Cross-training experiment: expectations, data and learnings

6 weeks on the bike

Another day on the bike in Amsterdam.

In the past six weeks, I’ve mostly trained on the bike, while recovering from a running injury. It has been an interesting experiment, and below I cover goals and expectations, changes in resting and exercise physiology, and some learnings.

Goals and expectations

I covered in a previous blog the research on cross-training and de-training (here). Based on previous studies, it seems we can maintain VO2max by e.g. cycling instead of running, but we would likely lose running economy.

This of course makes sense, as some aspects of training are easily transferrable (the way we use our heart is not so different), while others are quite specific to the sport (e.g. how we use our muscles).

Unsurprisingly, I wasn’t as optimistic about “maintaining VO2max” either, given my previous experiences a few years ago with other injuries, but I was ready to give it a try (nothing to lose after all).

I also thought that given all the running I did recently (up to 200 km / week), I would lose some speed and aerobic efficiency, but maintain endurance, which is not really what happened.

Let’s see how it went.

Execution

My goals for this training block were to 1) cycle as much as possible 2) run as frequently as possible, 3) walk as much as possible.

Running wasn’t really an option at the beginning, but given how much you can reduce de-training just by maintaining a little running in your week (see this), I tried frequently to run e.g. 5 minutes, then 10 minutes, then 30 minutes, etc. and stopped as soon as I felt any sharp pain in my calf.

Going all in

In terms of cycling, I suspect in the past I didn’t obtain good results (i.e. didn’t maintain any fitness) because I wasn’t cycling enough. So this time I went all in and cycled 19 hours the first week, 22 hours the second week, and 21 hours the third week (starting from zero).

Here is my running and cycling volume for the past 3 months:

Transitioning from 60 hours / month of running, to a similar amount of cycling.

For the first 3-4 weeks I kept going only because I was highly motivated, but I enjoyed almost none of it. Apart from the far-from-ideal weather and daily rides in storms, wind, and rain, I also went from being “happy with where I got with running” to being a beginner at a new sport, which was a bit depressing.

Somewhat interestingly, I felt like I was never challenged aerobically, but my legs just couldn’t do it. I went out for many rides at 100-110 bpm or similar (my maximal heart rate is ~182-187 bpm), feeling relaxed from a cardiorespiratory point of view, but somewhat unable to generate power in my legs for sustained periods of time.

After about ~60 hours or 4-5 weeks on the bike, I started feeling differently and actually enjoying more riding, as well as being able to push a bit more with my legs. Just your normal training adaptations probably, but I would guess that these issues are amplified for someone who has a weak spot in neuromuscular fatigue / endurance, as I covered elsewhere in the context of running and marathon performance (here).

Two easy rides at similar perceived efforts / heart rates, about 5 weeks apart.

Re-introducing running volume and intensity

I tried to run frequently to limit detraining, but also tried to run harder a bit too soon (also with the idea that it would help to limit de-training even more), and this caused some preventable setbacks.

After re-injuring myself with a hard session, I went back to building volume before intensity, and spent plenty of time on my legs either running or hiking for 2 weeks:

Full log, here.

In the last two weeks I had no pain and managed to run a bit more (100-120 km / week), even though much of this time I had to force myself to stay out on my legs while terribly sore.

While I am used to this feeling post-workouts, in the past weeks every run has felt like this. Given how much I ran in the past two years and especially in the past six months, it was somewhat unexpected to me to struggle this much from a muscular point of view, despite being aware of my limits.

Let’s see what parts of my fitness I could maintain and which ones I couldn’t, in the sections below.

Physiological changes

Below I cover changes in resting and exercise physiology during the past 6 weeks.

Resting physiology

In terms of resting heart rate variability (HRV), I was very stable during this entire time, which was good to see (stability is the ideal response, as I discuss here). Despite a change in the type of training stimulus, I was still training a lot as I was doing before while managing quite well non-training related stressors.

Baseline HRV always within normal range.

In terms of resting heart rate, I did have an increase when visiting friends at altitude, and then not feeling great for two days afterward, before re-normalizing:

Below we can see for context the same timeframe for training load, in terms of running:

The peak in running load on August 13th here is in fact a very long hike in the mountains, with limited running.

and cycling, where you see a very large increase, and a dip then for the week I spent driving to Chamonix and then driving back to Amsterdam:

New habits.

Exercise physiology

In terms of exercise physiology, in the past, I had seen very large changes in heart rate at a given pace when getting back into running after an injury, regardless of cycling. As mentioned above, the reason might have been that I simply did not cycle enough, which is something I certainly did this year.

Did it work then?

Much to my surprise, it worked really well.

Granted that I did keep running a bit here and there, but not much. In particular, when I started running, my heart rate at a given pace (what we call aerobic endurance in HRV4Training) had not changed at all. Similarly, after some intensity, it would re-normalize really quickly, clearly showing that my fitness was still there.

