Sweet Spot Training Method Explained

Your weekly hours9h

Fits your week. Coherent from 4 to 12 hours a week.

Where the time goes

Compare with

By time. The 26% between the two thresholds is minutes in zone; the 40–50% figure usually quoted counts sessions, not minutes.

A representative week

Show the week for

Across all three sports.

Mon

Swim 45 minAerobic and technique. Well under CSS — this is not a quality day.

Tue

Bike 95 min15 min warm-up, 3 × 20 min at 213–235 W (85–94% FTP) with 5 min easy between, 10 min cool-down. 60 min in zone. Quality day 1 of 3.

Wed

Run 50 minEasy aerobic, conversational throughout. No strides needed.

Swim 45 minAerobic set, first thing if it shares the day with the run.

Thu

Run 55 min12 min warm-up, 2 × 15 min at 4:30–4:40/km (threshold pace + 10–20 s/km) with 3 min jog between, 10 min easy. Marathon-to-half-marathon effort. Quality day 2 of 3.

Fri

Recovery spin 45 minUnder 55% FTP, or take the day off. Sitting between two quality days, this is the ride you do not turn into a third.

Sat

Bike 3 hEndurance, with 2 × 20 min at 213–225 W (85–90% FTP) in the back half. Quality day 3 of 3 — drop the intervals in a recovery week or if Thursday's run went badly. Nothing off the bike afterwards.

Sun

Run 80 minLong run, easy. Nothing steady-state in it — the week already has three quality days.

What Tuesday looks like

Bike, 95 min. Warm up 15 min easy, then 3 × 20 min at 213–235 W (85–94% FTP) with 5 min easy between, then 10 min cool-down — 60 min in zone. For a 250 W FTP that is 85% at the start of a block and 94% only once the athlete is holding the full 60 min in zone comfortably. Heart rate anchor if the power meter is out: 90–99% of threshold HR. RPE 6–7 — short sentences only, and you should finish believing you could have done a fourth. The 5 min between reps is part of the session, not padding: it is what makes three of these a week survivable. Next week the interval gets longer, not harder. Nothing follows the ride — no brick, no run off the bike — and the watts came from your FTP, not from any race's power.

Signature sessions

The classic 3 × 20: 3 × 20 min at 84–94% FTP (210–235 W at FTP 250), 5–8 min easy between. FasCat names this the staple for Olympic-distance triathletes; Form starts a block at the 85–90% end of it.

The practical floor: 3 × 5 min at 88–93% FTP, 5 min easy. Fifteen minutes in zone is the least that counts as a session.

Two documented FasCat progressions, not one chain. Progression A: 3 × 10 (30 min in zone), 4 × 9 (36), 3 × 15 (45). Progression B, their published base mesocycle: week 1 2 × 10 (20), week 2 2 × 15 (30), week 3 3 × 12 (36), week 4 2 × 20 (40), week 5 onward 60 min and up. Extend the interval, hold the watts.

Long-ride sweet spot: 2–3 × 20 min inside a 3–4 h endurance ride, placed in the back half so the work happens on partly depleted glycogen.

Over-unders, when you want the bridge toward threshold. TrainerRoad's canonical shape is 3 × 12 min alternating 2 min at 95% FTP and 2 min at 105% — equal halves. The sweet-spot variant is 4 × 12 min with 2 min valleys at 80% FTP and 2 min peaks at 95–100%. Both are harder than a plain sweet spot set; they count against the weekly budget as a quality day, not as an extra one.

Run steady state (the analogue, not the same thing): 2 × 15 min or 1 × 30–40 min at threshold pace plus 5–20 s/km — 4:25–4:40/km off a 4:20 threshold, closer to 4:25 for half-marathon effort and 4:35–4:40 for marathon effort. Roughly 2–3 mmol, 80–88% HRmax. Maximum two of these a week, and only one in a week that already carries two bike sessions.

Swim: CSS from a maximal 400 and 200 in one session with about 5 min recovery, CSS pace in s/100 m = (T400 − T200) / 2. The analogue is CSS pace plus 3–8 s/100 m (1:43–1:48 off a 1:40 CSS), as 10–20 × 100, 5 × 400, or 400/300/200 descending sets on 10–20 s rest.

Session ceiling: 60–120 min of total in-zone time for a well-conditioned rider. FasCat cite 2–3 h for the very well conditioned; the 6 h figure that circulates is an anecdote, not a target.

