Browsing posts in: Training 101

Heart rate monitors and calories…

There are a lot of devices out there that include heart rate monitors, from stationary devices like treadmills and ellipticals to fitness bands to bicycle computers. People often ask whether the calorie measurements are any good, so I thought it would be useful to talk about how they determine calories and what you can expect in terms of accuracy.

Metabolic testing

The gold standard for measuring calorie burn is through metabolic testing. You exercise on a treadmill or exercise bike wearing a mask that collects all of the air that exhale. By analyzing your exhalations, a very accurate measurement of how many calories you are burning can be determined.

The test is usually one that increases in intensity, and along the way, your heart rate is recorded. One of the results that you get out of the test (there is a ton of data collected, including your V02Max, a measure of your overall fitness) is a table that records your calorie burn rate at various heart rates.

Heart Rate

Calories per minute
118 0.117
135 0.173
144 0.199

In a real test, there would be a lot more data than this.

I can put the data in a graph, and do a little math on it:


The line through data point is a calculated curve it through the points. In this case, it’s a straight line, but in the real world, a curve will probably be a better choice.

We also get a formula out of the calculation:

Calories/Second = Heart rate * 0.0032 – 0.2568

If I know the heart rate on a second-by-second basis, I can use the formula to estimate the calorie burn for that second. If we were doing an experiment where we wanted to measure the calorie burn of a group of subjects, we would do this for each of them, and then we could just ask them to wear a heart rate monitor during the experiment, and use their personal equations to calculate calorie burn. This is a very common approach for exercise studies.

Note that we are predicting the total amount of calories burnt by the person at a given heart rate, not the excess number of calories due to exercise.

Heart rate monitors

Enter the general-purpose heart rate monitor. They would like to provide you a measurement of your calorie burn, but they have a serious problem – they do not have results from a lab test to use.

What to do, what to do?

Well, what they do is they recruit a bunch of volunteers, and they put each of them through metabolic testing, average all the data together, and come up with a single equation.

This does not work well. Not well at all. There is a ton of variability among people; some have large hearts, some have small hearts, etc. This means that the prediction is not very good.

So, they start collecting more data about each of the participants – what is their age, their height, their weight, their sex, their exercise level, the breed of their dog – and they feed this into some statistical software, and they end up with what is known as a model, This is just a very fancy equation where you plug in the demographic information (age/sex/etc.) and the current heart rate and you get a prediction for calories per minute. Add that up during the exercise, and you get total calories burned.

Sources of error

So, how well does that approach work, Well, the answer is *it depends*.

The following scenarios result in error in the prediction:

  1. Being dehydrated increases your heart rate because your blood volume decreases.
  2. Being sick or especially tired makes the prediction less good.
  3. Being non-average makes the prediction less good. If you have a larger or smaller heart than average, or you are more or less fit than average, the prediction will be poor.

Generally speaking, the more data your fitness device takes in, the better a prediction it can make. If there is no data entered, it will probably be pretty bad. Add age and sex, and it gets better. Add in fitness level, and it gets better still.  Some fitness devices even let you enter the results of your metabolic test, which can give you decent accuracy as long as your fitness level remains the same.

How bad is the error? It’s really hard to quantify.

If, for example, you have a small heart (high heart rate) and you are very out of shape, it might overestimate your calorie burn considerably. I’m not sure what “considerably” means, but I would not be surprised to see 50% or even 100% more.

On the other hand, if you have a large heart and you are in very good shape, the device will likely underestimate your calorie burn considerably.

Calorie Inflation

Manufacturers are exercise equipment (treadmills, elipticals, etc.) tend to use pretty simple models; they often don’t even use heart rate, and if they do, their models aren’t very good. They also benefit with more sales if people are happy about their exercise, and a higher calorie burn does that, so machines tend to inflate their calorie burn numbers a fair bit – perhaps up to 30%.

I don’t know whether heart rate monitors.

Direct measurement of calorie burn

There is one way to get a great estimate of calories burnt doing exercise, at least for one specific situation.

