... ... Power meters for trail runners - Mile27
Mar 142018

What do you use to determine the intensity or pace in races or training on trails?

For road runners pace is a good a metric as any to work from but pace on trails varies too much for it to be of any value.


The Stryd Power Centre

What about heart rate? Heart rate training has been around for a while and has some merits but also some limitations. It is subject to many variables – mental fatigue, physical fatigue, caffeine, adrenaline, stress, weather and cardiac drift to name a few. There is also a lag between your hearts response to an increase in effort.

Heart rate is a measure of the hearts response to the work being done by the body. Why not measure the work directly and use that instead of heart rate? Power meters allow us to do just that.

Power meters have been around in cycling for a while but in recent years they have become available for runners.
Initially they weren’t that applicable to trails as power readings when hiking weren’t accurate but the guys at Stryd have done a lot of work recently and have developed the software to the point where it works on all but the most technical of trails (in which case your skill rather than effort is likely to be the limiting factor). Without getting into too many details or getting too technical Styrd use a foot pod that measures 3-dimensional acceleration and then converts that into a power number through their algorithm.

Benefits of using power for trail runners.

Power gives you a reliable metric to base your training intensity on. Whether you are going up or down or flat you base your speed on power.
For example, if you are doing hill reps and the hill has a few changes in gradient. Pace will vary according to gradient so you cant use pace. Heart rate will take a while to adjust to any speed changes, so if the hill flattens off a little and you dont pick the speed up then heart rate will take a while to reflect that. Power on the other hand, will drop straight away allowing you to adjust your speed accordingly to keep the intensity at the desired level.

For long runs – are you running too easy or too hard? Are you pushing too hard uphills and too easy on the downhills? Speed varies too much to be of any value at all but with a power meter, since it is measuring the work done by the body regardless of terrain, can tell you.

This is where heart rate and perceived effort are the most unreliable. The increased levels of adrenaline in the body combined with fresh legs that cant wait to run, convinces us that we are running easy at the start. Not till its too late do we discover the initial pace was indeed too fast and we are now paying the price. With a power meter you can derive a power level to stay under to ensure you are still running well late in the race.

In essence, it can answer the question – what intensity should I run at. Whether you are running up or down or flat or even hiking you can determine a power zone to ensure you maximise your training and race to your potential.

Monitoring Training Loads.
How do you keep track your training volume each week ? For some its distance, others time but neither are that accurate. For example a week containing a hard 40km race plus a few easy 10 k runs is only 60km but would be a lot more taxing on the body than 80-100km of easy runs.
With a power meter we can objectively quantify how much work the body did and use that to base our training increases on instead of time or distance.

Measuring Efficiency.
We can use power to track our improvements in efficiency. For example on a regular training route if we notice that we are running the same time for less power or running faster at the same power then we know we have improved. You might argue the same can be said for heart rate – but given so many other factors affect heart rate we can never be truly sure it was an improvement in our running or our heart rate was lower due to cooler temps, or less stress etc.

With power meter we can also get feedback on fatigue and changes in running form. If our power levels are higher than usual for a typical run but pace is the same then suggests our running is less efficient (ie using more power to run the same pace).

Hopefully, by now you can see some of the advantages of using power – the next question is how do we use it.

Power training zones.
For those that have used heart rate zones before its basically the same approach with power.
We establish a series of zones targetting different areas of fitness- easy aerobic, tempo , intervals VO2 max etc

Power zones are based on your threshold power (loosely defined as the maximum power you can sustain for 30-60 minutes.) There are a number of tests you can do to determine this that don’t require any fancy equipment other than a power meter and software. Once out have threshold power training zones are based on a percentage of that. For example if your threshold power is 250W then easy runs should be 175-200W ( 70-80% of Threshold Power).

Is power the same for flat uphills running and hiking ?
Trail running is very different to road running in that we often have hills that we can run up easily, hills that are steep and are at the edge of our running ability and hills that we hike up. 
What I have found is for some athletes their Threshold Power is the same for all three but for others it’s very different.

Some can sustain higher levels power running uphill than they can on the flat – others the opposite. Hiking power threshold may be very similar or quite different. 

Some might see this as a drawback for using power but it just takes a bit more data and testing to ensure you know your threshold power for different terrain.

Pacing races on trails using power.
We know that all of us slow down in ultra’s 100km or more – what determines a good performance is minimising the slow down. In shorter races maintaining speed is the aim.

Trying to determine whether the runner ran a well-paced race or not was only possible through subjective means. With power we know have some objective data – we can see decreases in power throughout the race.

Lets look at some examples to see how it works

In the first example we see the graph of an elite athletes power and heart rate in a 100km trail race.
The light blue line is heart rate, dark blue is running power and purple is hiking power, representing average per hour throughout the race. You can see the hiking and running power are close in number and that heart rate differs considerably.

Note the initial decline in both over the first half of the race and then stabilises for the last 30km or so , even increasing slightly as he pushed for the finish. This was a good performance for the athlete given his fitness level at the time.


Example 1: A well paced 100km

In example 2 we see the same athlete as above in another 100km trail race. In this race he wasn’t as fit and although his power levels actually started lower and they stabilised sooner (but at a lower level) they continued to drop in the last 30km.
The athlete wasn’t as fit for this race and wasn’t able to sustain an even lower level of power than in the first example.


Example 2: Struggling in last 25km

In example 3 this athlete was also competing in a 100km race, started off too fast and power declined consistently throughout the race. Hiking power is less similar to running power in this athlete. The athlete’s heart rate varies significantly from the power trends and is a good example of why heart rate isn’t a great way to determine effort in a long trail race.

