Banister TRIMP
Training Impulse and Fitness-Fatigue Modeling · Also known as: TRIMP, training impulse, fitness-fatigue model
The Training Impulse (TRIMP) model, developed by Eric Banister and colleagues (1975), quantifies the physiological stimulus of a training session by combining duration and intensity. The Banister fitness-fatigue model proposes that training effects on performance follow two opposing dynamics: fitness (beneficial) accumulates with time constant tau_f (~42 days) and fatigue (temporary decrement) accumulates faster but decays quickly (tau_d ~5-10 days). By tracking TRIMP and modeling these two processes, coaches can predict performance trajectories and optimize training load. Although superseded by newer frameworks, the Banister model remains influential and intuitive.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Banister modeling is useful for research investigating training dose-response and for retrospective analysis of seasonal performance trends. The model is less practical for day-to-day coaching due to complexity and need for accurate TRIMP quantification. The method assumes clear performance metrics (race times, test results) and consistent training documentation.
Strengths & limitations
- Mechanistically intuitive: captures opposing effects of training stimulus
- Enables prediction of performance trajectories; optimize load timing
- Identifies over-reaching or under-reaching before problems emerge
- Quantifies individual training response time constants; personalizes model
- Foundational framework for modern training science research
- Requires precise TRIMP quantification; RPE-based TRIMP is subjective and variable
- Assumes exponential dynamics; human adaptation is often nonlinear or threshold-based
- Individual time constants vary widely; generalized estimates may not apply
- Performance must be measured objectively and regularly; difficult in team sports with infrequent testing
- Model fits past data well but predictive accuracy for future performance is modest
Frequently asked
What is TRIMP?
Training Impulse (TRIMP) is a composite score combining training duration and intensity. A simple version: TRIMP = duration (minutes) × intensity factor (0-1 scale). A 60-minute session at 70% max HR might be scored as 60 × 0.7 = 42 TRIMP units. More sophisticated versions weight intensity non-linearly. TRIMP allows comparison of different training modes on a common scale.
What are typical fitness and fatigue time constants?
Research suggests fitness time constant (tau_f) ranges 30-60 days; fatigue (tau_d) ranges 5-15 days. This means fitness develops slowly but persists for weeks, while fatigue appears and disappears quickly (days). Individual variation is large; calibrate from your own data.
Can the Banister model predict race performance?
The model predicts relative performance trends (peak timing, performance dips) better than absolute race times. Model predictions are typically within 2-5% of observed performance for athletes in familiar events. Use model for optimal load timing and periodization, not absolute time prediction.
How often should I update the model?
Recalibrate model parameters every 4-8 weeks using recent performance tests. Fitness and fatigue time constants can shift with training phase and sport-specific conditioning. Quarterly calibration ensures model stays relevant and predictive.
Is Banister model obsolete?
Modern training science uses more sophisticated models (nonlinear dynamics, machine learning) that account for multiple physiological systems. However, Banister's basic principle—opposing fast-decaying fatigue and slow-accumulating fitness—remains valid. The model is useful for understanding training physiology and retrospective analysis, even if newer tools offer better prediction.
Sources
- Banister, E. W., Calvert, T. W., Savage, M. V., & Bach, T. (1975). A systems model of training responses and its relationship to muscular strength. Transactions of the ASME, 97(3), 177-183. link ↗
- Morton, R. H., Frick, U., & Bazalgette, D. (2005). Modeling human performance in running. Journal of Applied Physiology, 92(6), 2393-2402. link ↗
- Clarke, D. C., & Skiba, P. F. (2013). Rationale and resources for teaching the mathematical modeling of athletic training and performance. Advances in Physiology Education, 37(2), 134-142. DOI: 10.1152/advan.00078.2011 ↗
How to cite this page
ScholarGate. (2026, June 3). Training Impulse and Fitness-Fatigue Modeling. ScholarGate. https://scholargate.app/en/sports-science/banister-trimp
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Acute-Chronic Workload RatioSports Science↔ compare
- Session RPESports Science↔ compare
- Time-Motion GPSSports Science↔ compare