If you have a bicycle with a power meter, the power meter measures how much force you are putting into the pedals, and therefore your bicycle computer can calculate how much work you did in kilojoules. You can then *roughly* convert that into the number of calories burned and get a decent estimate – way better than what you would get with a heart-rate measurement.

Training #3–Specificity

In the last post, I talked about the importance of achieving overload if you want to keep improvement, and talked about “spikiness” in workouts. You can get spikiness in different ways. The first thing that springs to mind is to segment our workouts into hard days and easy days. Will this help?

Well, it will let you get some overload when you start, but my guess is that over time, you will get used to the hard days, and you will plateau again.

The answer is “specificity”, which just means that you are going to focus a specific workout on a specific area. By focusing on that area – and getting sufficient recovery between workouts in that area – you can continue to generate overload, and continue to improve. There are literally hundreds of different kind of specific workouts you can do; here are a few:

  • 20 minute time trials (threshold repeats)
  • Hill repeats
  • Over/under intervals
  • Muscle tension
  • Single-leg drills
  • Cadence drills

Now, we have a different problem – the problem of figuring out what areas you need to work out, what workouts are most appropriate for those areas, and how to fit them all together along with appropriate recovery so that the program is just the right intensity – not too easy, and not too hard.

The best answer to this problem is “find a coach”, but I’ll share some ideas in future posts.

Next time, I think I’ll talk about macrocycles, mesocycles, and microcycles.

Training #1: Overload + Recovery = Improvement

The first in a series of posts on cycling training. 

Humans are quite adaptable, and we take advantage of that adaptability when we work out. Or, we *try* to take advantage of it, but we often don’t do a very good job of it, because we misunderstand how things work. It’s really quite simple:

Overload + Recovery = Improvement

Overload means applying training stress that exceeds what your body is already accustomed to.

Recovery is the time after the workout when you get better.

Improvement is when you get better.

When you first start riding, this is easy to do; you aren’t used to riding at all, so pretty much anything that you do overloads your system, and you improve. Over time, your body adapts to the way that you are working out, and your improvement plateaus. If you keep training the same way you were training in the past, you won’t see the improvements that you are looking for, since you are no longer generating any training stress.

In future posts I’ll explore ways to get more overload, and how to improve recovery.

You’re working too hard

I get the opportunity to see lots of new riders on our group rides, and there’s a common thread that shows up over and over.

Nearly everybody who is new is working out too hard, and many people are working out too much. What do I mean by this?

Well, to get better at cycling, you need to focus on both your anaerobic and aerobic systems. But the way you focus on these is totally different.

Your aerobic system training is best done at a fairly light intensity – one at which you can still talk comfortably. This is sometimes known as “base miles” or “LSD (Long Steady Distance)”. This sort of riding helps your body get better at getting oxygen to your muscles, better at using your stored flat, and (very importantly) builds up your muscles and ligaments to deal with increased loads. Pros and races put in 1000s of miles at these sort of intensities.

I think of this sort of riding as setting your baseline.

Once the baseline is set – which is a 6-8 week process (or more) – then you can add some intensity to the mix to work on the anaerobic system. This involves interval work – things like 1 minute all out, 1 minute recovery, repeated 6 times, or hill repeats, where you do the same hillclimb over and over. The details are a subject of another post, but these workouts increase your ability to produce power, and your ability to deal with short exertions and recover quickly. And they’re very small in quantity – you might only do 2 sets of 4 intervals in a workout.

This is also the time to add in tempo rides. Tempo rides are done right below around your lactate threshold (another big subject, but where your legs start to hurt when you’re out of breath is a decent starting point) and last from 15 minutes to perhaps 30 minutes. Tempo rides improve your ability to deal with lactic acid, so that you can ride at a higher heart rate but still staying aerobic.

What I see in beginning riders is working out as hard as they can for the whole ride. And that works fairly well when you start riding – frankly, pretty much anything works well when you start riding – but it has problems once you start to get some fitness. That pace is fast enough that there is a significant anaerobic component, which is much faster than optimal to train your aerobic system. And, even though it feels like you’re going really hard, you are riding too fast to be able to do your intervals all out.

So, if you want to get better, you need to get in the base miles, and then you need to have focused workouts of very high intensity with limited duration.