Cycling training terms,
explained plainly.

TSS (Training Stress Score)

TSS is a single-number summary of a ride's training load, combining how hard it was and how long it lasted. An hour at FTP equals 100 TSS by definition. TSS is the unit that lets you compare a long easy ride to a short hard one and stack both into a weekly total.

TSS = (duration_s × NP × IF) / (FTP × 3600) × 100
    where IF = NP / FTP

TSS was created by Dr. Andrew Coggan (after coach Hunter Allen flagged the need for such a metric) and became the foundational input to his Performance Management Chart (PMC). It lets coaches plan weeks by total load rather than time alone — a 60 TSS day is a 60 TSS day whether it came from 90 minutes of sweet spot or 3 hours of easy endurance.

Typical weekly TSS by training volume: 4–6 hrs/wk → 250–400; 8–10 hrs/wk → 500–700; 12–15 hrs/wk → 800–1100; and well above for WorldTour pros (often 1500+).

CTL (Chronic Training Load)

CTL is the 42-day exponentially-weighted average of your daily TSS. It represents your fitness — the accumulated effect of consistent training. Rising CTL means you're getting fitter; falling CTL means you're detraining. It moves slowly by design.

CTL_today = CTL_yesterday × e^(−1/42) + TSS_today × (1 − e^(−1/42))
    (EWMA with a 42-day time constant)

The 42-day time constant is Coggan's adaptation of the fitness side of Banister's original fitness-fatigue framework (whose fitted fitness time constant was longer — on the order of 50 days). At τ = 42, roughly 95% of CTL's response to a training stimulus comes from the last ~120 days — so your CTL today is a real reflection of your last 3–4 months of training, not just last week.

CTL in isolation isn't a performance predictor — plenty of high-CTL athletes race flat. What matters is CTL combined with ATL (fatigue) and TSB (balance), plus the rate of CTL change. A sustained CTL ramp above ~8/week over multiple weeks is a common overreaching warning sign, though the safe ceiling is highly individual — most riders tolerate roughly 5–8/week depending on age, training history, and current fitness.

ATL (Acute Training Load)

ATL is the 7-day exponentially-weighted average of your daily TSS — the fatigue side of the fitness/fatigue model. ATL responds fast: a hard block drives ATL up quickly, and a few easy days bring it back down. It's why you feel flat two days after a crusher session.

ATL_today = ATL_yesterday × e^(−1/7) + TSS_today × (1 − e^(−1/7))
    (EWMA with a 7-day time constant)

The 7-day constant approximates the time course of acute fatigue — most of the residual fatigue from a training day is gone within a week, though some adaptations and inflammation persist longer.

ATL alone isn't very useful — it's how ATL compares to CTL (= TSB) and how fast ATL is changing that carries information about your readiness. A sudden ATL spike with flat CTL is the classic overload block signature; the taper is engineered to pull ATL down while preserving CTL.

TSB (Training Stress Balance)

TSB is CTL minus ATL — your 'form.' Positive TSB means you're fresher than your training has been lately (rested, primed for a race); negative TSB means you've got fatigue debt (mid-training block, overreached, or undertrained the taper). The zero line isn't magic — context matters.

TSB = CTL − ATL

Rough heuristics: TSB > +10 typically means taper/race ready; 0 to −10 is maintenance; −10 to −30 is heavy training; below −30 is overreach territory (useful briefly, risky long-term). These are directional, not prescriptive.

Peaking for a race is the process of pulling TSB from deeply negative during the build phase up to approximately +15 to +25 on race day — the standard A-race target, though some riders race best only slightly fresh, around +5 to +10 — while preserving as much CTL as possible. That's what tapers do mathematically.

Intensity Factor (IF)

IF is a ride's Normalized Power divided by your FTP. It's a dimensionless 'how hard was this ride relative to my threshold?' number. IF 1.00 is an hour at FTP. IF around 0.90 is a typical sweet-spot effort. IF above 1.05 is only sustainable for short efforts.

IF = NP / FTP

IF is used to categorize a ride's overall intensity relative to threshold (Coggan's reference ranges): below 0.75 is a recovery ride; 0.75–0.85 is endurance; 0.85–0.95 is tempo; 0.95–1.05 is lactate-threshold work and shorter races; above 1.05 is short time-trial or interval efforts that can't be held for an hour.

IF is also the intensity factor in the TSS formula — it's how 'how hard' is encoded into a single number when computing training stress. A 60-minute ride at IF 0.70 generates 49 TSS; at IF 0.85 it generates 72 TSS.

