FTP rises from repeated, structured time near and above threshold — sweet spot, threshold and VO2max intervals — built on a base of genuinely easy riding. Most riders who stall are not doing too few intervals. They are riding their easy days too hard to recover for the hard ones.
The three intensities that move FTP
- Sweet spot (roughly 88–94% of FTP). Hard enough to force adaptation, gentle enough to repeat two or three times a week. The workhorse. Blocks of 8–20 minutes.
- Threshold (roughly 95–105%). Directly trains the line you are trying to move. Costly to recover from — once or twice a week at most. Blocks of 8–20 minutes.
- VO2max (roughly 106–120%). Raises the ceiling that FTP sits under. Short and sharp: 3–5 minutes, or repeated 30–40 second efforts. Once a week is usually enough.
None of these is magic on its own. What raises FTP is the accumulation: the same sessions, repeated across weeks, at an intensity you can come back from.
The reason most riders plateau
It is rarely a missing interval session. It is the easy rides that are not easy. Zone 2 that creeps to Zone 3 because the group went up the road, because the climb was there, because it felt good.
That middle intensity is expensive — it costs real fatigue — and it buys very little adaptation. Ride it habitually and you arrive at every hard session slightly too tired to hit the numbers that would actually have moved your FTP. The session gets completed. It just does not do what it was for.
The uncomfortable fix: make your easy rides genuinely, boringly easy. Most riders find this harder than the intervals.
Does the model matter as much as the argument suggests?
Polarized, pyramidal, sweet-spot-first — the internet argues about this constantly. A 2024 systematic review with meta-analysis is the closest thing to an answer, and the answer is deflating in a useful way.
Seventeen studies and 437 participants were pooled. Polarized training came out ahead for peak oxygen uptake, but only slightly (SMD 0.24, 95% CI 0.01–0.48), and only in interventions shorter than twelve weeks and in already highly trained athletes. For everything closer to what you actually care about, the distributions were indistinguishable: time-trial performance (SMD −0.01), and power or speed at the second threshold — the closest laboratory stand-in for FTP — came out at SMD 0.04, 95% CI −0.21 to 0.29 (Sports Med, 2024).
In other words: the label on your intensity distribution is not what decides whether your threshold moves. Consistency, the total amount of quality work, and whether your easy days are actually easy will decide far more than which camp you joined.
🔴 The limits: seventeen studies across several endurance sports, not cyclists alone; the randomised trials were rated as having some concerns for risk of bias. It does not prove all distributions are equal — it shows the differences are small enough that they are not your problem yet.
Which intervals actually moved threshold power
A 2025 crossover study put the question directly to well-trained cyclists. Twenty-two riders (mean VO2max 69.5 mL/kg/min) completed both a one-week block of moderate-intensity intervals — six sessions, 5–7 repeats of 10–14 minutes, RPE around 14.5 — and a one-week block of high-intensity intervals, five sessions of shorter, harder work at RPE around 17. The blocks were two months apart, and testing came after six days of easy riding (Eur J Sport Sci, 2025).
- 15-minute maximal power: both blocks improved it — moderate +4.9%, high +2.8% — with no significant difference between them (p = 0.44).
- Power at 4 mmol lactate (the threshold-like measure): moderate intervals won, +4.5% against +2.1% (p = 0.03).
- No difference between blocks in VO2max or in efficiency.
The practical reading matches the section above: the longer, slightly-less-brutal interval is at least as good for the number you are chasing, and easier to repeat.
🔴 The limits, and they matter. Twenty-two riders, one week per block, and already very fit — a VO2max of 69.5 is not the average club rider. And look at the spread: that 4.9% improvement carried a standard deviation of 8.7 percentage points — a spread nearly twice the size of the average gain, which tells you the group contained riders who did not improve at all. Group averages do not promise you anything individually. What works is whatever you can repeat and recover from — which is something you find out by watching your own responses, not by copying a study.
How to sequence it across weeks
A workable pattern is two to three hard sessions a week, everything else easy, then a lighter week roughly every fourth. The light week is not a reward for good behaviour — it is when the adaptation you have been forcing actually gets built.
Progress the work before you progress the intensity: add a repeat, or a minute per repeat, before you add watts. And change one thing at a time, so that when the number moves you know what moved it.
How long before you see it
Give any consistent block six to eight weeks before you judge it. Newer riders often see a lot in the first months, simply because there is a lot of untrained ground to cover. Riders with years of structure move in smaller increments, and that is not failure — it is what a mature aerobic system looks like.
Be careful what you attribute: a bigger number after a rest week may be freshness rather than fitness. The trend across several tests is the honest measure.
Your numbers, explained the moment you ask.
The part no training plan on paper can do is notice that today is not the day. A fixed plan written six weeks ago does not know that you slept badly, that work went sideways, or that Sunday's ride cost you more than it should have. So it hands you threshold intervals anyway, you complete them at 92% of the target, and the session that was supposed to raise your FTP quietly becomes another medium-hard day.
Peakfy reads what your rides and your recovery actually say and adjusts today's session to today, then tells you why it moved. When you disagree, you can say so — the conversation has no counters and no credits to top up, so arguing with your coach costs you nothing.
Questions riders ask
Keep reading
- What is FTP in cycling, and how do you test it?
- Recovery and readiness in cycling: reading the morning signals
- Critical power vs FTP: which number should guide your training?
- Cycling training over 40: what changes, and what to do with it