Deep dive/Training science
Intensity distribution from real training logs, elite versus recreational. The amateurs are the ones going hard.
Two ways to spend a week. The horizontal axis is intensity, the height of each curve is how much of the athlete’s training time is spent there.
Shape of curves:illustrative, drawn to match the distributions reported in Seiler & Kjerland 2006 and Esteve-Lanao et al. 2005 (references 1 and 4). These are not plotted from raw participant data. The measured percentages behind them appear in chapter 4.
On a Saturday morning, in any city with a running club, ten people set off on what the schedule calls an easy run. For the first mile it is easy. Then someone drifts to the front and the pace tightens by fifteen seconds. Conversation thins from paragraphs to sentences, then to single words, then to nothing at all. By the turnaround the run has quietly become a race that nobody agreed to enter. Everyone finishes it moderately wrecked, and pleased with themselves, having gained close to nothing.
Somewhere else that same morning, a marathoner with a personal best under two hours and ten minutes ran for two hours at a pace that, measured against what their body can actually do, was slower than the club run. They talked the entire way. They finished able to do it again tomorrow, and tomorrow they will.
The gap between those two mornings is one of the more consistent findings in endurance training research, and it points the opposite way from intuition. Go and measure what fast people actually do, session by session, week after week, and the striking thing is not how hard their hard days are. It is how genuinely, almost boringly easy their easy days are.
The most reliable difference between how elites train and how the rest of us train is not the ceiling. It is the floor.
This turns out to be measurable, and it has been measured many times: in Norwegian skiers and rowers, in Spanish distance runners, in professional cyclists across a full season, in the training year preceding an Olympic gold medal. The numbers differ between sports and between accounting methods, and we will get to why that matters. The shape does not differ much at all.
There is also a catch buried in the headline of this article, and it is the most interesting part of the story. The claim that elites train easier than you is, taken literally, false. We will get to that too, in chapter 8, and it will require throwing out the slogan that most people take away from this research.
Start with the physiology, because none of the training data means anything without it.
Your body does not experience effort as five coloured bands on a watch face. It experiences two transitions, and they are real physiological events rather than administrative divisions.
The first is the point at which blood lactate lifts off its resting baseline. Below it, whatever lactate your muscles produce is cleared about as fast as it appears, and you sit in a steady state you could hold for a very long time. Fuel comes substantially from fat. Breathing is deeper than at rest but unremarkable. This is commonly called the first lactate threshold, or LT1, and it tends to land somewhere near two millimoles of lactate per litre of blood.
The second is the highest intensity at which lactate production and clearance still balance. Push past it and lactate begins to accumulate rather than plateau, and a clock starts running. Depending on which laboratory you ask, this is the second lactate threshold, the maximal lactate steady state, critical power, or functional threshold. It often lands near four millimoles.
Those two points cut the intensity range into three zones. Below LT1 is easy. Above LT2 is hard. And between them sits a band of effort that is genuinely uncomfortable, genuinely tiring, and, for the purposes of getting faster, largely a tax.
The curve below is the classic lactate response to increasing intensity. Move the marker to see where the thresholds fall, what the effort feels like to speak through, roughly how long it could be held, and what it buys you in exchange for what it costs.
Curve: a smooth function fitted to pass through 2 mmol/L at LT1 and roughly 4 mmol/L at LT2, the conventional anchors used in the three-zone literature (references 2, 14 and 15). Holdable-for values: order of magnitude estimates for a trained athlete, not measured durations. Individual thresholds vary substantially, and the only way to know yours is to have them tested.
Stephen Seiler, the physiologist whose laboratory did much of the original work of mining athletes’ training diaries, has called that middle band the moderate intensity rut. Coaches call it the grey zone. The unkinder name, and the one that sticks, is the black hole: effort falls into it and very little comes back out.
