I READ EVERY STEPHEN SEILER PAPER I COULD GET MY HANDS ON. HERE'S WHAT I LEARNED.

I went down a rabbit hole over the past few weeks.

It started the way these things usually start. A member of Not Done Yet asked me in a weekly rider support call what I actually thought about polarised training. Not the Instagram version. Not the Peter Attia version. The real thing. And I realised that even though I've had Stephen Seiler on the podcast a few times, even though I quote his work to members every week, even though I build our whole NDY reverse-periodisation model on the back of what he figured out in a lab in Norway twenty years ago, I'd never actually sat down and read the papers.

Not the summaries. Not the Fast Talk articles. The actual peer-reviewed PDFs.

So I did. Over about eight weeks. Every Seiler paper I could find. The famous ones. The obscure ones. The Sportscience reviews. The 2013 interval study. The 2007 autonomic nervous system paper that nobody ever talks about. His 2024 piece on HIIT, which he's publicly said on X is the thing he's most proud of writing in his entire career.

Here's what I learned.

Key Takeaways

• Seiler didn't design the 80/20 model — he measured what elite endurance athletes were already doing and the same pattern kept showing up across sports.
• VT1 is a hard binary. Below it your nervous system barely notices a session. Above it, even briefly, you trigger a systemic stress response that delays recovery.
• The threshold zone — the grinding moderate-hard middle — is where most amateurs stall. Pros use it surprisingly sparingly.
• Volume and frequency sit at the base of Seiler's Hierarchy of Endurance Training Needs. Everything else is a multiplier on that base.
• The 4x8 at highest sustainable power is the highest-yield single session in the whole Seiler canon.
• "Maximal effort" means different things on a 4-minute vs a 16-minute bout. Your body silently recalibrates pacing around duration.
• Training is an optimisation problem across months and years, not a maximisation problem across single sessions.

The 80/20 rule wasn't invented. It was observed.

Seiler didn't sit in a lab in Kristiansand and design the 80/20 distribution. He just went and measured what the best endurance athletes in the world were actually doing — cross-country skiers, rowers, runners — and the same pattern kept showing up.

His 2006 paper with Glenn Kjerland is where it starts. They tracked eleven nationally competitive junior cross-country skiers over 32 consecutive days and analysed 318 of their endurance sessions. Whether they measured it by heart rate, by session RPE, or by blood lactate, the numbers landed in roughly the same place: about 75% of sessions clearly below the first lactate threshold, 5-8% in the middle zone, and 15-20% at high intensity.

Then came the rowing paper. Fiskerstrand and Seiler went back through thirty years of training data from Norwegian rowers who'd won European, World or Olympic medals between 1970 and 2001. Twenty-eight athletes. Twenty-seven still alive. Twenty-one of them filled out detailed questionnaires about what they actually did.

The headline is the one you'll have heard a hundred times. But the numbers are where it gets interesting. Annual training volume went up about 20%, from around 924 hours a year in the 1970s to 1,128 hours in the 1990s. Training below 2 millimoles of blood lactate — the stuff recreational riders would call "too easy" — nearly doubled, from 30 hours a month to 50. And very hard training in the 8-14 millimole range dropped from 23 hours a month down to about 7. VO2max went up 12% across those decades. Six-minute erg performance went up nearly 10%.

Read that again. The Norwegian rowers who kept winning medals weren't getting better because they trained harder. They were getting better because they trained more. And the extra hours they added were almost entirely easy.

That's not my opinion. That's what the data from three decades of Olympic medallists says.

VT1 is the line that changes everything.

The paper of Seiler's that blew my head off isn't one of the famous ones. It's a 2007 study in Medicine & Science in Sports & Exercise, co-authored with Olaf Haugen and Eike Kuffel, called "Autonomic Recovery after Exercise in Trained Athletes."

It doesn't get the clicks because autonomic nervous system research is boring to most people. But this is the paper that explains why the whole polarised model actually works at a physiological level, not just a "pros do this" level.

Here's the setup. Nine highly trained male runners with VO2max scores around 72, training 14 hours a week. They did four different sessions in random order: 60 minutes below VT1, 120 minutes below VT1, 60 minutes with 30 of them between VT1 and VT2 (a classic threshold session), and 60 minutes with an interval block above VT2. Then they measured heart rate variability — which is basically a window into how stressed the autonomic nervous system is — for four hours afterwards.

