Junk miles: what they are and when they aren't

"Junk miles" is one of those phrases everyone uses and nobody defines. Ask five riders what it means and you get five answers, most of them contradictory: easy riding that does nothing, hard riding that wrecks you, any ride without a plan, any ride without intervals. This post tries to pin it down. The short version is that the pace of a ride does not make it junk; the cost of it does. A slow ride that leaves you fresh for tomorrow's intervals is not junk. A medium-hard ride that leaves you too tired to do tomorrow's intervals properly, and was not hard enough to be worth that, is.

Two things people mean

The phrase is doing double duty, and the two meanings point in opposite directions.

• The first meaning is "too easy to count". Riding to the café, spinning for an hour at a heart rate that never leaves the bottom zone, the Sunday ride where nobody sweats. The accusation is that this riding is so gentle it produces no adaptation, and the time would be better spent on the sofa.
• The second meaning is "too hard to be easy, too easy to be hard". The group ride that is a rolling threshold effort with sprints, the solo two hours where you chase every rise, the commute that is always a bit of a race. The accusation is the opposite: this riding costs a lot and returns little, because it is fatiguing without being a stimulus.

The evidence is fairly clear that the first accusation is mostly wrong and the second is mostly right, with important exceptions to both.

Easy riding is not junk

Seiler and Kjerland's study of elite Norwegian cross-country skiers is the origin of the "80/20" idea. They logged every session by heart rate and blood lactate and found roughly three-quarters of sessions were below the first lactate threshold, only a small fraction in the middle, and the rest properly hard. Seiler's 2010 review pulled together similar patterns from rowers, runners and cyclists. These athletes do more easy riding than anyone, and they are not doing it by accident.

The clearest test of whether the easy time matters is Esteve-Lanao's study of sub-elite runners. Two groups trained with the same total load for five months, one with more time in the easy zone and one with more in the middle, and the group with more easy time improved more over 10 km. Easy riding does something, in other words, and van Erp's four-year analysis of professional road cyclists shows that the people with the most to lose from wasted hours still spend most of theirs well below threshold.

What easy riding does is worth spelling out, because "it builds base" is not an explanation. San-Millán and Brooks measured fat and carbohydrate oxidation and lactate across a range of intensities in professional cyclists and in less fit people, and found that the professionals burn far more fat at a given power and clear lactate far better, differences they attribute to the mitochondrial adaptations that long, moderate riding produces. That is the adaptation the easy ride is buying: a bigger, better fat-burning engine that makes every other intensity cheaper. It is slow to build and it does not show up on a twenty-minute test, which is why people undervalue it.

When easy riding really is junk

There are conditions under which the first accusation holds.

• When it displaces a hard session. Easy riding is the foundation, not the building. A week of six easy rides and no intervals is a plateau, and the evidence that easy time helps comes from studies where hard sessions were present as well. Do long slow rides still work covers the trade-off.
• When it is too short to matter. The intensity-and-duration study everyone cites, Dudley's, was in rats, and it is worth saying so. It found that mitochondrial adaptation depends on both intensity and duration, that at low intensities a session has to be long to produce much, and that returns flatten beyond a certain daily dose. The human version of the same idea is common sense: forty easy minutes is recovery, not training. Three easy hours is training.
• When the rider has almost no time. At four hours a week, an easy ride is an expensive way to spend a quarter of the budget. Our reasoning on volume is in how much should you actually train.

The middle is where the junk lives

Now the second accusation. Seiler's group measured how quickly heart-rate variability returned to baseline after sessions of different intensities and durations in trained athletes. Below the first threshold, recovery was fast regardless of duration, even for two-hour sessions. Above the second threshold, it was slow. The middle zone, between the two thresholds, cost noticeably more than easy riding without delivering the stimulus of a hard session. That is the physiological description of the rolling group ride: it uses up recovery capacity and gives back less than an interval session would have.

This is not the same as saying the middle zone is useless. Tempo and sweet-spot work have their place, and a lot of good plans use them deliberately. The problem is the accidental version: riding there because the group went there, or because it feels like proper training, or because a steady ride drifted up without anyone noticing. The rider ends up with a week that is all medium, no easy and no hard, which Seiler's review describes as the least effective distribution in almost every dataset he looked at. Whether the zone itself is "grey" is a separate argument we take up in is the grey zone real.

A test you can apply to any ride

Forget the pace. Ask three questions afterwards.

1. Did it have a job? Recovery, endurance, threshold, VO₂max, skills, a long day to build durability, or just enjoyment. All of those are legitimate jobs. "I rode because it was Tuesday" is not a job, and rides without one tend to drift into the middle.
2. Did it do the job? A recovery ride that had three hard hills in it did not recover you. An interval session where you sat at 90% of target for the last two reps did not deliver the stimulus. A long ride you cut short at two hours was a medium ride.
3. What did it cost tomorrow? This is the one that catches most junk. If the next hard session was worse because of this ride, and this ride was not itself a hard session, it was junk, regardless of how good it felt.

The simplest tool for the third question is Foster's session RPE: rate the whole ride from 0 to 10 once it is over, multiply by the minutes, and you have a load number that reflects what the ride took out of you rather than what the power meter says it should have. Rides that score like a hard session but were meant to be easy are the ones to look at. We cover the method in RPE and effort perception.

Things that look like junk and are not

• The coffee ride. If it is genuinely easy, it is a recovery ride that you enjoyed, which makes it better than a recovery ride you did not do.
• The commute. Twenty minutes each way at a steady pace, five days a week, is more than three hours of aerobic riding that costs almost no recovery and no free time. See commuting by bike as training.
• The ride with no numbers. A ride without a power meter is not a ride without a purpose. The professionals in the Norwegian studies were logging heart rate and feel long before power meters were common.
• The long day that was slower than planned. Duration is the stimulus on a long ride; if you finished it, it did its job.

Things that look like training and are junk

• The group ride that is a race, every week. One is a hard session. Two is two hard sessions. Three, on top of your intervals, is a rider who is always slightly tired and never improving.
• Easy rides that keep drifting up. If your "easy" ride averages the top of zone 2 and includes ten minutes above threshold, it is a tempo ride, and you now have no easy days.
• Intervals ridden at 90%. Not junk in the sense of doing nothing, but junk in the sense of costing a hard day and returning a medium one. If you cannot hit the numbers, you are probably not recovered, which points back at the rides earlier in the week.

How Moveee handles it

A fixed plan built from the wizard gives every session a job, which is most of the battle, and spaces the hard days so that easy really means easy. The calendar matches each imported ride to the day's planned session with a planned-versus-actual verdict, which is a direct answer to question two: the easy ride that came in at tempo shows up as exactly that. With the adaptive plan, a ride that comes in well over what was planned eases the next hard day, which is the engine answering question three for you. It cannot tell a good group ride from a bad one. It can tell you that Wednesday's intervals were worse than they should have been.

The short version

Easy riding is not junk, and the fittest riders in the world do most of their hours there. Unplanned medium riding usually is, because it costs recovery and returns little. Judge a ride by whether it had a job, whether it did it, and what it cost tomorrow, and the label takes care of itself.