One honest caveat before we build anything on top of this. The names of these thresholds are genuinely contested, and have been for fifty years. What LT1 and LT2 mean physiologically, whether lactate causes anything or is merely a convenient marker of something else, and where exactly the lines fall in any given person, are all live arguments. What is not seriously disputed is that the two transitions exist, that they are detectable, and that the region between them behaves differently from the regions either side.
Before the data, a detour that saves a great deal of confusion, because two completely different numbering systems are in circulation and they collide directly.
The physiological model in this article has three zones. Most watches, and most training apps, use five, and some use seven. In the three zone model, zone 1 is everything below the first threshold, which is a very large territory covering most of what an elite athlete does. In a five zone heart rate model, zone 1 is a narrow strip at the bottom that many people treat as barely training at all.
So when a coach says most of your training should be in zone 1 and your watch says you spent eight minutes there, the two are not talking about the same thing. The popular phrase zone 2 training makes it worse. In five zone language, zone 2 is a comfortable aerobic effort, which sits below the first threshold and is therefore easy in three zone language. In three zone language, zone 2 is the black hole, the thing you are supposed to avoid.
Three-zone easy, below LT1, corresponds to about 60 to 82 percent of maximum heart rate, which is five-zone zones 1 and 2. Three-zone moderate, between the thresholds, is about 82 to 92 percent of maximum heart rate, five-zone zones 3 and low 4. Three-zone hard, above LT2, is roughly 92 percent and up.
These conversions are approximate and individual. Heart rate also drifts upward over long sessions at constant effort, which is one reason threshold work is better anchored to pace, power or lactate than to heart rate.
For the rest of this article, zones mean the three physiological ones, and intensity is expressed as a percentage of VO2max rather than of heart rate, because that is how most of the underlying literature reports it.
In 2006, Seiler and Glenn Kjerland did something that sounds obvious and had rarely been done properly. Rather than asking athletes what they thought their training looked like, they took junior elite cross-country skiers, monitored them across eight weeks, and classified every session by its intended intensity.
About three quarters of sessions were aimed below the first threshold. Roughly eight percent sat in the middle band. The remainder, around seventeen percent, was unambiguously hard. The distribution had two peaks and a hollow between them, and it has been given the name polarized.
Then people went looking, and kept finding the same shape.
Each bar is one cohort, split into the three physiological zones. Note how the middle segment behaves.
Elite and national level
Norway, 8 weeks of logs
Seiler & Kjerland 2006, by session goal
Norway, national team
Fiskerstrand & Seiler 2004, by time in zone, approximate
Spain, 6 months to national championships
Esteve-Lanao et al. 2005, by time in zone
Full competitive season
After Lucía et al. 1999 and 2000, approximate
Recreational
Self-directed training, 3 to 6 hours a week
Composite of the pattern reported across recreational cohorts
Self-directed, group rides and studio classes
Composite of the pattern reported across recreational cohorts
Elite rows: figures as reported or closely approximated from references 1, 6, 4 and 12 respectively, with the exact study and accounting method printed under each bar. Recreational rows: flagged as composite. They represent the pattern reported across studies of self-directed recreational athletes rather than a single measured cohort, and should be read as illustrative of the shape, not as precise values.
The rowing row deserves a second look, because it is the most extreme and the most instructive. Alexander Fiskerstrand and Seiler went through thirty years of Norwegian national team rowing records, from 1970 to 2001. Across those decades the programme changed in a specific direction: total volume of low intensity work went up substantially, while time spent at and just below race intensity came down. Measured oxygen uptake improved. So did results. The team got faster by, in effect, training easier more often.
The Spanish running study is useful for a different reason. Esteve-Lanao and colleagues tracked sub-elite distance runners across six months and then looked at who ran fastest. Time spent in the easy zone was the variable that best tracked with performance. Two years later the same group ran the controlled version: two matched groups, one doing about eighty percent easy, the other about sixty-five percent, with the difference made up in the middle zone. The group that stayed easier improved more.