The result is the thing every serious cyclist needs tattooed on their forearm. Up to 120 minutes of running below the first ventilatory threshold caused basically no disturbance to autonomic nervous system balance in the highly trained athletes. Zero. The moment intensity went above VT1, recovery was significantly delayed. And once it went above VT1, it didn't matter whether they were at threshold or well above it — the delay was about the same. VT1 acted as a binary threshold. Below it, the body barely noticed. Above it, full stress response.

In the less-trained group, recovery of parasympathetic tone after the hard interval session was slower by roughly 60 to 90 minutes compared to the highly trained guys.

Think about what that means in practice. A two-hour zone 1 ride is, from your nervous system's perspective, essentially free. A 45-minute ride with 20 minutes of threshold work costs you the rest of the day. The cost isn't the calories or even the muscular fatigue. It's the systemic stress response that delays every adaptation happening in the background.

This is why the easy days have to be actually easy. Not "oh I'll just cruise at 78% of FTP." Actually easy. Below VT1. Because if you cross that line, even for a bit, the session stops being free and starts eating into the recovery you need for the hard stuff.

The full zone 2 guide goes deeper on what "below VT1" actually looks like in practice — nose-breathing, RPE, power caps, the lot.

The threshold rut is where most amateurs die.

If there's one phrase Seiler uses more than any other in his papers, it's this one: elite athletes train "surprisingly little at lactate threshold intensity." His 2010 review in the International Journal of Sports Physiology and Performance makes the point bluntly. Athletes training 10 to 13 times a week converge on the same pattern: about 80% at low intensity, about 20% dominated by high-intensity work near 90% of VO2max. Threshold work — that grinding, in-between intensity most amateurs live in — is used surprisingly sparingly.

There's an observation that keeps showing up across his papers. In the 2006 Kjerland study, he concluded that elite endurance athletes train surprisingly little at the lactate threshold intensity. In the 2010 IJSPP review, he reported that training intensification studies on already well-trained athletes have produced no convincing evidence of long-term performance gains. Put those two findings next to each other and you get a pretty clear message: the best in the world avoid living at threshold, and when you force already-fit athletes to spend more time there, you mostly get fatigue and very little adaptation to show for it.

And here's the uncomfortable part. When Seiler looks at recreational athletes training around 350 hours a year, their distribution tends to look nothing like the pros. They look like threshold athletes. They spend a huge chunk of their hours in zone 2 — the moderate, nearly-hard zone — because that's where every ride drifts when you only have ninety minutes and you want to feel like you worked.

The pros don't train that way. Not one of them. And the data Seiler has from well-trained junior skiers, from national-team rowers, from Grand Tour cyclists, all says the same thing: the moment annual volume starts pushing above 750 hours, the distribution naturally shifts either to polarised or pyramidal, and threshold drops out as the default intensity.

The threshold rut isn't where progress lives. It's where progress goes to stall.

Volume beats everything else. It's not close.

In 2016 Seiler gave an invited lecture where he proposed something he called the Hierarchy of Endurance Training Needs. Eight layers, Maslow-style, with the most important at the bottom.

The bottom layer isn't polarised training. It isn't intensity distribution. It isn't VO2max intervals or tapering or altitude camps or any of the sexy stuff.

It's frequency and volume.

According to Seiler, the single most beneficial thing you can do to improve your endurance performance is train a lot. The more hours, the more sessions, the better. Everything else in the pyramid sits on top of that base.

Above volume you get high-intensity training. Above that, intensity distribution. Then periodisation. Then micro-periodisation, which Seiler rates as "likely modest" in its impact. Altitude and heat? Potentially important but individual and condition-specific. It's the final 1-2%. If you haven't nailed the layers below, it does nothing for you. Pacing and tapering sit at the top — potentially decisive, but only once everything underneath is done right.

This is the hierarchy the internet has upside down. Every Instagram thread wants to talk about the top of the pyramid — the clever periodisation trick, the magic 4-minute interval, the 3-week altitude block — while ignoring the fact that if you're training six hours a week, none of that matters as much as adding a seventh.

When I had John Wakefield on the podcast, he said basically the same thing in different words. Dan Lorang too. The coaches at the top of WorldTour cycling aren't obsessing over periodisation models. They're making sure their riders are accumulating the hours week after week, year after year, without blowing themselves up.

Seiler's hierarchy is just the evidence base under that intuition.

The 4x8 is the session I'd pick if I could only do one.