Normalized Power (NP)

NP is a weighted average power that reflects the physiological cost of variable efforts better than simple average power. A surgy criterium and a steady tempo ride can have the same average power but very different NPs — the crit's NP will be higher because short hard efforts cost more than easy pedaling saves.

1. Compute rolling 30-second average power at every second.
2. Raise each value to the 4th power.
3. Average those.
4. Take the 4th root. That is NP.

The 30-second rolling window approximates the body's physiological response time — VO2, heart rate, and ventilation respond to changes in effort with time constants on the order of 30 seconds, so the rolling average tracks the body's lagged response rather than instantaneous power. Raising to the 4th power captures the non-linear cost of harder efforts.

NP is the 'fatigue-weighted' power. For an evenly paced ride, NP ≈ average power. A moderately variable road ride runs roughly 5–15% higher, and genuinely surgy efforts like criteriums or mountain-bike races routinely run 15% or more above average. The variability index (VI = NP / AP) quantifies how surgy the ride was.

Ramp rate (CTL ramp)

Ramp rate is how fast your Chronic Training Load (CTL) is climbing, expressed in CTL points per week. It's the single best summary of how aggressively you're building fitness. A common rule of thumb keeps sustained ramp rate in the range of about 3–8 CTL per week; push it much higher for too long and you tip from productive overload into overreaching and injury risk.

Ramp rate = ΔCTL over a week (CTL_today − CTL_7-days-ago)
    rule-of-thumb sustainable band ≈ 3–8 CTL/week
    (highly individual — depends on age, history, and current CTL)

Because CTL is a 42-day average of daily TSS, raising it requires each week's load to exceed that trailing average — and that surplus is the ramp. A ramp of +5/week means you're adding fitness quickly; flat or negative means maintenance or detraining. Early in a base phase or after a layoff, ramp can safely run high because you're rebuilding lost fitness; near your peak CTL, the same absolute ramp is far harder to sustain.

The 3–8/week band is a heuristic, not a law. Tolerance depends on age, training history, current fitness, life stress, and sleep — a young rider rebuilding may absorb 8+/week for a few weeks, while a masters athlete already at high CTL may find 3–4 is plenty. The right way to use ramp rate is as a guardrail with individual calibration, watched alongside TSB and subjective markers, not as a target to be maximized.

Sustained high ramp is the classic setup for non-functional overreaching: TSB stays deeply negative week after week, form fails to rebound on easy days, and performance stalls or drops. The remedy is periodization — recovery weeks that deliberately let ramp go negative so CTL consolidates — rather than an ever-climbing line. A ramp that only ever goes up is a plan with no off-ramp.

FTP (Functional Threshold Power)

FTP is the highest average power, in watts, that a cyclist can sustain for approximately one hour. It is the foundational reference value used to set every training zone and prescribe every interval workout. FTP is the single most important number in power-based cycling training.

FTP ≈ 95% of 20-minute all-out average power
    (20-min test protocol, per Coggan)

FTP is typically estimated rather than measured — a true 60-minute maximum effort is brutal and rarely done. The standard protocols are the 20-minute test (ride all-out for 20 minutes, take 95% of the average) and the ramp test (ride increasing-intensity ramps to exhaustion, FTP ≈ 75% of the best 1-minute power — roughly 75% of MAP).

More recent approaches — notably Kolie Moore's TTE protocol — argue for a time-to-exhaustion test at a power you can sustain for 30–70 minutes, which more directly measures the underlying physiology. Power-duration modeling (Critical Power, W') is another route that estimates FTP from multiple shorter-duration efforts.

Lactate threshold (LT1 & LT2)

Lactate threshold refers to the exercise intensities at which blood-lactate behavior changes, and there are two of them. LT1, the aerobic threshold, is where lactate first rises above resting levels (~2 mmol/L); LT2, the anaerobic threshold or maximal lactate steady state, is the highest intensity at which lactate production and clearance still balance (~4 mmol/L). Together these two points carve your intensity into the three physiological zones that sit underneath all endurance training.

