Training Concept

Adaptive Mountain Training: A New Category

Most training plans are static. They are written once, downloaded as a PDF, and frozen. The first missed week or surprise overshoot makes them obsolete. Adaptive mountain training is the opposite: a plan that reads what you actually did, then reshapes the next week based on it. Here is what it is, the science underneath, and what changes when you move from static to adaptive.

Why static plans break in week two

A typical mountaineering training plan is a 12 to 24 week grid. Monday is intervals, Tuesday is easy aerobic, Wednesday is rest, and so on. It is designed for a hypothetical athlete who never travels, never gets sick, never overshoots a session, and always recovers exactly the way the spreadsheet expects. That athlete does not exist.

The moment real life shows up - a missed week, a Tuesday session that came in 20 percent above target, an unexpected work trip - the rest of the plan no longer makes sense. The spreadsheet has no way to know. So the athlete keeps following an outdated plan, or improvises off it, or gives up entirely. Most amateur mountaineers fall into the second category, and that is the silent reason their training does not produce the fitness it was supposed to.

This problem is not new. The training science community has known about it for fifty years. The solution requires a plan that recalculates - not just once at the start, but every single day.

What adaptive training actually means

"Adaptive training" is not a marketing term. It refers to a plan that reads your actual training data and modifies upcoming sessions accordingly. Three things distinguish it from static training:

Compared to "personalised" training (a plan customised to you at the start, then frozen) or "AI coaching" (a chatbot that gives advice on demand), adaptive training is structural. The plan itself changes shape over time without you having to ask.

The science underneath

The intellectual foundation is the fitness-fatigue model, developed by Eric Banister and colleagues in the 1970s (Banister et al., 1975; Calvert et al., 1976). The model says any training session produces two effects: a positive fitness component that decays slowly (over weeks), and a negative fatigue component that decays quickly (over days). Performance at any given moment equals fitness minus fatigue.

In practical terms:

An adaptive system reads these values daily. If your ATL spikes 20 percent above your CTL after a hard week, the model knows you are accumulating fatigue faster than you are building fitness. It rebuilds the upcoming week to protect the fitness curve. If your TSB drifts deeply negative for too long, the system schedules forced recovery before you crack. (Busso, 2003 extended the model to handle individual response variation.)

A static plan assumes the athlete never deviates. An adaptive plan assumes the athlete always does, and treats that as data, not failure.

What changes day to day in an adaptive plan

In practice, four things shift continuously:

Session intensity

If you came in 15 percent above plan on Monday, Tuesday's session is adjusted - not because the plan is "punishing" you, but because the next session was assumed against a planned ATL that no longer matches reality. The session reshapes to land where it was supposed to physiologically.

Weekly volume

If you missed two sessions, the system does not just shove them into next week. It recalculates the whole block, possibly extending the build phase by a week, possibly accepting a small hit to peak fitness, possibly shifting a hard session forward.

Phase transitions

Base, build, peak, taper. The boundaries between phases are not date-based - they are fitness-based. An adaptive system shifts you into the build phase when your CTL has reached a target floor, not when 8 weeks have elapsed on the calendar.

Recovery prescription

Real recovery is prescribed when TSB drifts dangerously negative, not just on calendar rest days. The system watches the data and pulls back when needed.

The role of data

An adaptive plan is only as good as the data feeding it. For mountain athletes, three streams matter:

More data is not always better. A plan that obsessively reacts to every single session jitter creates noise. The point is to detect signal - sustained drift in fitness, fatigue, or form - and respond to that, not to every individual data point.

The takeaway

Adaptive mountain training is not magic. It is the application of a 50-year-old physiological model (fitness-fatigue) to the daily reality of an athlete with a real life. The plan reads what you did, recalculates what your body actually needs, and reshapes the upcoming week. For mountain athletes specifically, this matters more than for road runners or cyclists, because the consequences of arriving fatigued at a 4000m peak are bigger than arriving fatigued at a 10K. Alpine training, mountaineering plans, and climbing training apps all benefit from being adaptive. The category is small today; it will not stay that way.

Train adaptively. Arrive ready.

Train to Mountain reads your Strava data, calculates fitness and fatigue daily, and rebuilds the upcoming week so it matches the body you actually have. Not the body the PDF assumed.

Join Early Access →