In 2013 Seiler and his team published one of the most practical papers in the entire endurance training canon. It's the 4x4 vs 4x8 vs 4x16 study, in the Scandinavian Journal of Medicine & Science in Sports.

Thirty-five recreational cyclists with a VO2max around 52. They trained about six hours a week. They got split into four groups for seven weeks: one group did only low-intensity training, one group did 4x4 minutes with 2-minute recoveries, one group did 4x8 minutes, and one group did 4x16 minutes. All the interval groups did two sessions a week alongside their base training.

VO2max went up 10.4% in the 4x8 group, 6.5% in the 4x16, 5.6% in the 4x4, and 3.4% in the low-only group. Threshold power jumped 16% in the 4x8 group. Sixteen percent. In seven weeks. In riders who were already training six hours a week.

Why did the 4x8 win? This is where Seiler's own explanation is gold. He's suggested that 4x4 sessions are often over-paced. The temptation to go too hard on the shorter bouts is real, and failed sessions are common. The 4x16 sets tend to get pulled down to a slightly easier intensity and become mentally demanding in a different way. The 4x8 hits the sweet spot — enough duration to accumulate real time at high intensity, short enough that you can hold something meaningful without the pacing falling apart.

If you're a time-crunched amateur training seven to ten hours a week and you want one HIT session a week that earns its keep, it's 4x8 at the highest sustainable power. Two-minute recoveries. Done. More detail on how to execute it here.

Maximal session effort isn't what you think.

In a 2017 paper, Seiler and Øystein Sylta analysed the data from a larger HIIT intervention study: over 1,400 HIIT sessions performed by 63 well-trained cyclists doing 4x4, 4x8 and 4x16 prescriptions over 12 weeks. Every session was told to be "maximal session effort" — go as hard as you can for the full session.

Here's what they found, and it's one of the quiet revelations of the whole Seiler canon. The 4x4 prescription was four to six times more likely to elicit a near-maximal RPE of 19-20 on the Borg scale than the 4x16. Across nearly 500 completed 4x16 sessions, only 8% hit a peak RPE of 19 or 20. In the 4x4 sessions, 61% hit that peak.

Same instructions. "Maximal effort." Completely different physiological outcomes. The longer the work bout, the less likely you are to push yourself to an all-out place — because your body self-regulates based on how much more work it thinks is coming.

This is the thing I want you to sit with. Your body doesn't interpret "maximum" the same way on a 16-minute effort as on a 4-minute one. Not because you're soft. Because that's how pacing works. Which means the kind of stress you generate in a session isn't just about how hard you try. It's about the duration of the work bouts, which silently recalibrates what "hard" even feels like.

Training is an optimisation problem, not a maximisation problem.

Seiler published a paper in 2024 in Applied Physiology, Nutrition, and Metabolism called "It's about the long game, not epic workouts: unpacking HIIT for endurance athletes." He said on X that it's the piece of science writing he was most happy with in 2024. I can see why.

The argument goes like this. A lot of HIIT research treats interval prescription as a maximisation problem. You're trying to find the single protocol that produces the biggest acute VO2 response, the highest peak heart rate, the most time above 90% of VO2max. That's how the studies are designed.

But that's not how successful endurance athletes train. Successful endurance athletes perform the vast majority of their training below the first lactate turn point, with threshold work and HIIT as integrated but smaller components. The molecular signalling that drives long-term adaptation involves multiple overlapping pathways, many of which show substantial feedback inhibition at high intensities — which means the previous training content and longer-term training history are critical modulators of how you respond to any given HIIT session. And the timeframe for long-term maximisation of endurance capacity extends over years, not weeks.

His conclusion is the sentence that should be taped to every cyclist's stem. It's not "epic" HIIT sessions but the effective integration of intensity, duration, and frequency of all training stimuli over time that drives endurance performance success.

Read it twice. The reason most age-groupers plateau isn't that their hard days aren't hard enough. It's that they're trying to maximise every single session instead of optimising the whole week, the whole month, the whole year. Training is not a series of hero workouts. It's an accumulation.

Zone 2 got hijacked, and the scientists had to take it back.

If you've been anywhere near cycling Twitter in the last two years you know Zone 2 became a religion. Peter Attia, Andrew Huberman, every longevity podcast on the planet. Suddenly everyone has a lactate meter and a nasal-breathing protocol and a very strong opinion about whether you can talk in full sentences.