LT1 (aerobic threshold)          ≈ first rise above baseline, ~2 mmol/L
LT2 (anaerobic threshold / MLSS) ≈ ~4 mmol/L, highest lactate steady state
    (2 and 4 mmol/L are conventional anchors, not universal constants)

The two thresholds define three domains. Below LT1 is easy endurance: lactate stays near baseline and fat dominates as fuel. Between LT1 and LT2 is the moderate-to-heavy band where lactate is elevated but stable. Above LT2, lactate climbs inexorably toward exhaustion. The familiar 2 and 4 mmol/L figures are reference conventions, not physiological laws — an individual's true thresholds can sit meaningfully above or below them, which is exactly why lab testing measures the inflection rather than assuming a number.

Because most riders won't draw blood mid-ride, lactate threshold is often anchored by heart rate — LTHR, the lactate-threshold heart rate, commonly estimated from the average HR of a hard 20–30 minute time trial. Joe Friel's heart-rate zones are built as percentages of LTHR (%LTHR), which is a different anchor from %HRmax. LTHR is more stable day to day than power output, but it still drifts with heat, dehydration, and fatigue.

Lactate threshold is not the same construct as FTP, even though the two usually sit close. FTP is defined operationally (the power you can sustain for about an hour); LT2 is defined by lactate kinetics, and maximal-lactate-steady-state testing is its gold standard. The practical upshot is that FTP is the convenient field proxy for the threshold that actually matters physiologically — good enough to set zones from, without a blood-lactate meter.

VO2max

VO2max is the maximum rate at which your body can take in and use oxygen during exhaustive exercise, measured in millilitres of oxygen per kilogram of body mass per minute (ml/kg/min). It is the ceiling on aerobic power production and one of the strongest predictors of endurance performance.

For trained amateur cyclists, VO2max typically ranges from 55 to 70 ml/kg/min; elite road pros are often 70–85+, with the highest outliers above 90. Genetics set a plausible range; training determines where within that range you operate.

VO2max is trainable primarily with short, hard intervals near maximum sustainable aerobic output — 3–8 minute efforts at 105–120% of FTP are the classical prescription. 30/30s and 40/20s (short bursts with incomplete recovery) work a related but partially different mechanism.

Critical Power (CP)

Critical Power is the highest power a cyclist can sustain in a metabolic steady state — the asymptote the power-duration curve flattens toward as effort duration grows. At or below CP the effort is theoretically sustainable for a long time; above it you're on a countdown, draining your finite work capacity (W'). It's the physiologically-grounded cousin of FTP, derived from a mathematical model of your power-duration curve rather than a single all-out test.

P(t) = W' / t + CP
    (hyperbolic power-duration model; Monod & Scherrer 1965)
    equivalently: t_to_exhaustion = W' / (P − CP)

Monod and Scherrer introduced the critical-power concept in 1965 for isolated muscle groups; later work extended it to whole-body cycling, and modern researchers — notably Jones, Vanhatalo, and Poole — formalized the two-parameter model now in common use. The model has exactly two parameters: CP, the aerobic asymptote in watts, and W', the finite work capacity available above it in kilojoules. Both are estimated by having a rider perform two to five maximal efforts of different durations (classically in the ~2–15 minute range) and fitting them to the hyperbola.

Physiologically, CP marks the boundary between the heavy and severe exercise-intensity domains: it is the highest intensity at which oxygen uptake and blood lactate can still stabilize. Ride above CP and VO2 drives toward maximum while lactate accumulates relentlessly until you're forced to stop. That makes CP a more principled threshold than FTP's operational 'power you can hold for an hour' heuristic, because it's defined by the underlying steady-state physiology rather than a clock.

CP and FTP are close but not identical. Because CP is a steady-state asymptote, it typically corresponds to power a rider can hold for roughly 40–60 minutes, so it often lands a few percent above a 20-minute-test FTP estimate — how far depends on the test durations used and the size of the rider's W'. The classic estimation mistakes are fitting the model to efforts that are all too short (which inflates CP), pairing durations that are too close together, or not going genuinely maximal on each test effort.

W' (anaerobic work capacity)

W' (pronounced 'W-prime') is the fixed amount of work, measured in kilojoules, that you can perform above Critical Power before exhaustion. Think of it as a battery: every second spent above CP drains it, and time spent below CP recharges it. When W' reaches zero you're done — you have to drop below CP or stop.

W' = (P − CP) × t_to_exhaustion
    (constant effort above CP; W' in joules, P and CP in watts)
    typical range ≈ 10–30 kJ

W' is the second parameter of the critical-power model, alongside CP. It represents a finite, largely anaerobic reserve — glycogen-derived substrate, phosphocreatine, and tolerance for the metabolites that accumulate above CP. Typical values sit around 15–25 kJ (with a broader range roughly 10–30+). Explosive pursuit and sprint-type riders carry a large W', while pure diesel time-triallists tend to have a small one; genetics and training both shape it.