Who it suits

Time-limited cyclists and short-course triathletes training 4–12 h/week with at least 3 h on the bike, especially indoors where holding a steady power number is easy. It suits athletes whose racing happens to land near this intensity — sprint bike legs are conventionally 85–95% FTP and Olympic 80–90% — though the sessions are still written off FTP, not off a race target. It suits athletes returning to structure who need a repeatable, tolerable stimulus rather than a heroic one, athletes practising a TT position under load, and riders with high aerobic durability whose LT1 sits high enough that 88% FTP is close to an endurance effort for them. It is also the method that survives contact with a real calendar — every session has a defined target, missing one costs you a known amount, and the load steps up in 3-weeks-on, 1-week-easy blocks.

Not for you if

Ironman and long-course athletes in the specific phase. Ironman bike power is conventionally put at 0.65–0.75 IF for an age-grouper, which is tempo and Z2, so sweet spot is not race-specific there and it displaces the low-intensity durability work that decides the bike split. The same applies to a 70.3 athlete riding over 3 hours, who sits at the bottom of the conventional 0.78–0.85 IF band for that distance or below it. Both bands are published convention rather than measured values; use them to decide that this band is too high, not as a prescription.

Athletes training more than about 12 h/week — the opportunity cost of not riding easy is real.

Athletes with an unreliable or stale FTP. Wong et al. (2022) had 13 cyclists last 33.7 ± 7.6 min at FTP with end-test lactate 6.7 ± 2.1 mM, so FTP is not a maximal metabolic steady state to begin with. If a number is set 5–10% high, "90% of FTP" sits at or above the athlete's true steady state and every sweet spot session is quietly a threshold session. Prescribing off an untested or algorithm-estimated FTP is the single most common way this method becomes something else.

Athletes carrying high non-training stress, poor sleep, or under-fuelling. Sweet spot is glycogen-expensive and there is no easy-day relief valve if it appears four or five times a week.

Anyone using it as their only intensity. FasCat are explicit that it cannot substitute for VO2max, anaerobic or neuromuscular work.

Athletes new to structured riding with a thin aerobic base. Building the base at 90% FTP works for a while and then stalls.

In a triathlon week

The bike transfers unchanged. Dose it against race distance: sprint, sweet spot as the primary bike intensity through base and race prep; Olympic, sweet spot up to threshold with the 3 × 20 as the staple; 70.3, sweet spot for athletes going under 3 h on the bike, lower sweet spot or tempo for those over 3 h; Ironman, shift progressively to tempo and Z2 as the specific phase approaches, because sweet spot is not race-specific there. Note what that dosing is not: the intervals are still written as a percentage of FTP, and the race distance only decides how much of them you get and when they stop. The constraint is never the ride itself — it is what comes after. A 3 × 20 compromises the next day's run quality, so it gets placed against the run schedule, not the bike schedule.

The run transfers only by analogy. There is no legitimate run "sweet spot percentage". The nearest physiological equivalent is steady-state running at marathon-to-half-marathon effort: roughly 2–3 mmol, 80–88% HRmax, or threshold pace plus 5–20 s/km (closer to +5 for half-marathon effort, +15–20 for marathon effort). Treat that as an approximation to calibrate, not a formula. Running carries eccentric and impact load the bike does not, so 1–2 steady-state runs a week is the ceiling.

The swim transfers by analogy too, anchored to Critical Swim Speed. CSS pace in seconds per 100 m is (T400 − T200) / 2 from a maximal 400 and 200 in one session. The analogue is CSS pace plus 3–8 s/100 m as long aerobic sets. Most age-group swim training already lives here, and the swim's real limiter is usually technique rather than fitness, so time in zone buys little if the stroke is inefficient.

Whole-athlete rule that overrides everything above: the 2–3 sweet spot sessions per week is a total across all three sports, not per sport, and it is the same 2–3 quality days the rest of Form counts — nothing gets added beside it. And if you are borrowing a cycling plan's volume tier for a triathlon week, drop a tier: a cycling plan's hours assume the bike is all the training. Purpose-built triathlon tiers already count all three sports and should be taken at face value.

What the evidence says

WELL SUPPORTED. Work in this band raises FTP and lactate threshold in the short term in under-trained and moderately trained riders; this is not seriously disputed. It is time-efficient per hour relative to low-intensity work. And it is more repeatable than threshold or VO2max work, which is the actual practical selling point. Structured, target-defined, progressively loaded training also beats unstructured riding for time-limited athletes, largely because it optimises adherence.