A team spent thirty years quietly moving work out of the middle zone. Their oxygen uptake went up and their race results improved.
Norwegian international rowers, 1970 to 2001The distributions are not identical across sports. Rowers and cyclists accumulate enormous low intensity volume partly because their sports are not weight bearing and the tissue damage per hour is lower. Runners cannot simply add hours without accumulating impact. But the direction is the same everywhere, and the middle band is thin everywhere.
Scroll through the two shapes side by side, because the picture carries the argument better than the percentages do.
01 of 3
One broad hump, centred in the middle of the intensity range. Not easy, not hard. A pace that feels like training, held for most of the week, on most of the days.
02 of 3
Two peaks and a valley. A large mass of work well below the first threshold, a small spike of genuinely hard work above the second, and remarkably little in between.
03 of 3
The band between the two thresholds is where the recreational athlete lives and where the elite athlete refuses to go. Everything else in this article is an attempt to explain why.
Chart: the same illustrative curves as the opening figure, shaped to match the reported distributions in references 1 and 4. Not plotted from raw data.
If the middle zone is such a bad deal, why does almost everybody training without a coach end up living there? Because it is the only zone that reliably feels like training.
Easy work, done properly, feels close to pointless. You finish a genuinely easy hour and your internal accounting says you did not really do anything. Hard work is unpleasant enough that most people will not choose it without a plan telling them to. The middle is the compromise: it hurts a respectable amount, it leaves you tired enough to feel virtuous, and it does not require the specific commitment that intervals demand.
There is a second reason, which is social. Groups drift upward. Any run with more than about four people in it will settle at the pace of the fittest person having a good day, which is moderate for them and moderate-to-hard for everyone else. Nobody decides this. It just happens, every Saturday.
The physiological problem with the middle is easiest to see if you hold fatigue constant instead of holding time constant. Give three athletes the same fatigue budget for one session, let each spend it entirely in one zone, and ask what each one comes home with.
One fixed fatigue budget, spent entirely in each zone. The session lengths fall out of the cost, and the returns fall out of the length.
Every number above follows by arithmetic from four inputs. Fatigue per minute is set at 1 for easy, 2.6 for moderate and 6.5 for hard. Each column then spends the same fatigue budget of 120 units, which is what fixes the session lengths. Aerobic stimulus per minute is set at 1.0, 1.35 and 1.5, and top end stimulus per minute at 0, 0.15 and 3.2.
Those coefficients are a defensible reading of the physiology, not measured constants. Change them and the columns move. What does not move, across any plausible set of values, is the shape of the result: the middle column is never first at anything.
This is a model, not a measurement. It is arithmetic from the four stated coefficients, which are a reading of the mechanisms described in references 2, 3 and 18. No study reports these exact values. The chart is included because the argument is easier to check when the assumptions are written down than when they are left implicit.
The easy column buys roughly twice the aerobic stimulus of the moderate column, for the same fatigue, because it buys more than twice the minutes. The hard column buys an order of magnitude more top end. The moderate column comes second in both categories and first in neither, which is the entire case against it.
And there is a cost that this chart does not capture, because it looks at one session in isolation. Moderate work is contaminating. A Tuesday spent in the middle does not just fail to produce much on Tuesday. It makes Thursday’s intervals slower, which means Thursday buys less top end than it should have. Run that pattern for a month and both ends of the distribution quietly degrade.
The case for easy volume is not that it is pleasant. It is that several of the adaptations that make someone a good endurance athlete respond primarily to how long you go, and are close to indifferent to how hard.
Mitochondria. Endurance training increases both the number and the density of mitochondria in trained muscle, which is what raises the ceiling on how much oxygen the muscle can actually use. The signalling that drives this responds strongly to accumulated contractile time. Long easy work is an efficient way to accumulate a great deal of it.