Skiba and colleagues extended the static parameter into a dynamic, real-time model — W'balance — that tracks depletion during hard efforts and reconstitution during easier ones. Reconstitution is neither instantaneous nor linear: it follows a recovery time constant that depends on how far below CP you drop, so the deeper the recovery, the faster W' comes back. This is precisely why interval sessions live or die on the quality of their recovery valves.

W' explains why you can complete, say, 5×3 minutes above CP even though no single 15-minute effort at that power is possible — the recovery windows partly refill the battery between reps. Make those recoveries too short or too intense (still near CP) and W' never recharges, so the final reps collapse. Good coaching back-solves interval targets from W' so the last rep is still completable rather than a guaranteed blow-up.

Power curve (mean maximal power)

Your power curve — also called the mean maximal power (MMP) curve — plots the best average power you've produced for every duration, from one second out to several hours. Read left to right it falls steeply through the sprint durations and flattens toward your threshold. Its shape is a fingerprint of what kind of rider you are, and it's the raw material for estimating FTP, Critical Power, and W' without a formal test.

MMP(t) = max rolling-average power sustained for duration t
    (taken across your entire ride history)

The curve is built by sliding a window of each length across every ride and keeping the single best value at that duration. The result is an envelope, not any one ride: your best 5-second sprint and your best 3-hour effort almost certainly came from different days. Handling gaps and dropouts correctly (regridding the stream to a steady sample rate) matters, because a naive window can understate power on rides with pauses or missing data.

The shape reveals rider type. A pure sprinter shows enormous short-duration power that drops off a cliff; a time-triallist posts modest sprint numbers but a nearly flat curve out to an hour. The steepness between roughly 1 and 20 minutes reflects anaerobic capacity (a large W') versus aerobic endurance (a high CP), and comparing your curve against normative power-profile charts locates your strengths and weaknesses.

Because it captures maximal efforts across durations, the curve is what lets models estimate FTP (e.g. ~95% of best 20-minute power, or via the CP model) and fit CP and W' — no dedicated test day required. The important caveat: the curve is only as good as the efforts in it. If you've never gone genuinely deep for 5 minutes, the model has no anchor there and will underestimate that part of your ability.

Durability (fatigue resistance)

Durability is your resistance to the fatigue-induced decline in performance that shows up late in long rides — the erosion of power, efficiency, and even your thresholds after several hours or a few thousand kilojoules of work. Two riders with identical fresh FTPs can differ enormously in what they can still produce at hour four. It's increasingly described as endurance cycling's 'fourth dimension,' alongside VO2max, threshold, and efficiency.

Durability ≈ decline in a performance marker after accumulated work
    e.g. Δ FTP or Δ 5-min power, measured fresh vs. after ~2000–3000 kJ
    (no single standard metric yet — an emerging research area)

A growing body of work, notably by Maunder and colleagues, has formalized durability as the deterioration of physiological and performance characteristics over prolonged exercise. After heavy accumulated work, power at a given effort falls, thresholds shift downward, and gross efficiency drops — so the traditional fresh-state lab numbers (VO2max, FTP) stop describing what a rider can actually do deep into a race. It helps explain why a pro who can't out-sprint an amateur when fresh will ride that same amateur off the wheel after four hours.

Durability is driven largely by substrate — progressive glycogen depletion and the shift toward fat oxidation — compounded by muscular fatigue and thermoregulatory strain. You build it with long rides that accumulate time and kilojoules, with carefully-dosed low-carbohydrate-availability endurance work, and increasingly by placing key intervals late in a long ride so the hard efforts land on already-tired legs rather than fresh ones.

It matters most for long-event riders — gran fondos, marathon MTB, long road races, ultra-endurance — where the decisive moments come after hours of work, and much less for a 40-minute criterium. Measuring it is still unsettled: there's no single agreed-upon metric, so most practical approaches compare a performance marker fresh against the same marker after a fixed amount of accumulated work.

Aerobic decoupling (cardiac drift)

Aerobic decoupling — also called cardiovascular or cardiac drift — is the tendency for your heart rate to climb over the course of a long, steady effort even though your power (or pace) stays constant. Quantified as the Pw:HR ratio, it compares the power-to-heart-rate relationship in the first half of a ride against the second. A drift above about 5% on an aerobic ride flags that the effort taxed your aerobic base or durability more than it would a well-adapted rider.