CONTESTED. There is no good randomised trial of "sweet spot" as a named protocol against polarised or pyramidal training in trained cyclists. The closest evidence is indirect. Stöggl and Sperlich (2014) took 48 well-trained endurance athletes (VO2peak 62.6 ± 7.1) through 9 weeks in four arms — high-volume 83/16/1, threshold 46/54/0, HIIT 43/0/57, polarised 68/6/26 — and polarised produced the greatest improvement in most key endurance variables, while the threshold arm showed only slight gains in work economy. Against that, the 2024 meta-analysis (Oliveira, Boppre and Fonseca) found polarised training's VO2peak advantage was small overall (SMD 0.24, 95% CI 0.01–0.48, p=0.04, n=284), present only in studies under 12 weeks (SMD 0.40) and in highly trained athletes (SMD 0.46), absent at 12 weeks or longer (SMD 0.04, p=0.83) and in merely trained athletes (SMD 0.08, p=0.62), with no difference at all on time-trial performance (SMD −0.01, p=0.92) and no significant difference on any outcome for polarised versus threshold specifically. Filipas et al. (2022) ran 60 well-trained male runners through 16 weeks in four arms — pyramidal, polarised, pyramidal-then-polarised, and polarised-then-pyramidal. Pure pyramidal was not inferior to pure polarised, but switching from pyramidal into polarised after 8 weeks beat both, with the largest gains in every variable (about 3.0% in relative VO2peak) and at least 5 s more improvement in the 5 km time trial. That is close to FasCat's own sequencing: build the base in the middle zone, polarise as racing approaches. Read together, the middle zone is not poison. It is just not magic.

THE FATIGUE-ACCUMULATION CRITIQUE — the strongest objection, and it deserves to be stated fairly. Seiler, Haugen and Kuffel (2007) measured heart-rate-variability recovery in 9 highly trained runners (VO2max 72 ± 5, 14 ± 3 h/week). Running 60 or 120 min below the first ventilatory threshold returned HRV to baseline within 5–10 min. A 60-min session containing 30 min between the two thresholds delayed HRV recovery to about 30 min — essentially identical to 6 × 3 min at 96% VO2max. The implication is precise: the middle zone costs you autonomically about what a VO2max session costs, without delivering the VO2max stimulus. Two caveats matter. The threshold session was only 30 min in zone at 2.7 ± 0.4 mM, which is low sweet spot or high tempo rather than a 3 × 20 at 92% FTP; and the subjects were highly trained runners, so carrying the number across to a 6 h/week cyclist is extrapolation. It is a strong argument, not a settled verdict.

SECONDARY CRITIQUE, physiologically reasonable but less directly evidenced: fat oxidation peaks around 60–65% of VO2max, roughly 65–75% of FTP, and falls away above it, so a base built at 88–94% FTP under-develops fat oxidation and durability — the qualities that decide long-course racing.

LORE, NOT EVIDENCE. "Five hours of sweet spot equals 10–20 hours of traditional base" is marketing copy. "Most bang for your buck" as a universal claim. The published race intensity factors for 70.3 and Ironman, which are convention rather than measurement. And the magic ranges themselves — 84–97 versus 88–93 versus 88–94 — presented as though the boundaries were physiologically meaningful. They are conventions, and since FTP is not a valid steady-state marker in the first place, a four-percentage-point argument about zone edges is noise. TSS and CTL are bookkeeping units, not physiological measurements, and weekly TSS is not comparable across phases with different intensity profiles.

How it goes wrong

Doing it four or five times a week. Past 2–3 sessions you accumulate fatigue with no clean adaptive target. This is the failure the whole critique is aimed at, and FasCat's own instruction is blunt: do not sweet spot every day.

Progressing watts instead of minutes. Creeping the power turns sweet spot into threshold within about three weeks, and then you are doing threshold work three times a week and wondering why you are flat.

Shortening the recoveries. Five to eight minutes easy between reps is what makes the session repeatable; cut it to a minute and you have written a threshold session with a sweet spot label on it.

Anchoring to a wrong FTP, so that the band is nominally 88–94% and physiologically 100%+.

Translating the percentage across sports. This is the big one for a triathlete: 84–97% is a percentage of POWER, and percentages of power do not map to percentages of PACE. Running at 90% of threshold pace is an easy jog. Swimming at 90% of CSS speed is a warm-up. There is no valid arithmetic conversion — the run and swim equivalents have to be set by physiology.

Counting the budget per sport. Two sweet spot bikes plus two steady-state runs plus three CSS swim sets is seven moderate sessions and no easy days. This is the most common way triathletes wreck an otherwise sensible method.

Running it year-round with no VO2max block and no genuine easy riding, then concluding you have hit your ceiling.

Believing the marketing. "Five hours of sweet spot equals 10–20 hours of traditional base" is TrainerRoad's line, and TrainerRoad's own founder has said the Sweet Spot Base name "shot ourselves in the foot" — those plans were never mostly sweet spot. Take the training, leave the equivalence claim.

Sources: fascatcoaching.com fascatcoaching.com fascatcoaching.com trainerroad.com trainerroad.com trainerroad.com trainerroad.com trainerroad.com trainerroad.com support.trainerroad.com pubmed.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov pmc.ncbi.nlm.nih.gov pubmed.ncbi.nlm.nih.gov topendsports.com trainingpeaks.com