Capillaries. More capillaries per muscle fibre means more surface area for delivering oxygen and clearing metabolites. This remodelling is driven substantially by sustained blood flow and shear stress, which again is a function of duration.
Stroke volume. Prolonged submaximal work loads the heart with a high volume of returning blood at a manageable rate, which over time increases how much blood the left ventricle moves per beat. A bigger stroke volume is the main reason a trained resting heart rate is low, and it directly raises the ceiling on oxygen delivery.
Fat oxidation. Training the body to use fat at a given intensity spares the limited store of muscle glycogen. That matters enormously in events lasting more than about ninety minutes, and the adaptation is driven by working at intensities where fat is actually the dominant fuel, which means below the first threshold.
Each of those responds to time under load. And time under load has a hard constraint: you can only accumulate a lot of it at intensities you can repeat day after day without accumulating damage. That constraint is the whole reason the easy zone is large in elite training. It is not a philosophy. It is arithmetic about how many hours a human body can absorb.
This is also why the largest and most direct predictor of performance in the studies keeps turning out to be something unglamorous. When Casado and colleagues looked at what best predicted world class long distance running performance, volume of easy running was among the strongest factors, alongside deliberate practice of short intervals and tempo work. Not the hard sessions alone. The easy volume that made them possible.
None of this is an argument for training slowly all the time. That misreading is common, and the data does not support it. The elite distributions contain a real, deliberate spike of very hard work, and it is there because some adaptations simply cannot be bought any other way.
To improve maximum oxygen uptake, you generally have to spend time at or near maximum oxygen uptake. That means intensities at which the cardiovascular system is operating at its ceiling, which by definition cannot be sustained for long, which is why this work arrives as intervals. Once an athlete is already well trained, the evidence is fairly clear that further gains in VO2max and in race performance require this kind of stimulus, and that adding volume alone stops producing much.
High intensity work also drives the machinery for moving lactate between and out of cells, improves neuromuscular recruitment and running economy at speed, and produces the specific tolerance for racing that no amount of easy running produces. An athlete who has only ever run easily is aerobically fit and racing badly.
Easy work builds the engine. Hard work raises its ceiling and teaches it to be used. The middle mostly builds fatigue.
So the polarized distribution is not a compromise between the two. It is a recognition that they are separate purchases with separate price tags, and that the sensible move is to buy each one at the intensity where it is cheapest. Aerobic base is cheapest well below the first threshold. Top end is only available well above the second. Buying either one in the middle means paying more and getting less.
Here is where the popular version of this research falls apart, and it falls apart on something as basic as division.
Eighty percent easy and twenty percent hard is a ratio. Ratios are silent about totals. Two athletes can have identical distributions and completely different training loads, and the one with more hours is doing more of everything, including more of the hard work.
Take the numbers seriously. A recreational runner training four hours a week, splitting eighty twenty, does about forty eight minutes of hard work. An elite marathoner training twenty hours a week, with the same split, does about four hours of it. The elite is doing five times as much hard training, and roughly five times as much easy training, and about five times as much of everything.
Move the slider to your own weekly hours and compare the absolute quantity of hard work, not the percentage.
Absolute hard minutes per week, everyone at the same 80/20 split
Four sessions a week
Six sessions, one long
Thirteen sessions, about 200 km
4 hours a week
Athlete archetypes: representative weekly volumes, not measured individuals. Twenty hours and roughly thirteen sessions a week is a conventional figure for elite marathon training, consistent with the volumes described in references 10 and 11. The 80/20 split applied to every row is deliberately identical so that only total volume varies.
So the honest version of the finding is not that elites train easier than you. It is this: elites train easier per hour, and vastly more in total, and the first thing is precisely what makes the second thing survivable.
That reframing matters, because it changes what you are supposed to do about it. If elites simply trained easier, the lesson would be to slow down. But they do not simply train easier. They use easy intensity as the mechanism that allows a very large total volume, and that large volume is what builds the capacity to absorb four hours a week of genuinely hard work without falling apart.
