Skip to content
Running Browser

Training fatigue

How distance and Effort build into a 21-day training load, with decay, rest days and worked examples.

A modeled estimate from recent training load, not a direct measurement of how tired or recovered you feel.

Physiological Background

Training fatigue accumulates over time and dissipates through recovery. This follows the fitness-fatigue model (Banister, 1975), where each workout adds to both fitness and fatigue, but fatigue decays faster than fitness gains. [1]

The Fatigue metric represents accumulated fatigue from recent training, with more recent workouts contributing more heavily. This helps identify:

  • Overreaching risk - when fatigue climbs too high
  • Taper effectiveness - fatigue should drop before races
  • Training consistency - stable fatigue indicates regular training
  • Recovery status - how much fatigue remains from recent work

Formula

Fatigue is calculated as the weighted sum of training load over the past 21 days, with exponentially decaying weights: [2]

F = (1/1000) · Σi=020 (di · Ei + si) · Di

Where:

di
Distance in meters on day i
Ei
Effort value on day i
si
Secondary distance (strength work) on day i
Di
Decay factor for day i

Decay Factors

The decay factors model how fatigue dissipates over time. As in the fitness-fatigue model, fatigue fades exponentially: every day a workout keeps the same share (87%) of the weight it had the day before. Today's workout has full impact (1.0), while a workout from 20 days ago contributes only about 7% of its original load: [1]

Di = e−0.135·i

Where:

i
Days ago (0 = today, up to 20)
0.135
Decay rate per day: a time constant of about 7.4 days, so a workout's weight halves roughly every 5 days
051015200.000.250.500.751.00Days agoDecay factor Dday 5: 0.51day 20: 0.07
Decay factor D by days ago. Today's workout counts in full; after 5 days about half of its load remains, after 20 days about 7%, and from day 21 it no longer counts toward fatigue. The shaded area shows how one workout's load is spread over three weeks.

The decay curve follows research on glycogen replenishment, muscle repair, and neuromuscular recovery. Most acute fatigue clears within 7 days, but residual effects persist for 2-3 weeks.

A rule of thumb

If every day were the same, Fatigue would settle at one day's load times the sum of the 21 decay factors, 7.46. A day's load is its kilometres times its Effort, so with the week spread evenly:

F ≈ 1.07 · kmweek · Eavg
The Fatigue a week settles into, by weekly volume and average Effort.
Weekly kmE = 1.2E = 1.4E = 1.6
30384551
50647585
7089104119
90115134153

Real weeks are not even: a long run yesterday weighs more than one six days ago, so daily Fatigue swings around this level (figure 3.2). Still, it says what a week of training settles into.

How fast Fatigue responds

After a lasting change in training, Fatigue covers about 65% of the way to its new level in one week, 90% in two and all of it in three, when the oldest days leave the window. The other way round, a week off takes away about 65% of it. That is why a taper of two or three weeks lowers Fatigue clearly, while fitness, built over months, holds.

What counts

  • Runs: distance in metres times Effort.
  • Strength and other training: its secondary distance in metres, added as it is, so 1,000 m of secondary distance weighs like a kilometre at Effort 1.0.
  • Planned runs: at their planned pace, or the default Z2 pace without one, so you see where a plan takes your Fatigue before you run it.
  • Several workouts on one day: all of them.

When Fatigue is shown

Fatigue needs an entry on each of the last 15 days: a run, other training or a rest day. With gaps the app shows nothing rather than a number that would read too low, since a missing day may be a run that was never logged. Log rest days as workouts in the Rest category; plan templates do it for you. Fatigue is rounded to whole numbers.

Interpreting Fatigue Values

Guideline ranges for the fatigue value F.
FLoadMeaning
< 30Low LoadRecovery phase, detraining risk if prolonged
30–50Moderate LoadMaintenance training, good for base building
50–70High LoadPeak training, monitor for overreaching
> 70Very High LoadRisk of overtraining, consider recovery

Note: These ranges are individual. Athletes with years of training may handle higher fatigue, while beginners should stay conservative. Track your personal patterns over time.

Using Fatigue for Race Preparation

These metrics help you plan training load during peak phases leading up to a race. The key principle: avoid exceeding fatigue levels significantly higher than your previous training peaks.

When building toward a goal race:

  1. Review your fatigue values from previous successful training blocks
  2. Use those peaks as a ceiling - don't exceed them by more than 10-15%
  3. Build gradually toward that ceiling over several weeks
  4. Allow fatigue to drop during taper (typically 30-50% reduction)
NovDecJanFebMarApr0255075100DateFatigue FTaperRacePeak 10456
Daily fatigue of the sample runner from the home page demo through a half-marathon build, computed with the app's own formulas. Fatigue climbs to a peak of 104, falls by 46% during the three-week taper (shaded) to 56 the day before the race, and keeps dropping in the recovery weeks after it.

Pushing fatigue far beyond what your body has previously handled increases injury risk and often leads to arriving at race day fatigued rather than fresh. Progressive overload means small, sustainable increases - not dramatic jumps that overwhelm your recovery capacity.

Sources & further reading

Scientific References

These sources provide the scientific background discussed in this article. The formulas and coefficients used by Running Browser are described above, including their assumptions and limits.

  1. Banister, E.W. (1991)

    Modeling elite athletic performance

    Physiological Testing of the High-Performance Athlete · Library record

  2. Foster, C. et al. (2001)

    A new approach to monitoring exercise training

    Journal of Strength and Conditioning Research

View the complete bibliography

Running Browser

Use the models in your own training

Plan and log your runs in Running Browser. Try it without an account; register to save your training across devices and connect Strava.

Try Running Browser