Detraining: how fast you lose fitness (and how fast it comes back)

Every rider asks this at some point, usually while staring at a cast, a work deadline or a fortnight of hotel rooms: how much fitness am I actually losing? The honest answer is less dramatic than the panic suggests in the first two weeks, and more serious than most people expect after two months.

Sports science has studied this carefully, mostly by taking trained athletes and stopping them. What follows is the pattern that emerges. Two warnings before the numbers: everything below is a population average, and individual variation in detraining is large — and the figures come from studies of already-well-trained people, which is the group that has the most to lose.

Two different things are happening

Detraining is conventionally split at about the four-week mark, and the split matters because the two halves feel completely different.

The week-by-week picture

Read this as a rough map, not a prediction of your own numbers. A pooled analysis of 21 detraining studies found VO2max fell by about 3.9% on average with up to 30 days off and about 9.4% beyond 30 days — but individual studies in that range report anything from almost nothing to a fifth of a rider's aerobic capacity.

Why the first week feels fine — and then doesn't

The earliest casualty is plasma volume. Training expands the liquid part of your blood; stop training and it contracts, with reductions of around 5% reported within a fortnight and blood volume down roughly 9% after two to four weeks off. Less blood volume means less filling of the heart, a smaller stroke volume, and a higher heart rate to move the same oxygen.

This is why the classic detraining experience is not "I have no power" but "my heart rate is twenty beats higher for the same power and I don't know why". It also explains why the first week can feel deceptively good: you have shed fatigue and refilled glycogen while the structural losses have barely started.

The muscle side takes longer to show and longer to fix. Citrate synthase — a standard marker of mitochondrial density — has been reported to fall substantially over weeks rather than days, and after around eight weeks off, studies of runners and cyclists have documented a shift of muscle fibres away from the fatigue-resistant type IIa profile. That is the part of detraining that genuinely costs you a training block to rebuild.

How little is enough to hold on?

This is the practical question, and the good news is that the answer is "not very much". Detraining is driven by an insufficient stimulus, not by a reduction in volume as such. Reduced training — fewer or shorter sessions that keep some intensity — preserves far more than complete rest does.

• Keep intensity before you keep volume. When training has to shrink, the sessions with some hard work in them do the most to defend what you have built.
• Two or three short sessions a week is not nothing. It is the difference between the reduced-training and the complete-cessation columns of every study on this.
• If you're injured, keep whatever moves. Any aerobic work you can do without aggravating the injury protects blood volume and a good deal of the metabolic machinery.
• A planned two-week break is not detraining. It is the thing that lets you train properly for the next ten months. A taper is the same mechanism, aimed deliberately.

Coming back

Retraining is faster than the original build, and every experienced rider knows it — the first three weeks back feel like fast-forward. Plasma volume rebounds within days to weeks, which is why the heart-rate-for-power relationship often normalises long before your ceiling does.

Beyond that, the mechanism is genuinely contested. The popular explanation is muscle memory: that myonuclei gained during training are retained through detraining, giving previously trained muscle a head start. The direct evidence for that is largely from rodent work. A human training–detraining–retraining study found strength and cross-sectional area returned readily while myonuclear number did not behave as the theory predicts, and the authors of that field have been arguing about it ever since. So: retraining is reliably faster; the reason is not settled. Be sceptical of anyone who states the mechanism as fact.

Three things riders get wrong about this

• "Two weeks off and it's all gone" — Two weeks costs a few per cent of VO2max on average, and a good chunk of what you feel is lost plasma volume — which comes back quickly.
• "I'll just train through the illness" — Training through costs you more than the week you were trying to protect. The week-one column of that table is the honest answer.
• "Retraining starts from scratch" — It does not. Riders who have been trained before regain fitness faster than they built it the first time — although exactly why is still argued about.

Where a training log actually helps

The useful thing during a layoff is not a number telling you how much you have lost — no model can know that — but an honest record of what you did and did not do. Moveee's Fitness & Form page shows Fitness decaying while you are off, which at least makes the trade-off visible: how much a reduced-training week holds onto versus a week of nothing.

On the way back, the more useful feature is the plan that eases when you are tired. A returning rider's biggest risk is enthusiasm, and an adaptive plan that backs off after a session you clearly didn't absorb is more valuable in week one back than any test protocol. Retest your FTP after two or three weeks of consistent riding, not on day three — see FTP explained.

And if you are reading this because you are about to take time off: take it. The floor you fall to is much higher than zero, the way back is quicker than the way up, and a season interrupted by two honest weeks of rest beats one ground down by guilt.