We can analyze this a bit more systematically by looking at the aerobic endurance feature in HRV4Training Pro. Here are my settings for this analysis:

The goal of this configuration is to capture easy aerobic runs and see how the change over time.

and here is the data, which basically looks at how heart rate changes at a given pace for filtered workouts, over time. If anything, it seems aerobic endurance has even improved a little, or in other words, my heart rate is similar or slightly lower at a given pace, with respect to before getting injured:

You can learn more about aerobic efficiency, here.

This was nice to see while running, as I was expecting to get a bit worse.

Today (August 24th, 2023), after almost 6 weeks since the injury, I did the first proper workout in a while (3 x 2000m), and even during this session, I was running similarly to before the injury. For example, I had run a similar session in July at 3’50”/km with the third interval at 170 bpm (activity here):

While I ran 3’46”/km today, also with the last interval at 170 bpm (activity here):

While the session in July was a bit of a bad day, these sessions are very similar and the pace I could hold today is pretty good for my standards.

Based on the data above, we could say that indeed cycling a fair amount over the past month allowed me to maintain my VO2max and aerobic efficiency.

So far so good. Let’s now look at what I wasn’t able to maintain.

Limiters

What I could not maintain during this period of cross-training is my ability to endure sustained efforts, from a muscular point of view. I feel really fatigued and sore in my legs all the time, and I tire very quickly with respect to before.

Here is my subjective muscle soreness, which goes quite high as soon as I start running more:

I log muscle soreness subjectively every morning in HRV4Training.

In today’s 3 x 2000m session, I was completely cooked at the end. Normally I would do 4 or 5 reps in this session, but I had to stop today. Last Sunday I also went out at marathon intensity and had to call it a day after only 12 km at intensity (and 17 km total). Normally I would do twice that (35 km with 25 km at intensity).

It was interesting to see how some aspects of fitness were almost perfectly maintained - even when quite different aspects such as speed near threshold and heart rate and low intensities - which are as good as before, while other aspects, such as durability, have decreased quite dramatically.

As mentioned above, this might be amplified by the fact that economy and durability are limiters for me even when I run a lot, and others might experience it differently or less.

Learnings and next steps

Overall, I enjoyed this experiment, it was a good excuse to learn from the scientific literature on de-training and cross-training, and I always enjoy a bit of self-experimentation.

I also believe I am slowly picking up a bit of passion for another sport, which is a good thing, and I am looking forward to doing more cycling in the future. This was also a good reminder that it is always hard to start something new when we are so bad at it and the conditions might not be ideal. Giving it enough time our relationship with it might change dramatically.

The long day.

As running remains my priority, and my limiters (i.e. economy and durability) are heavily impacted by how much I run (as shown once again by the past 6 weeks), my goal is to keep running as much as I can.

If I had a different physiology, maybe lacking in VO2max and good in running economy, I would probably consider quite a different plan, with more cycling and more high-intensity cycling (ideally indoor for safety reasons) to push VO2max higher and reduce injury risk.

However, in my case, I have to run as much as I can to keep improving in terms of economy and durability, while VO2max is quite high (for my level). I do have of course to balance this with the difficulty of staying injury-free, which we all know, is not as easy for runners.

So here is my plan for the future:

• Go back to one day per week off running, and use this day to go for a medium / long ride. Having an alternative works well for someone who tends to behave obsessively (see later).

• Add one indoor high-intensity session on the bike / zwift, whenever possible. This type of training should help in terms of fitness, takes little time and has minimal or no impact on running in my experience. I will start by doing it the same day I do a hard running workout, so that I do not compromise running.

• Add another ride on a day in which I don’t run much (e.g. 60 minutes run, then 2-3 hours ride). This would be the only activity that aims at maintaining cycling fitness more than helping running.

The goal with this plan would be to 1) reduce injury risk by not running every day, which had worked well before I abandoned it in favor of more volume 2) maintain some level of cycling fitness all year round, so that I can enjoy a good adventure on the bike here and there. I have no intention to race my bike (I’d be terrified), but I always love a good day out, and finally I need to 3) manage my personality: as I tend to overdo things, getting some of that behavior away from running, and into another activity I can do at certain times (e.g. cycling after a hard race, instead of running even more, which was what caused this injury), is probably a better way to manage my health and long term performance. I suspect I might not be alone here.

Alright, I hope there is something useful for you as well in there!

Keep training.

Marco holds a PhD cum laude in applied machine learning, a M.Sc. cum laude in computer science engineering, and a M.Sc. cum laude in human movement sciences and high-performance coaching.

He has published more than 50 papers and patents at the intersection between physiology, health, technology, and human performance.

He is co-founder of HRV4Training, advisor at Oura, guest lecturer at VU Amsterdam, and editor for IEEE Pervasive Computing Magazine. He loves running.

Social:

• Twitter: @altini_marco.

• Personal Substack.

• Strava.

New to HRV4Training?