Easy training, in other words, is not the goal. It is the thing that pays for the goal.
Which raises an uncomfortable question for anyone training four or five hours a week, and we should deal with it honestly rather than pretend the research says something tidier than it does.
One study is worth describing on its own, because it looks at the extreme end of the distribution: athletes in the twelve months leading up to winning a world championship or an Olympic title.
Tønnessen and colleagues assembled training data for a set of gold medal endurance performances and looked at what the final year actually contained. The pattern is not exotic. Very high total volume, the overwhelming majority of it at low intensity, maintained with remarkable consistency across months. A modest, stable amount of high intensity work rather than a dramatic build. And in the final weeks before the target, volume comes down while intensity is largely preserved, which is what tapering actually means in practice.
The thing that stands out reading this kind of data is how unremarkable any individual week looks. There is no session in there that a committed amateur could not complete. What an amateur could not do is complete that week, and then the next one, and then eighty more, without breaking. The training is not impressive. The accumulation is.
A useful way to think about it: an athlete who trains at seventy percent of their theoretical capacity for fifty weeks will beat an athlete who trains at a hundred percent for six weeks and is then injured or exhausted for four, repeatedly. Most amateur training failures are not intensity failures or volume failures. They are consistency failures, and the middle zone is one of the most efficient ways to produce them.
This is the chapter that most articles about polarized training leave out, and leaving it out is how a reasonable finding turns into an internet rule.
The research is descriptive first.Most of the foundational work observed what successful athletes already did. That is genuinely useful evidence, and it is not the same as proving that the distribution caused the success. Elites might be training this way partly because it is what their sport’s culture teaches, or because it is what survives contact with very high volume. The controlled trials that followed, from Esteve-Lanao, Neal, Stöggl and Sperlich, generally favour the polarized approach, but they are small, short, and run on already trained athletes.
The meta-analyses are more cautious than the slogans. When Rosenblat and colleagues pooled the polarized versus threshold comparisons, the advantage was real but more modest and less certain than the individual studies suggest. A separate review of periodization and distribution in distance running concluded that no single distribution has been demonstrated to be optimal for everybody. That is the honest state of the evidence.
Low volume changes the maths.This is the important one. The argument for a large easy fraction depends on easy volume being large enough to drive adaptation by itself. At twenty hours a week, sixteen easy hours is a powerful stimulus. At four hours a week, three easy hours is not very much of anything, and strict eighty twenty leaves you under an hour of hard work. For genuinely time limited athletes, the evidence supports a higher intensity fraction, because intensity is the only variable left to manipulate. If you have four hours, you may well be better served by two hard sessions and two easy ones than by trying to imitate a professional’s ratios at a fifth of the volume.
Sport matters. Non weight bearing sports tolerate far more low intensity volume than running does. Rowers and cyclists can push the easy fraction to ninety five percent because the hours are survivable. A runner adding the same hours accumulates impact and gets injured.
Some middle work is deliberate and correct. Marathon pace for a good marathoner sits close to the middle zone, and you cannot race a distance you have never practised at its intensity. Threshold work has a real place, especially in the specific phase before a race. The Norwegian middle distance approach that has attracted so much attention recently is built around large volumes of controlled work close to the second threshold, carefully monitored with lactate testing. The argument is not that the middle is forbidden. It is that the middle is where the largest number of amateurs accidentally spend most of their week.
The defensible version of the finding is narrower than the slogan and still useful: the distribution most self-coached athletes arrive at by default puts far more work in the middle zone than the distribution of nearly any successful endurance athlete ever measured, and moving work out of that band is one of the highest-return changes available to them.
One more piece of methodological honesty, because it explains why the numbers in chapter 4 do not all agree, and because anyone who starts measuring their own distribution will run straight into it.
There are two reasonable ways to classify training intensity, and they produce meaningfully different answers from identical training.