In 2025 Seiler was one of fourteen co-authors on a consensus commentary published in the International Journal of Sports Physiology and Performance called "What Is Zone 2 Training?" It's not a Seiler paper exactly, but he's on it, and it reflects the cleanest current thinking on what Zone 2 actually is.

The expert panel reached consensus that Zone 2 training should preferably be performed at intensities located immediately below the first lactate or ventilatory threshold — LT1 or VT1 — and that it can be delivered through continuous, variable or interval-type sessions.

At this intensity, you should see a steady-state physiological response. Blood lactate holds flat or drifts up only transiently to around 1-2 millimoles. Heart rate sits at roughly 70-80% of maximum, or 80-90% of your LT1 heart rate. RPE stays low, about a 10 on the 6-20 Borg scale. Power output lands around 75-80% of critical power.

Here's what's important. That's not "chat pace." That's not "zone 2 as a feeling." That's a specific physiological window immediately below the first threshold, where your body is working at a rate it can sustain almost indefinitely without meaningful stress — which is exactly the condition Seiler identified in that 2007 autonomic nervous system paper as the threshold for zero ANS disturbance.

The science is internally consistent. Zone 2 isn't a vibe. It's the zone where you can accumulate huge amounts of aerobic stimulus without dipping into the stress response. The Instagram version of zone 2 is usually higher than that. And the reason your friend swears she's doing zone 2 but never improves is almost certainly because her zone 2 is actually somebody else's tempo.

What pros actually do differently isn't mysterious.

Across every paper I read, one pattern came up again and again. The best endurance athletes in the world — the ones Seiler studied in cross-country skiing, rowing, cycling, running — aren't doing anything exotic. They're doing the obvious things, consistently, for years.

They train a lot. They train most of that volume easy, below VT1, where the body is undisturbed and can accumulate work without cost. They include a small but non-negotiable dose of genuinely hard training, usually one to three sessions a week, often in the form of 4-8 minute intervals at 90-95% of VO2max. They don't live in the threshold zone because they know it's the worst of both worlds — too hard to be recovery, not hard enough to force real adaptation. They don't chase epic workouts. They chase months and years of accumulated work.

And when Seiler and his team compared 38 years of training distribution data from Olympic speed skaters, or 30 years from Norwegian rowers, or a year of training leading into a gold medal performance from a range of endurance champions, the picture was the same in every sport. Volume up. Low-intensity volume up relatively more. High-intensity volume refined and carefully applied. Threshold work kept small on purpose.

So what do you actually do with this?

Here's where I'll keep it simple, because you don't need a periodisation chart — you need a rule of thumb.

If you're training less than eight hours a week, the first thing you need to do isn't optimise your intensity distribution. It's add a session. Seiler's hierarchy is very clear about this: frequency and volume are the foundation, and nothing higher up the pyramid matters as much as that base. Ride more. Even if it's another hour of easy. Our base training guide is the place to start.

If you're training eight to twelve hours a week and you're stuck, there's a high chance your distribution is wrong. Get honest with yourself about how much of your riding drifts into that moderate-hard grey zone where nothing good happens. Make the easy rides actually easy — below VT1, below first ventilatory threshold, the place where you can breathe through your nose and hold a conversation in full sentences. And make the hard rides actually hard — one or two sessions a week, 4x8 at the highest sustainable intensity, two-minute recoveries.

And if you want one thing to take from eight weeks of reading the papers, it's this. The guys who studied the best endurance athletes in the world, for thirty years, from a lab in Norway, didn't find a secret. They found a pattern that's been hiding in plain sight. Train a lot. Train most of it easy. Do a small amount really hard. Don't live in the middle. Don't chase hero sessions. Let the weeks stack up into months and the months stack up into years.

That's it. That's the whole thing.

Want this applied to your training?

If any of this sounds familiar, it's because it's exactly what we build the Not Done Yet coaching on. Reverse periodisation. Heavy low-intensity base. Small, precise doses of real HIT. No time wasted in the threshold rut. It's not because I invented it. It's because Seiler, Fiskerstrand, Tønnessen, Haugen and about twenty years of Norwegian Olympic data all point in the same direction — and I'd rather build on that than on the latest hot take.

Applied to your power numbers, your events, your calendar, your life — that's what Not Done Yet coaching does. Seven-day free trial. We'll build you the plan a lab in Norway would have built for a pro twenty years ago.

There's more in me. There's more in you.

Now go and make your easy rides easier.