Pw:HR decoupling = (ratio_first_half − ratio_second_half) / ratio_first_half × 100%
    where ratio = average power / average heart rate for that half
    < ~5% = 'coupled' (well-adapted); > ~5% = decoupled

The mechanism is straightforward. Over a long steady effort, heart rate rises for the same power because of cardiovascular drift: plasma volume falls as you sweat, core temperature climbs, stroke volume drops so heart rate compensates, and accumulating muscle fatigue recruits less-efficient fibers. The less aerobically fit or durable you are for that duration and intensity, the sooner and steeper the drift.

Joe Friel popularized the ~5% threshold as a practical marker of aerobic fitness for a given intensity. To measure it, ride a steady aerobic effort (endurance to low tempo, no big surges, minimal coasting) of an hour or more, split it in half, and compute the change in Pw:HR. Under 5% means you're well 'coupled' — aerobically adapted to hold that effort. It has to be measured on genuinely steady rides; interval or hilly rides full of surges and coasting make the ratio meaningless.

Rising decoupling as ride duration grows is a durability signal: your late-ride efficiency is fading. Large drift on rides that should be easy suggests the aerobic base isn't there yet — or that you rode too hard, or it was hot, or you were dehydrated or under-fueled. Because those confounders all move heart rate independently of effort, a single high reading isn't a verdict; the trend across similar rides is the real signal.

Zone 2 (aerobic endurance)

Zone 2 is low-intensity aerobic endurance riding — in Coggan's power model, roughly 56–75% of FTP. Physiologically it sits at or below the first lactate/ventilatory threshold: an effort you can hold for hours and talk through in full sentences. It's the unglamorous zone where the bulk of most riders' base volume should live, because it drives the aerobic adaptations that raise your ceiling at very little fatigue cost.

Coggan Zone 2 ≈ 56–75% of FTP
    physiologically: at or below LT1 / VT1 (first threshold)
    'conversational' — full sentences, nasal breathing

'Zone 2' means different things in different systems, which trips people up. In Coggan's 7-zone power model it's the endurance zone (56–75% FTP). In Seiler's 3-zone physiological model the same easy riding is called Zone 1 (low-intensity, below VT1) — and confusingly, that model's 'Zone 2' is the moderate grey zone instead. This page uses the common cyclist meaning: easy aerobic endurance below the first threshold. What matters is the physiology, not the label.

Sustained low-intensity work is the primary driver of mitochondrial biogenesis and capillary density, improves the muscle's capacity to oxidize fat and spare glycogen, and expands plasma and blood volume and stroke volume — the aerobic base everything else is built on. It's trainable in high volume precisely because it's low-fatigue: you can accumulate many hours a week without needing days to recover from any single ride.

Zone 2 surged in popularity after Iñigo San Millán's work with WorldTour riders and his interviews describing its role in mitochondrial function and fat oxidation. The most common amateur mistake is riding it too hard — letting 'easy' creep up past 80% of FTP into the grey zone, which pays threshold-like fatigue for sub-threshold benefit and quietly undermines the next hard session. The discipline to keep easy genuinely easy is the entire point of the zone.

Tempo (Zone 3)

Tempo is moderate-intensity riding — Coggan Zone 3, roughly 76–90% of FTP. It's 'sustainable but working': harder than conversational endurance, easier than threshold, the pace of a strong steady group ride or a long climb ridden briskly. It sits in the physiological middle ground, which makes it both genuinely useful and, if overused, the classic grey-zone trap.

Coggan Zone 3 (Tempo) ≈ 76–90% of FTP
    RPE ~5–6/10 · breathing rhythmic but controlled
    upper tempo blends into sweet spot (~88–94% FTP)

Tempo lives between the aerobic threshold and threshold proper — above LT1, below LT2. Ridden in volume it extends aerobic endurance, builds muscular endurance and fatigue resistance, and teaches sustained sub-threshold pacing, which makes it a staple of high-volume base blocks and long climbing days. The top of tempo shades into sweet spot (~88–94% FTP), where the adaptation-per-minute improves.

Tempo is also exactly the 'moderate middle' that polarized-training advocates like Seiler argue amateurs overdose on: it accumulates real fatigue and sympathetic stress while adding only a modest aerobic stimulus beyond what easier volume already delivers. The failure mode is unintentional tempo — endurance rides that drift up to 80% FTP because it feels productive — which leaves you too tired to hit the genuinely hard sessions that actually drive adaptation.

