g. sodium, carbs, proteins, etc. for a healthy diet)
