Every second post on cycling social media mentions 80/20 or polarised training. It has become the go-to answer for everything. Not improving? You need more Zone 2. Going too hard? You need 80/20. Want to get faster? Polarised training. The problem is that most people throwing the term around have never actually looked at what Stephen Seiler found, how he measured it, or who he was studying. And the way it gets applied by most amateur cyclists has almost nothing to do with the original research.
I coach around 50 athletes and I have this conversation constantly. Someone comes to me already "doing 80/20" and they are confused about why they are not getting faster. Nine times out of ten, when I dig into their training, they have taken a concept designed to describe elite athletes training 20 plus hours a week and tried to squeeze it into 8 hours of riding while skipping the bits that actually make it work.
WHAT SEILER ACTUALLY FOUND
Stephen Seiler is a sports scientist based in Norway. He spent years studying how elite endurance athletes across multiple sports actually trained. Not how they should train in theory. How they actually trained in practice. What he found was a pattern. Across cross-country skiing, rowing, cycling, and distance running, the best athletes in the world kept landing on a similar distribution. Roughly 80% of their training sessions were low intensity, below the first lactate threshold. About 15 to 20% were high intensity, above the second lactate threshold. And very little, maybe 5 to 10%, sat in the middle.
That middle zone is what most of us would call tempo or sweet spot. The zone that feels productive. The zone that hurts a bit but not too much. The zone where most amateur cyclists spend almost all of their time.
SESSIONS, NOT TIME
This is the single biggest thing people get wrong. The 80/20 split refers to sessions, not time spent in each zone. That is a massive distinction and it changes everything about how you should think about this.
The athletes Seiler studied were doing 10 to 13 sessions per week. If 80% of those are easy, that is 8 to 10 easy sessions. If 20% are hard, that is 2 to 3 hard sessions. When you look at those same athletes and measure actual time spent in each zone rather than counting sessions, it works out closer to 90/10. Seiler himself has said this. The reason is obvious. An easy session might be 2 to 3 hours. A hard interval session might be 60 to 90 minutes. The easy sessions take up way more clock time even though they are "only" 80% of the session count.
BY SESSIONS
BY TIME IN ZONE
So when someone training 8 hours a week looks at their TrainingPeaks and says "I need 20% of my time at high intensity, that is about 96 minutes of hard work per week," they have already started from the wrong number. If anything, using the time-based figure, you are looking at maybe 50 to 60 minutes of genuine high intensity work across an entire week. That is two solid interval sessions. Not ten hours of smashing yourself.
WHO WAS BEING STUDIED
This is the other bit that gets completely ignored. The athletes Seiler was looking at were training 15 to 25 hours per week. Some were doing over 1,000 hours per year. These are full-time or near full-time athletes. Norwegian cross-country skiers. International rowers. Elite distance runners. Their easy volume alone was 12 to 20 hours a week.
If you are training 8 hours a week across 5 sessions, your training week looks fundamentally different to someone doing 22 hours across 12 sessions. Taking the same intensity distribution and forcing it into a third of the volume does not automatically give you the same outcome. The sheer amount of easy aerobic work those elites were doing is the foundation. The 80% is not filler. It is the entire point. It builds the engine. Without the volume to support it, you are just copying the shape of the programme without the substance.
THE AMATEUR VERSION OF "80/20"
Here is what I see over and over again. Someone reads about 80/20 and decides to apply it. What actually happens is this. They ride easy most of the time because that is the fun, comfortable bit. The part that feels virtuous. Look at me, riding in Zone 2 like the elites. But when it comes to the hard 20%, they either skip it entirely, go half-hearted, or end up doing something that is not actually hard enough to count. A "tempo" ride that sits right in the middle ground Seiler specifically said to avoid.
The bottom bar is the one I see most. Nearly half the training time spent in that no-man's land. Too hard to build aerobic capacity properly. Too easy to drive real high intensity adaptation. Every ride feels like it is "something" but nothing is actually pushing the needle. Seiler calls this moderate intensity trap or training monotony. Everything ends up at the same effort. Every ride a 6 out of 10. Never a 3 and never a 9.