Used deliberately, tempo earns its place: muscular-endurance work, big aerobic days, and a bridge intensity in base. As a default, though, it's the textbook definition of junk intensity. The whole distinction is dose and intent — a planned 2×20 tempo block is training; letting every ride settle into tempo because it's comfortable is the trap.

Sweet Spot training

Sweet Spot is sustained effort at roughly 88–94% of FTP — the upper end of tempo running into the lower edge of threshold. It is hard enough to drive most of the aerobic adaptations of true threshold work, but easy enough that you can accumulate far more weekly minutes at high intensity without the recovery cost. That favorable adaptation-to-fatigue ratio is the 'sweet spot.'

Sweet Spot ≈ 88–94% of FTP (Coggan power zones: top of Zone 3 / bottom of Zone 4)
    ≈ IF 0.88–0.94 · RPE ~7/10 · 'comfortably hard,' conversational only in short phrases

The concept grew out of Andrew Coggan and Hunter Allen's power-zone framework, formalized in Training and Racing with a Power Meter (first edition 2006). The label itself is usually credited to Frank Overton of FasCat Coaching around 2004–05: after Coggan presented his power-based impulse-response performance model at USA Cycling's Coaches Summit in Colorado Springs, coaches working to validate it noticed the most productive band of intensity sat just below threshold, where adaptation stayed high but fatigue dropped off sharply. FasCat then built much of its commercial coaching around dedicated sweet-spot plans, which is why the term is now near-universal in amateur cycling.

The physiological rationale is sub-threshold aerobic stimulus at a sustainable fatigue cost. Work in the high-tempo-to-low-threshold band sits below or right at the maximal lactate steady state, so blood lactate stabilizes rather than runs away. That lets you hold the effort for 10–30 minutes at a time and stack several intervals in a session — driving the central and peripheral aerobic adaptations (stroke volume, capillary density, mitochondrial volume and oxidative-enzyme activity) that underpin a higher FTP. You give up a little of the top-end stimulus of riding at 100% FTP, but you gain a lot back in total trainable volume and faster between-session recovery — the trade-off that makes sweet spot attractive for amateurs on 6–12 hours a week who can't absorb a steady diet of full threshold or VO2max work.

Common sweet-spot workout shapes: 2×20 min, 3×15 min, 3–4×10–12 min, or a 'sweet spot pyramid' (e.g. 10/15/20/15/10 min), all at 88–94% FTP with 5–10 minutes of easy spinning between reps. A practical session ceiling is roughly 40–70 minutes of total time-in-zone; weekly sweet-spot volume tends to top out around 90–150 minutes (spread over 2–3 sessions) before cumulative load starts to degrade execution and the work quietly slides into the very junk-intensity trap it's meant to avoid.

Where it fits in a periodized year: sweet spot is most valuable in the base-to-build transition and the early specific-prep block, where the goal is to raise sustainable power efficiently before layering on race-specific threshold and VO2max work. It is the workhorse for time-crunched indoor training and for riders rebuilding after a layoff. It is a poor primary stimulus during a peak or taper, when the plan needs sharpness from short, very-high-intensity efforts rather than more sub-threshold volume.

The classic mistakes are two sides of one coin. First, turning every ride into sweet spot: doing it 4–6 days a week leaves you chronically under-recovered, erodes the easy aerobic base that low-intensity volume builds, and produces a flat, monotone fitness that plateaus by year two. Second, the 'grey zone' trap — letting endurance rides drift up to 80–87% FTP without it being a structured sweet-spot effort, so you pay near-sweet-spot fatigue for sub-sweet-spot adaptation. Sweet spot earns its place as a deliberate, dosed block of work, not as the default intensity for everything. Critics — including Seiler's polarized camp — argue amateurs over-prescribe it precisely because it feels productive; the honest framing is that it is a high-value tool used sparingly, not a complete training philosophy.

Polarized training

Polarized training is an intensity distribution where roughly 80% of training sessions (or time) sit at low intensity — comfortably below the first lactate/ventilatory threshold — and about 20% are genuinely hard, at or above the second threshold (threshold-to-VO2max work), with deliberately little in the moderate middle. The model was characterized and popularized by sport scientist Dr. Stephen Seiler, now at the University of Agder in Norway.