Example 3: Continual decline of power but with heart rate not following same decline as power

Note the 100km races these athletes competed in had approx 4000-5000m of vert with a mix of trails and stairs.

Example 4 is of an athlete in a much shorter race but highlights the large difference between running power an hiking power in this athlete. This may or may not mean he needs to work on his hiking pace, or it may just mean we need to use different power levels when hiking. Analysing his traiing data would tell us that.


Example 4: Difference between hiking and running power

Example 5 is in a 50km race taking 6 hours 20 minutes. You can see the runner was able to stabilise the decrease of power much sooner and finish strongly. After reviewing the data with the athlete we concluded that he could have pushed harder in the middle section for an even better result


Example 5: Strong finish

As you can see from the above examples – we now have a means of determining how hard we should be running in a long trail race. We expect a slow down – how much of a slow down is what governs the race performance, start too hard and we lose any time gained in the first half in the second half. Knowing what kind of levels to start at in order to minimise the slow down can have a large impact our overall race time.

Without power many runners pour over previous years results calculating split times to help them achieve a goal time. The problem, of course, is having split times doesn’t tell us how fast to run at any given moment. It tells us once we get to a checkpoint whether we have covered that last section faster or slower than our pre-race plan. There is nothing you can do about it then. Power allows us to develop a race strategy that we can use at any moment throughout the race and adjust our speed accordingly.

Limitations with power.
One of the main ones is training and racing in the heat – performance and the ability to hold a certain power level will be decreased in hot humid conditions (this limitation also applies to pacing by heart rate or pace).

If you run on soft sand or into strong headwinds then the power numbers won’t reflect that but again the same limitations apply to pace and heart rate. Your heart doesn’t know if you are running on soft sand or into a headwind, it just responds to the work being asked of it. If you use pace then you dont know how fast to run in soft sand unless you have done some training in soft sand – the same applies to power.

Other advantages of power.
The Stryd power meter also offers a large range of metrics including ground contact time, leg spring stiffness, form power and vertical oscillation which we can use to analyse our running in more depth.

The distance measuring accuracy of Stryd is by far the best out of any GPS Watch ( see here for a summary of GPS watch accuracy)

Perceived effort vs Power meters.
Does having a power meter mean we can do away with perceived effort? Definitely not. Power is another metric we can use to help guide our running. It’s the best metric we have at the moment. Relying only on power is like driving a car and relying only on the speedo. If we dont have other feedback systems like revs, engine temperature, petrol gauge etc we are going to get into trouble at some stage. The combination of power and a well-developed sense of perceived effort can give you the best information to base your training and racing on.

Disclaimer – I have no financial connection with Stryd – its just a great product.

NOTE – there is debate whether the power numbers Stryd shows is really a measure of power or something similar but not power and should be called something else; for the purposes of training and racing – all that matters is it is a reliable metric that we can base training on.

This article is a brief overview on the benefits of power training. For more detailed discussion on power training have a look at the Stryd facebook page.

  4 Responses to “Power meters for trail runners”

  1. I’m using Stryd from the beginning and for almost all of my training. For flats it’s good, but so far I haven’t found it very useful for (ultra)trails. Proponents tell us “power” is the most reliable and objective metric of the work we do, but I’m not convinced. Power meters do not take in account the metabolic cost of temperature and altitude (racing in 15ºC or 35ºC, at 100m or 3500m altitude makes a huge difference). The power measured by Stryd is indirectly calculated and not directly measured, which is not a problem on flats, but output on long and steep grades (30-40 degrees) are completely unreliable and useless. On the downhills the measured output is very low, but a power meter does not take into account the metabolic cost of eccentric muscle movement. Ergo: for long and steep trails I prefer HR and perceived effort 🙂 It is good for relatively flat runs, and for controlling interval intensity (but on steep inclines only when the output is high enough).

    • Hi Emmanuel – thanks for your comments. A few thoughts on the points you rasied
      re temperature and elevation – correct it doesnt – neither does pace or heart rate and experience in training will help adjust power levels for differing altitudes and temperatures

      True power cannot be measured on the trials or roads directly this is true – but what I am interested in is – is it a valuable metric we can use in training and in my experience it is

      Re steep grades- there is a new software update out today that gives far greater accuracy for hiking up steep slopes and makes it a very useable metric

      Re downhill – heart rate is a very poor metric to gauge effort on downhills as it doesn’t take into account the eccentric loading. It is just a reflection of the effort in running at that speed. The new software update also gives much better downhill readings that you can base effort level from. I would never use heart rate for downhill running.

      I suggest you try the new update and see if you have a better experience .

      The races I discussed in the blog all had long steep sections at times and power numbers were very reliable.

  2. Hi Great thoughts,
    I have really enjoyed playing with the stryd and agree that whilst people argue about what exactly it measures the main benefit is the reliability of the measurements and therefore about able to use it for pacing and training zones in addition to heart rate which at certain times is quite unreliable (especially with an Optical HR which may drop out)
    I was wondering if Scotty or yourself had any figures to compare (as inaccurate is it is to do so) the pros from the bros.

    • Hi Andre

      Yes optical HR is still in its infancy has its problems.

      Re comparing pros from the bro’s – I dont have a lot of data at the moment – most of the Stryd Athletes I have are more to the front end of the field. I suspect and initial data seems to confirm it that elites tend to maintain a higher level of power for longer and can even increase it towards the end.

      How much of that is due to the extra motivation for elites in racing for positions compared to slower runners being happy to finish we dont know.

 Leave a Reply