Building My Fellrunning Season: Part 1
Foundational performance testing to set up the coming macrocycle
In the last three years I've learned a lot about running. Early this year I took an extended break to spend some time at home in Australia. I used a lot of that time to read, reread, listen and re-listen to an unhealthily-obsessed amount of athletic training and performance content.
As much to share what I’ve learned as to organise my own thoughts, I’m putting together a little series that outlines how I’m tackling this year’s fellrunning season.
Part 1 - Testing
If you read all that follows, I encourage you not to take my word for it. I have included a list of recommended sources that I have, and to continue to use, that are well worth the time to digest.
The content I have found to be most applicable comes from people with PhD’s, or at least practitioners with a distinctly scientific approach. There is plenty to learn from a variety of athletes and coaches, but much comes from N=1 experience rather than fundamentals of physiology.
On the topic of N=1 however, this is perhaps the most important element; the need for individualisation was stressed by pretty much all sources.
Everyone responds to training in a multidimensionally different way. There is no point copying anyone else’s program, you have to figure out what you respond to, shape that toward the demands of your goal event, and do that training - not what you saw an elite runner doing on YouTube.
So, how to know what you respond to and adjust training according to race goals?
Testing
I made big progress in 2023, culminating in a 50 minute PB at Trigger Fell Race in early January 2024. The trouble is that this isn’t that precisely measurable. How much of that came from fitness vs better weather? What fitness improvements actually occurred?
I can’t wait another full year to do Trigger again to see if my training has been effective. In order to prescribe training intensities, and diagnose what I respond well to (or not), I need some simple, and regularly repeatable testing.
While lactate profiling has been getting some hype in recent times, it's neither practical nor regularly repeatable for me. It would be super insightful, but the best tool you can't use isn't the best tool.
Maximal Testing
Both Olav Alexsandr Bu (coach of Kristian Blummenfelt, Gustav Iden) and Dr. Philip Skiba (Breaking 2 Project) agree that time/distance trials in the field are probably the best and most practical testing an endurance athlete can do.
Why? Because these tests capture actual performance, rather than testing the underlying physiological markers that indicate ability to perform. At the end of the day, the objective is to get faster, not lower your heart rate or blood lactate concentration1.
Furthermore, they agree the tests should be done across a range of durations to create a velocity curve. Both emphasise that this curve is one of the most powerful tools an athlete has to view their current fitness, prescribe training, and measure response to training.
The recommendation is a minimum of two tests; one of 3-5 minutes2 and another of 15-20 minutes. Both need to be maximal efforts. Olav likes a further test of ~60 minutes, however for me this seems too taxing outside of a race, so I will stick with the two shorter tests for the time being.
What it will create is the below. In this example, it allows an easy comparison of changes in performance between two rounds of testing. The dark line from the second test here shows maximal speed at all durations has increased, a great indicator that whatever training occurred between the two tests was effective.
Depending on where I see changes in this curve, I can measure the effectiveness of different training blocks. If I’ve just done a block of VO2MAX work, I can look for a change in my 5 minute performance compared to other durations. If I see very little movement, I know that something needs to change.
In practice, because I’m a fellrunner trying to improve primarily uphill speed I am doing these tests on a treadmill set to a 15% gradient, where I set a fixed speed and run to exhaustion. There is a bit of guesswork required in setting the speed, but as long as I go to exhaustion within the aforementioned timeframes it doesn’t matter too much.
Testing frequency will be roughly once a month. I don’t want to disrupt too much training with testing of this nature. Additionally, the scientific literature suggests it takes 4-8 weeks to adapt to a training stimulus, so testing every 4-6 weeks makes sense.
Critical Speed Modelling
A further reason for the maximal tests above is to calculate Critical Speed. Dr. Skiba is much more emphatic on this topic than some other sources, but all agree that some measure of this threshold is important. Important for both measuring fitness, and prescribing paces to target improvement at this intensity.
The threshold that Critical Speed represents is measured in a variety of different ways and has numerous names that for some bizarre reason the exercise physiology world can’t agree on. For clarity, it is the threshold that is often called simply ‘threshold’, or anaerobic threshold, or maximal lactate steady state, or second ventilatory threshold.
In short, Critical Speed (CS) is the border between stable, and unstable. At or just below CS you can perform for 30-40 minutes, heart rate and blood lactate concentration will peak and stabilise. Even just slightly above CS and heart rate will rise continually, you will eventually reach VO2MAX and exhaustion much faster. For a reasonably well-trained runner, CS is probably close to 10k race pace.
