Are the calorie counts on elliptical cross trainers accurate? I am a 5'7", 135-pound woman, and the readout says I burn between 14 and 17 calories a minute! This seems to good to be true, even in a high-intensity workout. And what about on other cardio machines, for that matter?

The calorie counter on any device is, at best, an estimation of your work output. Some machines make more accurate estimations, some less accurate. At the minimum, the machine's information panel must allow you to enter your weight, since the bigger you are, the more calories you burn -- no matter what you're doing. If the machine doesn't ask you to enter your weight, the information is useless.

Even if the machine factors in weight, it still can't account for variations in muscularity and cardiovascular efficiency between people of similar weight. If one person can perform an exercise more economically than another person of the same weight, the first person will expend fewer calories per hour than the inefficient exerciser, assuming that the workload and duration of the exercise are consistent.

I performed a highly unscientific experiment my gym. I input your weight and mine to see what the "Calories Per Hour" (cal./hr) display on our treadmills and steppers would read.

At 3.0 miles per hour (mph) and no incline, a person weighing 135 pounds (you) would burn 206 cal./hr. A 160-pound person (that's the default setting on our treadmill) would burn 244 cal./hr. And a 225-pound person (guess who?) would burn 343 cal./hr.

If we bump the speed up to 4.0 mph, you would burn 373 cal./hr. A 160-pound person would burn 443 cal./hr. I would burn 622 cal./hr.

This pattern repeats itself when holding speed constant and adding an incline to the equation. An investigation of our steppers yielded similar results.

None of those machines factors in individual fitness levels and the specificity of the exercise. To examine that, I wore a heart-rate monitor while walking and stepping. When my heart rate leveled off at a particular workload, I noted what it was. Then I held on lightly while exercising on each type of machine. My heart rate dropped by a few beats per minute. Then I leaned heavily on each machine the way you see the cheaters in the gym do. My heart rate dropped by as many as eight beats per minute, which represented a five percent drop in my training heart rate. That's a big dip in output. How would you like a five percent cut in pay? So if you crank the intensity up so high that you have to hang on for dear life, the calorie-output display can't be accurate.

Let's say two people of equal weight use a stair-climbing machine at the same workload. One person usually runs for fitness. The other person cycles outdoors and uses the stair machine exclusively indoors:

We know that running primarily recruits the lower back, hamstring and calf muscles. Cycling is more dependent on the quads as well as the calves; stepping uses those muscle groups as well. The runner probably has the aerobic capacity to perform the exercise but may have to work harder at that level of intensity than the cyclist, who is calling on the same muscles while stepping that she uses when she rides her bike. How accurate can the calories-per-hour display be if it reads the same for both athletes?

The numbers that you see apply only to you, and only for that particular piece of exercise equipment. You can use the readout to compare workouts on that machine, but don't try and correlate the information you receive from two different types of equipment. Use a heart-rate monitor instead.

Got a fitness question or comment? Post it on the Fit by Friday message board!