A Change of Pace
For almost four years I've been training very seriously as an endurance cyclist. The effects have been amazing: I lost over 30 kg (70 lb), reduced previously sky-high blood pressure, lipid and blood sugar numbers to the low end of the normal range, and transformed the way I feel about myself. All great things.
But now I'm abandoning endurance training in favour of sprint cycling.
Here's why.
First, I'd taken it just about as far as I could. My best aerobic power and VO2max numbers (see below) are either awesome, quite good or mediocre, depending on whether you compare me to the average man, the average weekend club rider, or the average competitive amateur racer.
3 hours: 244 Watts 1 hour: 272 Watts 20 minutes: 307 Watts 5 minutes: 392 Watts VO2max: 61 mL/kg/minute
At this point, I doubt I could add more than a handful of extra Watts to my 5+ minute power numbers, even with a hard focus on longer duration work. Apart from my 5 minute power (more on which below), these numbers haven't really increased much over the last 18 months; it's pretty likely that I'm close to my physiological ceiling.
Actually, it was clear to me quite early on that I'm naturally far stronger anaerobically than aerobically, and this idea has subsequently been borne out by all the data I've collected.
Comparing my data with those from other riders makes this crystal clear. For example, intervals.icu currently has over 5,000 male cyclists in my age bracket. Presumably, a large majority of these cyclists are quite serious and well-trained (since they've each taken the trouble to invest in a power meter and sign up for a service to analyze their data). My power percentiles in this group are:
5 sec: 96th 15 sec: 97th 30 sec: 98th 60 sec: 99th 3 min: 96th 5 min: 93rd 20 min: 82nd 60 min: 59th
Quite a drop-off! And this is after almost four years of strongly endurance-focused training.
Look again at my best 5-minute power, and compare it with my best over 20 minutes. Most riders who could do only 307 W for 20 minutes would be somewhere around 350 W for 5 minutes, not 392. This relatively high 5-minute power is due to my huge anaerobic capacity (estimated by three different training tools at close to 40 kJ — itself in the top fraction of the 99th percentile), and so a five-minute effort tends to greatly overestimate my aerobic capacity (VO2max), which is actually around 350 W. But clearly this 5-minute:20 minute ratio is another big clue that I'm an anaerobic kind of guy.1
Unfortunately, anaerobic capacity on its own doesn't really do much; if you also have a large aerobic capacity you could be a good criterium racer, while a large anaerobic capacity + high thresholds would make you a good road racer. But I don't have those. However, a huge anaerobic capacity plus large anaerobic power (i.e., peak power) would make me a good sprinter, especially over longer efforts of 30-60 seconds. And that's what I'm now working towards: keep topping up my capacity while aiming to increase my peak power via weight training in the gym and sprint work on the bike, and then see what kind of performances I can do over 1 kilometre.
Interestingly, it is commonly reported in the scientific literature that the anaerobic:aerobic crossover point (the duration at which there is a 50% contribution from both (shorter-term) anaerobic and (longer-term) aerobic energy sources) is around 60 seconds. But it has recently been argued that, due to a decline in metabolic efficiency over durations of up to several minutes, anaerobic energy contribution to total energy expenditure for maximal efforts is underestimated by about 30% in studies that do not account for this (i.e., just about all of them to date; see Haugen et al., 2021).
This means that the crossover point could in fact be around 3-4 minutes, making events like the 400 metre run and 1 kilometre cycling time trial overwhelmingly anaerobic, rather than somewhere close to a 50/50 split. This makes a lot more sense to me intuitively, as well as in explaining my own numbers and when looking at the training (and physiques) of athletes like Michael Johnson and Chris Hoy.
Furthermore, for sprint exercise, it appears that the musculoskeletal intensity level that can be achieved determines energy release and sprint performance. This is in contrast to endurance exercise, where sustainable aerobic energy availability determines performance (Bundle & Weyand, 2012). In other words, your sprint power is not limited by metabolic factors; it's determined by the force you can generate through the pedals.
The relevance of this is that if I want to maximize my 1-minute power (and I do), then I need to focus far more on anaerobic training than aerobic training. This is very important, due to the other reason I switched to a sprint plan: arrhythmia.