By session goal. An interval session is a hard session, and all of it counts as hard, including the warm up and the recovery jogs. This is how coaches and athletes naturally think, and it is what Seiler and Kjerland used.
By time in zone. Every minute is classified by the intensity actually being held during it. A hard interval session then turns out to be mostly easy minutes, because warm up, recoveries and cool down dominate the clock.
The same eight week block of skiing can come out as seventy five percent easy by session goal and above ninety percent easy by time in zone. Neither number is wrong. They answer different questions. Time in zone tells you about physiological dose. Session goal tells you about how the week was structured and how much recovery each day demanded.
The practical consequence is that if you compare your own time in zone data from a watch against a published session goal figure, you will conclude you are doing better than you are. Worse, heart rate lags: the first ten minutes of a hard interval can be logged as moderate while you are already working hard, and heart rate drift on a long easy run can log genuinely easy work as moderate. Both errors push your apparent distribution toward the middle.
Count sessions, not minutes, and be strict about the label. For each session in the last two weeks, ask one question: was this unambiguously easy, meaning you could have held a conversation throughout, or was it not? Sessions that were not clearly easy and not clearly a planned hard workout are your middle zone. Most people are surprised by that count.
Everything above is an argument. Below is that argument turned into something you can push on until it gives way.
Set your weekly hours and your distribution. The model responds according to the logic of this article: easy volume builds the capacity that lets you absorb hard work, moderate work costs more than it returns, and time between the thresholds degrades the quality of the sessions around it. Try the presets first, then try to find the settings that break it.
It is worth trying two things in particular. Take a five hour week and move it from the typical amateur preset to polarized, and watch the fatigue meter fall while the stimulus meters hold. Then take a twenty hour week and set easy to forty percent, and watch what happens to a training load that would end a professional’s season.
108 minutes a week between the thresholds
A third of the week sits between the thresholds. It is buying roughly what easy training buys, at more than twice the fatigue cost, and it is blunting the hard days.
One possible week at these settings
Blocks are scaled by duration. Rest days are the empty columns.
A teaching tool, not a prescription and not a performance predictor. It encodes the argument of this article so that you can push on it: easy volume builds the capacity that lets you absorb hard work, moderate work costs more than it returns, and time spent between the thresholds degrades the quality of the sessions around it. The coefficients are the ones listed in the matched fatigue chart above.
Entirely a model. No part of this simulator is fitted to measured performance data, and its outputs are not predictions. It uses the coefficients stated in chapter 5, plus two further assumptions: that recoverable load scales with accumulated easy volume, and that hard session quality degrades as the moderate fraction grows. Both encode the argument of references 2 and 3. Treat the numbers as a way of reasoning, not as a training plan.
The research is descriptive, the model is a toy, and the honest caveats fill an entire chapter. What survives all of that is still a short list of things worth doing.
The test is conversational, not numerical: full sentences, comfortably, for the whole session. It will feel too slow. That feeling is the adjustment working, not a sign that something is wrong. Most people need to drop more pace than they expect, and the first two weeks are psychologically harder than physically.
The point of protecting the easy days is to arrive at the hard ones capable of doing real work. If your intervals are not clearly faster than your tempo pace, you are running two versions of the same moderate session and calling one of them speed work.
Look at the last fourteen days and label every session easy, hard, or neither. The neither pile is the thing to shrink. This is a more reliable audit than any zone chart your watch will draw for you, for the reasons in chapter 11.
If you have room to train more, the first hours you add should be easy ones. They are the cheapest to absorb and they raise the ceiling on how much hard work you can eventually handle. Adding intensity to a small volume base is the fastest available route to being injured.
At four or five hours a week, copying a professional athlete’s percentages is the wrong move. You do not have enough volume for volume to do the work. Two hard sessions in a small week is a defensible plan. What is not defensible is four moderate ones.