THE HARD PART IS ACTUALLY THE HARD PART
People love to talk about Zone 2. It is the sexy topic right now. Every podcast, every YouTube video, every Instagram post. Zone 2 this, Zone 2 that. And yes, easy aerobic training is important. Nobody is arguing against that. But the part of polarised training that most amateurs skip or water down is the bit that arguably matters most for someone with limited training hours: the genuinely hard sessions.
When Seiler talks about the high intensity 20%, he is talking about work above the second lactate threshold. In practical terms for a cyclist, that means intervals at or above VO2max pace. Four by eight minutes at an effort where you are hanging on by the last rep. Not sweet spot. Not "comfortably hard." Properly hard.
If you only have 6 to 8 hours a week and you spend 80% of that going easy, you have maybe 60 to 90 minutes for hard work. That is not a lot. You need to make those minutes count. And "count" means the intensity has to be high enough to actually drive adaptation. A tempo ride at 85% of FTP is not doing that job.
WHAT IT LOOKS LIKE IN THE REAL WORLD
Let me put this in practical terms for someone doing 5 to 6 rides a week with maybe 8 to 10 hours total.
Look at the difference. The elite athlete has massive amounts of easy work either side of their hard sessions. They can absorb the intensity because the volume of easy training is doing the heavy lifting in terms of aerobic development and recovery. The amateur has much less buffer. Which means the easy sessions need to be genuinely easy, not moderate, and the hard sessions need to be properly targeted. There is no room for junk miles in between.
THRESHOLD IS NOT BANNED
Another thing that gets lost in the 80/20 conversation. Polarised training does not mean you should never ride at threshold. Even in the research, 5 to 10% of sessions sat in the moderate zone. That is not zero. There are times in a season where threshold work makes sense, particularly as a bridge between base and race-specific preparation, or for athletes who respond well to sustained efforts. The point is not to eliminate it. The point is that it should not be the default for every session that is not "easy."
The problem with threshold or sweet spot as your primary training stimulus is that it is hard enough to generate real fatigue but not intense enough to drive the top-end physiological adaptations you get from VO2max work. And because it generates fatigue without maximum adaptation, it eats into your ability to recover for the sessions that really matter. You end up tired but not improving. That is the moderate overkill trap.
SO WHAT SHOULD YOU ACTUALLY DO
Stop worrying about hitting a perfect 80/20 split. That was never the point. The actual takeaways from Seiler's work are simpler than the internet has made them.
Make easy genuinely easy. Below the first lactate threshold. A pace where you could hold a full conversation without any effort. Most people's "easy" is way too hard. If you are breathing through your mouth, it is not easy enough.
Make hard genuinely hard. Above the second lactate threshold. Intervals where you are at or near VO2max effort. The last rep should feel like you are barely holding on. If you could do two more reps at the same power, it was not hard enough.
Minimise the middle ground. That "comfortably hard" zone where you feel like you are doing something but the stimulus is not strong enough to drive meaningful adaptation. Some threshold work is fine. Living there is not.
Respect your volume. If you are training 8 hours a week, you cannot just copy a 20 hour programme and shrink it. The principles still apply but the emphasis shifts. With less total volume, every session has to be more intentional.
The best thing you can take from Seiler's research is not a specific ratio. It is the principle that training should not all feel the same. Easy should feel genuinely easy. Hard should feel genuinely hard. Most amateurs exist in a grey zone where everything is a 6 out of 10 and nothing ever really moves the needle.
If you are a working cyclist doing 6 to 10 hours a week and you want to get faster, you do not need to obsess over percentages. You need to be honest about whether your easy rides are actually easy, whether your hard sessions are actually hard, and whether you are doing enough of both. That is the actual lesson. The numbers are secondary.
So next time someone tells you that you need to do more Zone 2, ask them how much they are doing. Ask them how hard their hard days are. Ask them if they know the difference between counting sessions and counting time. And if they cannot answer those questions, they probably got their information from the same Instagram post as everyone else.