top of page

How important is braking in mountain biking?

I was never the most fit guy at the races, but by using my brakes smarter than others I was able to put together some good races. My strategy was always to minimize the time I lost on the climbs and to make up time on every corner and every descent.

10 years ago, power meters weren’t popular in MTB, but we really didn’t have any great way to compare ourselves anyway just by looking at power.

OK, you’re normalized power was this and mine was that. So….?

Well as it turns out, pedalling hard is only one important aspect of mountain biking–it’s no wonder power output only has a pretty good relationship with performance on an XC lap. Time and time again the research has backed this up.

This isn’t to say that pedalling isn’t import—it is very important—but in mountain biking especially, there are so many factors working for and against each other for speed.

I always ask this one important question: What wins races?

Speed wins races.

And how do you go fast? Well, lots of ways.

Anyone who has every mountain biked would probably know that braking is important. But how important?


In the first ever study using a brake power meter (well, second actually. The first one proved that the device worked!), we set 10 riders off through a short XC lap. There were three short climbs and three short descents. We measured their propulsive power (the power on the pedals using a standard power meter), their brake power using our newly invented brake power meter, and recorded the time it took to complete the lap.


OK, the world’s first brake power meter is a bit of a dinosaur by now, but it did the job!

It turned out that average propulsive power was strongly related to lap time. That is to say, those with higher [relative] propulsive power output we able to go faster around the track. But this was no surprise.

We also found that braking was important, too.

With combined measures of braking and propulsion, we could explain more of the variation in lap time than pedaling alone. Science calls this a regression equation and it was the first time we saw the important of braking and pedaling together.

What this equation meant was that by measuring brake power AND propulsive power, we had a much better idea of what was going on out on the trails. Namely, a combination of pedaling hard and braking for a short time went along with really fast lap times.

Riders had the same lap time every lap- but how did they do this?

In our test, riders did 3 laps and there was no statistical difference in lap time between the first, second or third lap. However, when we compared averages of propulsive power, this was higher in the first lap and then dropped off for laps 2 and 3. That is to say that they did not pedal as hard on lap 2 or 3 than in the first lap.

But laps were the same time–how could this be?

This answer came after looking at the braking, and results showed how braking changed across laps.

By lap 3, riders had significantly changed how much braking they were doing—both brake work and brake time were reduced! This reduced in braking appeared to counterbalance the reduction in propulsive power.

In other words: by changing braking, riders could go just as fast around an XC lap even if they didn’t pedal as hard!

Brake Power and Propulsive Power Graph

This rider’s average propulsive power was 328 W on lap 1 and 306 W on lap 3. He reduced his braking from 20,300 J to 18,500 J and was able to maintain his lap time to within 0.5 seconds. You can see where he saved time by braking differently.

To me, this one of the coolest findings. And on a personal note, I thought this did a pretty good job at explaining why I was able to finish with XC riders who were much more fit than myself. I couldn’t pedal as hard, but was able to make up for some of this by braking better.

At the moment, it isn’t totally clear why riders changed their braking. This might have something to do with learning the course at speed, though they did get a few laps of practice before testing started. If you look at the above figure there are a few instances of braking that were not present in lap 3, so this gives a good idea.

So how much braking do we actually do?

In our group of well-trained mountain bikers, their average propulsive power was 275 W across the entire lap. This included uphills, flats and -if it happened- on the descents. Across the ~5 minutes it took to do one lap, this was 85,000 joules of energy expended for pedaling.

Braking on the other hand was primarily done on the descents. The energy removed during braking was 21,000 J.

Riders used the brakes about 1/4 as much as they pedaled. That’s a lot!

The time they spent braking was short overall at 28 seconds per lap.

Brake power was calculated using the brake work divided by brake time. This was 800 W as an average during the lap. The fastest riders were doing all of their brake work in a short window, which meant that they had the highest brake power.

Of course, this is course dependent, especially since some trails might be very very fast and could require you to slow down almost to a stop while other may be slow and require minimal slowing.

It’s interesting because we can easily feel that work going through our legs to the pedals, but we can’t feel what we are taking away with the brakes. Holding 800 W while pedaling would be pretty hard, but we can do this same thing with the brakes without much thought.

But let me ask you this: how many 800 W (or 200 or 400) intervals have you done? Probably many.

It’s time we start thinking more about our braking.

It’s not just in the numbers

This was the earliest braking research, and we’ve come a long way since then.

Some of the more important things we’ve learned is that it’s not just how much you brake or how long your brake, but where you brake that is important.

You can brake really really hard for a short time, but if you do this at the wrong spot, you could be doing it better.

BPM helmet cam

Linking brake power with a helmet camera is one of the most telling ways to use the data.

A lot of this we learned by linking braking data to a helmet camera, but also in a number of studies that followed on from the original one.

More on this soon. And more

Until then, check out these other articles:

bottom of page