My first bout of tachycardia happened when I'd been cycling for just under a year. I was making my first attempt at the hardest ride around (the fast ride in Savannah). I managed to hold onto the back of the group for about 10 minutes, but just as I was getting dropped and easing off I got several consecutive heart rate spikes (my HR instantly increased by about 40-50 bpm, well above maximum) of up to 20 seconds each. The only symptom was a fluttering feeling in my chest. A similar thing happened a few weeks later when I sprinted up the first part of a climb on the Asheville Gran Fondo.
This 30-second ride excerpt shows tachycardia in action.
At first these events were few and far between (sometimes months apart), and only ever occurred when I went very deep for several minutes and during an extended period of increased training load. So I didn't worry too much about it. But this spring (again, several weeks into a training load build) they started happening regularly. As in earlier cases, hard riding was still needed to trigger them, but the level of exertion at which this happened was considerably lower than before: they'd started showing up on many of my regular intervals sessions.
I did a lot of research on this phenomenon, and it's actually quite common amongst older endurance athletes. It's apparently most strongly correlated with total lifetime endurance hours (I'd already racked up 2,500 hours in the first 3 ½ years!), and to a lesser extent with time accrued in the heavy-severe intensity domains (i.e., doing long aerobic efforts between threshold and VO2max), due to damage caused by increased pressure and volume of blood flow.
Arrhythmias in athletes are related to the cardiovascular changes that universally occur in response to endurance training (so-called Athlete's Heart): heart and blood vessel remodelling, low resting heart rate, heart murmors, etc. These changes are usually benign, but scarring of the heart chambers can also occur, which in some individuals can interfere with the electrical signalling that controls the heart rate (see O'Keefe et al., 2012).
So it usually shows up in people after a couple of decades of endurance activity, but in my case I imagine various 'lifestyle factors' may have begun the scarring process earlier, long before I started serious cycling!
In any case, I obviously don't want this to progress; at this point none of my individual tachycardia episodes have lasted longer than a few seconds, and have always resolved themselves on their own. However, if ignored, the disorder can progress to a point where the arrhythmia lasts for hours or days, necessitating a visit to the emergency room to have your heart restarted. No thanks.
Given all the above, it wasn't a difficult decision to switch from endurance training to sprint training: to resolve (or at least avoid any progression of) my symptoms, I have to let myself detrain enough so that my heart can repair the damage. The most important thing for me to do is reduce training volume, and also lay off regular high-intensity aerobic riding.
This whole saga has the benefit of giving me an excuse to spend many enjoyable hours researching the best way to go about this (I'd got bored of revisiting my old endurance training plan)2, and I now also get to train how I really like to train. No more tempo or threshold or VO2max intervals; no more interminable endurance sessions. Just short base and recovery rides (often commutes), various sprint workouts, and a few anaerobic sessions (typically widely-spaced 30-second efforts), combined with plenty of time lifting weights in my new garage gym.
In other words, just the fun stuff: flat out, or dead slow. It's great.
The one downside is that group rides aren't part of the equation, as they fall squarely into the 'endurance' category, and don't really help me towards my new goals (which revolve around things like 1-rep maxes in the squat and deadlift, and increasing my 5-, 30- and 60-second maximum power on the bike). But I'll try to find a way to sneak in the occasional social ride every now and then.
In the meantime, hopefully my new approach to training will result in a resolution of my symptoms, so I can keep on riding.
UPDATE: 12 June
Soon after writing this article, it came to my attention that there have been increased reports of myocarditis/pericarditis (inflammation of the heart, one of the symptoms of which is arrhythmia) in individuals taking two does of mRNA COVID-19 vaccinations.
I had my second dose on 30th March, and the increased incidences of tachycardia began a couple of weeks later. It's quite likely only a coincidence, but there is still a possibility that some slight inflammation occurred which had an additive effect with whatever (presumably fibrosis) has caused my tachycardia previously.
Nevertheless, this doesn't really change anything: I'm happy that sprint training is the right way for me to go and I'm really enjoying developing and implementing my new plan.
1 Even the 20 minute:60 minute ratio is much bigger than normal (13% difference). This demonstrates why you can't just use predefined formulas (like 20-minute power – 5%) to set training zones: if you want to know your real 60-minute power you have to ride as hard as you can for 60 minutes! Sorry.
2 My new sprint training plan has gone live, here.