Three-zone model (per Seiler):
    Zone 1 / LIT  — below VT1, < ~2 mmol/L lactate  →  ~75–80% of training
    Zone 2 / MIT  — between VT1 and VT2, ~2–4 mmol/L  →  ~5% or less ('the grey zone')
    Zone 3 / HIT  — above VT2, > ~4 mmol/L           →  ~15–20% of training

The pattern was crystallized in Seiler and Kjerland's 2006 study of nationally competitive junior cross-country skiers, set against a wider body of observational work on elite endurance athletes across cycling, rowing, skiing, and running. The recurring finding was a roughly 80/20 split of low to high intensity — strikingly consistent across sports and despite athletes often feeling they 'should' train more in the middle. It is critical that polarized uses the three-zone physiological model anchored to the two lactate/ventilatory thresholds (VT1 and VT2), not the 5–7 power zones cyclists usually count in; an effort at 88–94% FTP (sweet spot) is moderate Zone 2 in this model, not part of the hard 20%.

The physiological case for skipping the middle is that moderate, sustained 'grey zone' work — above the aerobic threshold but below the anaerobic one — accumulates meaningful fatigue and sympathetic stress while delivering a relatively modest additional aerobic stimulus over what easy volume already provides. Low-intensity volume builds the aerobic base (mitochondrial and capillary adaptations, fat oxidation, blood volume) cheaply, and a small dose of genuinely hard work supplies the high-end VO2max and lactate-tolerance stimulus. The middle is criticized as buying neither efficiently — hence Seiler's 'no man's land' framing.

Polarized is one of three commonly described distributions. Pyramidal training has the most volume easy, a moderate amount at threshold, and the least very-hard — a descending pyramid with a meaningful Zone 2 block; it is the distribution many elite athletes actually log in high-volume base periods. Threshold-focused training concentrates work at or near the lactate threshold (the sweet-spot/threshold-heavy model favored by many time-crunched amateurs). Polarized hollows out that middle deliberately. The honest reading of the literature is that pyramidal and polarized both beat threshold-heavy approaches in most trained populations, and that the two are close enough that phase, volume, and the individual matter more than dogma.

Distribution should shift with the training phase. In a high-volume base period, a pyramidal lean (more sustainable Zone 2 / sweet spot) is common and effective for building durability and raising the aerobic ceiling efficiently. As the season moves toward specific preparation and peaking, training tends to polarize: the hard 20% becomes more VO2max-specific while easy volume protects recovery. A frequently cited supporting trial is Stöggl and Sperlich (2014), in which a 9-week polarized block (an LIT–MIT–HIT split near 68–6–26) produced larger gains in VO2peak (about +11.7%) and lactate-threshold power than threshold-focused, high-intensity, or high-volume blocks — though it is one study on a modest sample, so treat the magnitude as illustrative, not a guarantee.

For time-crunched amateurs, strict 80/20 is hard to execute well. The 80% refers to the bulk of training, so on 5–7 hours a week the easy portion can feel almost too easy and most riders unconsciously push it into the grey zone — which collapses the polarization the model depends on. Practical execution: keep easy rides genuinely easy (below ~75% FTP / conversational), make the hard sessions count (true threshold or VO2max intervals, 2–3 a week at most), and resist filling the calendar with moderate 'tempo for tempo's sake' rides. The most common misapplications are calling a sweet-spot-heavy week 'polarized' because it has one hard day, counting by session rather than by time on lopsided week structures, and going so polarized on low volume that there isn't enough total aerobic stimulus to drive base fitness at all.

Periodization

Periodization is the deliberate structuring of training into phases that progress toward a goal — classically base, build, peak/specific-prep, and taper — organized across nested time scales. The premise is that you can't hold peak fitness or train every quality at once, so you sequence blocks that develop general fitness first and race-specific sharpness last. It's the scaffolding that turns a pile of workouts into a plan.

Nested time scales:
    Macrocycle  — a whole season / to a goal event (months)
    Mesocycle   — a training block or phase (≈ 3–6 weeks)
    Microcycle  — typically one week

Modern periodization is usually traced to Leo Matveyev's systematization of Soviet training theory in the 1960s. The traditional model moves from high-volume, low-intensity base work (building the aerobic engine) through a build phase that layers in threshold and VO2max intensity, into a specific/peak phase that sharpens race-relevant abilities, and finally a taper that sheds fatigue for the event. Volume and intensity trade off across the arc — lots of easy volume early, less-but-harder work later.