Social running is worth doing for reasons that have nothing to do with physiology. Just stop counting it as an easy day, because it almost never is one, and plan the rest of the week knowing where it actually sits.
None of that requires a laboratory, a lactate meter, or a new watch. It requires being honest about which of your sessions were actually easy, which is uncomfortable in a different way than training hard is.
Every link below is a title search rather than a direct identifier, so it resolves to the correct paper even if a volume or page number is misremembered. Where this article used a figure approximately, or assembled a composite, that is stated both here and under the chart itself.
Three of the visuals present measured or closely approximated published values: the zone distribution bars in chapter 4 are the primary one. The rest are explicitly models, drawn to illustrate an argument: the distribution curves are shaped to match published percentages rather than plotted from raw data, and the matched fatigue chart and the week simulator are arithmetic from stated assumptions. Nothing in this article should be read as a measurement unless its source line says so.
Quantifying training intensity distribution in elite endurance athletes: is there evidence for an optimal distribution?
Used for: The 75 / 8 / 17 distribution for junior elite cross-country skiers, and the session-goal method of accounting.
What is best practice for training intensity and duration distribution in endurance athletes?
Used for: The three-zone framework, the case for a polarized distribution, and the moderate intensity rut.
Intervals, thresholds, and long slow distance: the role of intensity and duration in endurance training
Used for: The synthesis of why duration drives aerobic adaptation and intensity drives the top end.
How do endurance runners actually train? Relationship with competition performance
Used for: The 71 / 21 / 8 time-in-zone distribution for sub-elite Spanish distance runners.
Impact of training intensity distribution on performance in endurance athletes
Used for: The controlled comparison in which the group with more easy training improved more.
Training and performance characteristics among Norwegian international rowers 1970-2001
Used for: Thirty years of national team rowing logs, and the shift toward more low intensity volume alongside improving performance.
Polarized training has greater impact on key endurance variables than threshold, high intensity, or high volume training
Used for: The head to head trial of four distributions, in which the polarized group gained the most.
Six weeks of a polarized training-intensity distribution leads to greater physiological and performance adaptations than a threshold model in trained cyclists
Used for: The crossover trial in trained cyclists comparing polarized against threshold training.
Does polarized training improve performance in recreational runners?
Used for: Evidence that the pattern holds in recreational runners, not only in elites.
World-class long-distance running performances are best predicted by volume of easy runs and deliberate practice of short-interval and tempo runs
Used for: The finding that easy running volume is among the strongest predictors of world class distance performance.
The road to gold: training and peaking characteristics in the year prior to a gold medal endurance performance
Used for: The training year preceding a world or Olympic title, including the balance of volume and intensity.
Heart rate and performance parameters in elite cyclists: a longitudinal study
Used for: Season long intensity distribution in professional road cyclists. Figures in this article are approximate readings of that work.
Training and bioenergetic characteristics in elite male and female Kenyan runners
Used for: Training characteristics of elite Kenyan runners, used as context rather than as a plotted figure.
The anaerobic threshold: 50+ years of controversy
Used for: The state of the threshold terminology, and why the naming is genuinely contested.
The maximal metabolic steady state: redefining the gold standard
Used for: The definition of the second threshold as the highest sustainable metabolic steady state.
Polarized vs. threshold training intensity distribution on endurance sport performance: a systematic review and meta-analysis
Used for: The more cautious reading of the evidence, and the basis for the limitations chapter.
The effect of periodization and training intensity distribution on middle- and long-distance running performance: a systematic review
Used for: The review finding that no single distribution has been shown to be optimal for all runners.
The scientific basis for high-intensity interval training: optimising training programmes and maximising performance in highly trained endurance athletes
Used for: Why high intensity work is required for further gains once an athlete is already well trained.
This article is written for general interest and is not medical or coaching advice. If you have a condition that affects exercise tolerance, talk to a clinician before making changes to training.