Fast twitch, slow twitch: what fibre type means for how you should train

The idea is seductive and everywhere: find out whether you are fast twitch or slow twitch, and train accordingly. Sprinters get short maximal work, diesels get volume, and everybody finally stops doing sessions that were never going to suit them.

Most of the underlying physiology is sound. Muscle fibres really do come in types with genuinely different properties, people really do differ enormously in their proportions, and training really does change some of that. What does not hold up nearly as well is the last step — the inference from "I am probably fast twitch" to "therefore my training should look different". This post covers both halves, including why the practical payoff is smaller and stranger than the premise suggests.

What the types are

Human skeletal muscle fibres are classified by the myosin heavy chain isoform they express, which is the molecular motor that determines how fast the fibre shortens. Three matter in humans: Type I, Type IIa and Type IIx.

Two details are usually lost in the popular version. First, there is no Type IIb in human limb muscle. The fibre type that older textbooks and a great deal of internet writing call IIb corresponds to IIx in humans; true IIb exists in small mammals. Second, a large fraction of fibres are hybrids, expressing more than one isoform simultaneously. The tidy three-box diagram is a simplification of a continuum, and the boxes are where we chose to draw lines rather than where nature drew them.

What training actually changes

Here the evidence is clear and also narrower than people expect. Endurance training, strength training and power training all shift fibres within the fast family: IIx proportions fall and IIa proportions rise. This shows up consistently enough that a meta-analysis found it across all three training modes, which is itself a useful clue — it means the shift is a response to being used at all, not to any particular kind of use.

What is far harder to produce is conversion between the fast and slow families. Reviews of the field describe Type I to Type II conversion as rare under normal training, requiring extreme or very prolonged stimuli. The practical implication: you can make your fast fibres more fatigue-resistant, and you cannot meaningfully turn yourself into a different rider at the myosin level.

There is one genuinely odd finding worth knowing. After a period of heavy strength training, IIx content falls as expected. If the training then stops, IIx does not merely return to baseline — it overshoots above it for a period before settling. That is a strange and well-replicated quirk, and it is part of the rationale for a short taper after a strength block rather than a long one.

Fibre type is also not the only thing that changes. Single-fibre work on runners going through a marathon build showed alterations in fibre size and contractile function that a type-percentage figure would miss entirely. Counting types is a coarse summary of what training does to muscle.

Why you cannot find out your own composition

The reference method is a needle biopsy, and there is no close second. A sample takes a few hundred fibres from one site in one muscle — usually the vastus lateralis — and your quadriceps contains millions distributed unevenly by depth and region. A single sample is an estimate with real uncertainty attached, which is why research studies take care over sampling depth and site and why one result from one needle should not be read as a fixed personal fact.

Everything else is a proxy. Some are better than they look and some are considerably worse.

The power-curve proxy is the one most riders will actually use, and it is worth being clear about what it can and cannot tell you. The ratio between your short-duration and long-duration power reflects fibre composition, but it also reflects gearing, body mass, skill at sprinting and how long it has been since you last did any. Moveee's power profile will show you that shape from your own rides, and what your power curve says about you covers how to read it — as a description of how you currently ride, not as a genetic readout.

There is also a real population signal behind the proxies. When world-class cyclists across different disciplines were typed non-invasively, the typology differed systematically by event: track and sprint specialists sat at the fast end, endurance riders at the slow end. That is exactly what you would expect, and it still does not let you place any individual rider confidently.

So what actually changes in your training?

Less than the framing promises. Here is the honest list.

What does not change

The sessions themselves. A rider with more fast-twitch muscle still needs a large aerobic base, still needs threshold work, still benefits from VO2max intervals. A rider with more slow-twitch muscle still needs to sprint occasionally, and still benefits from lifting weights. There is no training method that is exclusively for one type, and no controlled trial showing that typing athletes and matching sessions to type outperforms simply training them well.

Worse, the advice often runs backwards. "You are fast twitch, so focus on sprints" doubles down on what is already your strength and neglects the aerobic capacity that determines almost everything else in road cycling. If you are a sprinter, the sensible response is usually more base, not less — the argument made in base miles.

What might genuinely change

• How much recovery you leave between hard sessions. Fast-typed athletes were found to take considerably longer to recover from high-intensity exercise than slow-typed ones. If you consistently need three days to feel right after intervals while your training partner needs one, that is a plausible explanation, and it is an actionable one.
• How you interpret a bad session. Repeatability across efforts is partly a fibre-type trait. A rider who produces a huge first interval and fades badly is not necessarily undertrained or badly fuelled — that pattern is also what a fast-typed muscle does.
• What you build a race plan around. If your repeatability is poor, racing to make one decisive effort rather than five is a rational tactic. This is a pacing decision, not a training-type decision.
• How long you taper after a strength block. The IIx overshoot after stopping resistance training gives a mechanistic reason to leave a proper gap, rather than lifting until the week of the event.

The conclusion nobody sells

Fibre type is a real, important source of variation between riders. It explains why sprinting comes easily to some people and never to others, why two riders with the same threshold recover at different rates, and part of why efficiency differs between them. Knowing that the variation exists is genuinely worth something, because it stops you interpreting every difference between yourself and your training partner as a difference in effort or discipline.

But you cannot measure it, the proxies are loose, the training changes it only within the fast family, and the sessions that make you faster are broadly the same either way. "Train to your type" sounds like precision and functions, in practice, mostly as permission to avoid the work you already dislike. The far duller advice — train the system your event demands, and adjust your recovery spacing to what your own body keeps telling you — will beat it.