Traditional (linear) periodization develops several qualities concurrently with a gradually shifting emphasis. Block periodization, associated with Issurin, instead concentrates a small number of training targets into focused blocks, exploiting the residual training effects of previously-developed qualities so one is trained intensively while others are merely maintained. Block approaches suit time-crunched and already-trained athletes; the honest reading is that both work, and the choice depends on the athlete, the calendar, and the event.

The nested cycles carry real weight: microcycles (usually a week) alternate load and recovery days; mesocycles (a few weeks) group microcycles into a phase, typically closing with a recovery week to consolidate; the macrocycle spans the season to the A-race. The common mistakes are skipping the base entirely, never de-loading (no recovery weeks inside a mesocycle), and peaking too early or with no taper at all. Periodization is precisely what stops a year of undifferentiated sweet spot from plateauing.

Taper

A taper is the progressive reduction in training load in the days and weeks before a goal event, designed to shed accumulated fatigue while holding on to the fitness you built. The winning combination, well established in the literature, is to cut volume substantially — roughly 40–60% — while keeping intensity and frequency high, over about one to three weeks. Done right, it converts hidden fitness into fresh legs, typically worth a couple of percent of performance.

Taper recipe (per Bosquet et al. meta-analysis):
    Volume    — ↓ ~40–60%
    Intensity — hold (keep the hard efforts)
    Frequency — hold (don't stop riding)
    Duration  — ~1–3 weeks; progressive (exponential) decay works best
    Result    — ~2–3% mean performance gain

Bosquet and colleagues' meta-analysis of taper studies found the most effective approach is a progressive (exponential) reduction of training volume of roughly 40–60%, maintaining intensity and frequency, over about two weeks, yielding on the order of a 2–3% improvement in performance — a large margin at the sharp end of a race. The critical, counter-intuitive point is that you cut volume, not intensity. Dropping the hard efforts detrains you; dropping the volume is what actually removes the fatigue.

The mechanism lives in the fitness/fatigue framework, where performance is fitness (CTL) minus fatigue (ATL). Because fatigue decays faster than fitness (a 7-day versus a 42-day time constant), reducing load lets ATL fall quickly while CTL barely moves — so form (TSB) rises. The taper is engineered to land TSB in a fresh-but-not-flat window on race day, commonly around +15 to +25 for an A-race, while preserving as much CTL as possible.

Taper length scales with the block that preceded it and the event itself — a big endurance build needs longer than a weekend criterium. The classic errors are tapering too hard (going fully easy, losing sharpness and feeling sluggish), tapering too long (fitness starts to decay), and doing nothing at all (racing buried in fatigue). Keeping short, sharp efforts — openers — through the taper preserves neuromuscular feel without adding meaningful fatigue.

Relative Effort (Strava)

Relative Effort is Strava's training-load score for a single activity — a heart-rate-based number that estimates how hard a session taxed you from how much time you spent in each HR zone. It's the rebranded successor to Strava's old 'Suffer Score,' and it rolls up into Strava's Fitness & Freshness chart (their take on CTL/ATL). Because it's built from heart rate, it is not directly comparable to power-based TSS.

Relative Effort ≈ time-in-HR-zone weighted by zone intensity (a TRIMP-style score)
    higher HR zones are weighted more heavily
    HR-based — NOT interchangeable with power-based TSS

Relative Effort is a TRIMP-style metric (training impulse): it weights the time you spend in each heart-rate zone, counting minutes in higher zones more heavily, to produce one load number per activity. Strava introduced it as the more approachable, heart-rate-driven replacement for the Suffer Score, and it feeds their Fitness (a chronic average) and Freshness (form) curves — Strava's own rendering of the CTL/ATL/TSB model.

The reason it isn't comparable to TSS is that they measure different things. TSS is computed from power relative to FTP; Relative Effort is computed from heart rate relative to your HR zones. Heart rate lags effort, drifts with heat, fatigue, caffeine, and sleep, and can't see very short high-power surges — so the same ride can produce quite different Relative Effort and TSS values. Summing the two, or comparing your Relative Effort to a friend's TSS target, is an apples-to-oranges error: each is internally consistent, but they're different currencies.

Relative Effort earns its keep when you don't have a power meter — it's a legitimate, heart-rate-grounded way to quantify and trend training load, and for many riders it's the only load metric they have. Its limits are heart rate's limits: it under-rates short, punchy efforts and gets thrown off by anything that moves HR independently of effort. As a within-athlete trend it's genuinely useful; as a cross-athlete or cross-metric absolute, it isn't.