Trail Running vs Road Running: What’s the Difference?

Trail Running vs Road Running: What’s the Difference?

Forget everything you think you know about trail running. Seriously.


It’s a funny thing how we humans come to acquire knowledge about the things we observe around us. The even funnier thing is that we often don’t question why we believe certain things to be true. Here’s one for you, “what variables affect trail running performance?” If you’ve been paying attention up until now – you’ll remember I wrote a post back in March about the Three Key Aspects of Running Performance: VO2MAX, lactate threshold, and running economy…right? Wrong! Hell if Tim Noakes can switch from hi-carb to bye-carb… I can change too! The idea that these three variables are the key indicators of running performance is known as the ‘classic endurance model’.


You see recently I’ve been really thinking about my PhD that I’m about to embark on and started to do some serious research on the matter. I’ve always felt that there’s a difference between trail and road running, but I’ve never questioned what affects race performance in trail vs road running. Back to my statement about how we acquire knowledge for a second. Why do I know the three variables above to be the key factors of running performance? Because someone told me so in a sport science class about running, and because for almost 100 years now, this has been known to be true! So here’s the problem – bias. Nobody ever intended to be biased when they told me these things, and no-one is trying to be biased when they tell you that you need to have a high VO2MAX, lactate threshold and good running economy to be an elite runner – they are simply being biased because of the lens they are looking through, here’s why: trail runners do not fit the mold of the classic endurance model.


Let’s flashback 100 years. Someone observes a fellow human being running incredibly fast for an incredibly long time (okay there was no Eliud Kipchoge back then but the world record for the men’s marathon in the 1920s was around 2h30min) and decided – hey, let’s try and figure out what makes this guy so good at running! So this person (an exercise physiologist) or a group of people (exercise physiologists) come up with a test (VO2MAX test) which can test the maximal amount of oxygen this runner can utilize to produce energy during high intensity running. They need someone to compare this runner against, so they pop down to the local promenade and recruit a couple of recreational runners, and boom! The VO2MAX test is born. Everything sounds good so far right? The elite runner has a high VO2MAX and the recreational runners have low VO2MAXes, so story solved? Well…not quite. The researchers go on to figure out that when they compare the elite runner to another elite runner – their VO2MAXes are the same! So what makes the one better than the other? And so the process begins again. They establish this time that lactate threshold is the difference between the two runners’ performances (repeat this process for running economy). This process has taken us about 100 years to understand and develop (just by the way), and so it’s easy to see how this information has come to be both commonly understood and accepted.


So here’s the crux of the message I want to share with you. Knowledge is developed and distributed through the lens of both the time/era that the knowledge is developed in and within the boundaries of the questions asked in that time/era. Every time someone (myself included) tells you as a trail runner that the classic endurance model are the key indicators of running performance – they are perpetuating this bias. The bias is that the questions asked over the last 100 years have sought to answer, “what distinguishes running performance amongst runners?” but, what they’ve actually been researching is, what distinguishes running performance amongst roadrunners?”. You see, I venture a guesstimate that of all the 1000s of pieces of research that have been published about running performance since the early 20th century, 95% of them have been on roadrunners and the vast majority have been done on a treadmill.


There are two types of bias present here namely; sampling bias and design bias. The sampling bias is that trail runners have been systematically excluded from research studies (no – not on purpose, but research in trail running is only now picking up), and the design bias is that most of these studies have been done on treadmills (replicating road over ground conditions, not trail over ground conditions) or on road or tartan track. The real question we should be asking is, “to what extent can we apply what we know about the classic endurance model to a trail running population?”. The reality is – not very much! If you’re at all interested in the argument I’ve presented above, please feel free to get in touch with me. I’m going to be discussing some of the key literature I’ve read recently that has helped shaped my perceptions about the difference between trail and road running in a series of quick-fire posts about trail running performance. I’d love to have your thoughts on the matters above, and what that means for us as trail runners.

No Comments

Post a Comment