My actual CS itself doesn’t matter, I won’t race much at that intensity. All I care about is making it faster, to create room underneath it for my goal race pace to be faster. Positive change in CS is associated with improved performance at intensities both above and below it; it’s hard to make your CS faster without getting faster across the entire curve.
In order to calculate it from a two parameter test, I follow the below steps (taken directly from Dr. Skiba’s book, Scientific Training for Endurance Athletes).
Subtract short test duration from long test duration (in sec).
Subtract distance covered in the short test from the long test (in metres).
Divide the resulting distance by the resulting duration to get CS in metres/second.
With that calculation done, I have a third marker in addition to the raw test data from my 3-5 and 15-20 minute tests to monitor fitness.
It would be remiss of me to not point out that it’s often recommended to do three tests to calculate CS, adding a third middle duration test of 10-12 minutes and keeping the longer test more toward the 20min mark. As it’s more practical to fit into regular training, I’m going to work with the two parameter model for the time being, which is the minimum required.
If all the testing outlined so far works as intended, I will have a consistent view of my fitness in terms of real performance output across a range of intensities, so I can periodise training based on my physiology and how it responds (or doesn’t) over time.
Submaximal testing
I’m going to knowingly contradict myself slightly now. Above I highlight the importance of testing performance, and not indicators of performance like heart rate. My purpose for submaximal testing is twofold; get a view of fitness at lower intensity and to spot potential overreaching. I will do this with a simple fixed speed + duration test to monitor changes in heart rate.
The submax tests will be done once or twice weekly as part of a warm up for an interval session. The protocol is fairly simple.
Warm up for 10min, 0% gradient, starting at 8kph increasing by 1kph every 2min
1min before test start, drop speed to 4kph and set gradient to 15%
5 minutes at 7kph on 15% gradient
Record average heart rate from the final minute of the test
All going well, with the speed of the test not changing, test heart rate should decrease over time with improved fitness. Because the intensity of the test is low and functions as a warm up, I can repeat this 1-2x weekly without compromising training.
While I can’t plot this on my velocity curve, it gives a bit of insight into my efficiency that the maximal testing might not capture. The frequency that the low intensity enables will also be much more effective in catching overload problems early when I increase intensity/volume.
A note on the prolific use of treadmills that I’ve described throughout. I’m giving up some specificity in favour of controlling the test environment. The nature of field testing for fellrunning would often mean testing literally in a field, subject to wind, rain, mud, sheep and so on. In order for the tests to be effective, they need to comparable, so as far as I can see the treadmill is the only practical way forward.
Thanks for reading how I will be using testing as the foundation of my 2024 fellrunning season. In the next part, I’ll cover how this will inform my training intensity, in both volume and speed terms.
As promised, listed below are some of my sources.
Books
Scientific Training for Endurance Athletes - Dr. Philip Skiba
Training for the Uphill Athlete - Scott Johnston, Steve House, Killian Jornet
Advanced Marathoning - Pete Pfitzinger
Uphill Techniques for Off-Road Runners - Keven Shevels
Substack Publications
Endurance Essentials - Gordo Byrn
The Science of Maximal Athletic Development - Alan Couzens
Podcasts
The Norwegian Method (all episodes), Olav Alexsandr Bu
That Triathlon Show, #425, James Spragg, PhD.
That Triathlon Show, #311 , Dr. Philip Skiba
That Triathlon Show, #173 , Dr. Philip Skiba
Fitter Radio, #545, Dr. Philip Skiba
Fitter Radio, #412, Dr. Philip Skiba
Upside Strength, #212 , Dr. Philip Skiba
Koopcast, #128 , Dr. Philip Skiba
There is an argument that you could get the necessary insight from monitoring performance during interval workouts, ie. seeing a lower heart rate for a given speed + duration would indicate improvement. The problem with this is that heart rate varies day-to-day dependent on many factors, and diagnosing training response based on 5 - 10 beat changes becomes guesswork unless the change is truly substantial and sustained over weeks.
Speed achieved in a 3-5min maximal effort is apparently fairly well correlated with max speed achieved at the end of a VO2MAX ramp test, which is commonly referred to as velocity at VO2MAX (vVO2MAX), so this test, to an extent, can replace formal testing purely to discover that metric.