Start with observable facts
Distance, altitude, elevation and geometry come from the file. Technicality, weather and trail condition still need confirmation.
Race plan masterclass
An elevation profile does not predict your day. It provides the structure for sustainable effort, honest time windows, realistic self-sufficiency and prepared decisions.
Scientific revision 1.0 · 31 July 2026Demonstration
This fictional simulation teaches a method. Its times must not be reused for another race.
Distance, altitude, elevation and geometry come from the file. Technicality, weather and trail condition still need confirmation.
Core principle
A useful plan stays readable when weather changes or a split takes twenty minutes longer. Every value needs a clear status.
GPX distance, track altitude, official aid-station location or cut-off.
Even facts may be incomplete: noisy GPS, route changes or relocated support.Segment duration, passage time or carrying need computed from explicit assumptions.
An estimate is not a promise; confidence falls as context goes missing.Effort intent, planned stop, carrying strategy or action triggered by a deviation.
A rigid decision becomes unsafe when it ignores terrain, symptoms or race instructions.Useful precision is not decimal places. It is the ability to explain where a value came from, what can move it and which decision it supports.
Step 1
Equal kilometre splits create noise. A segment starts when effort, risk, self-sufficiency or logistics change.
Start or end of a sustained climb, summit transition or long descent.
Moving from runnable to technical, mud, snow, rocks or narrow trail.
Altitude, wind, sun, cold, no-exit zone or likely night section.
Confirmed water, official aid station, life base or authorised crew access.
Cut-off, mandatory gear, closure or special race rule.
Where you must check effort, refill, layer up or reassess the plan.
×Poor segmentation: equal distances mixing climb, summit and descent.
✓Good segmentation: each unit can receive one effort intent, probable duration and simple instruction.
Step 2
Locomotion cost changes sharply with grade. Pace per kilometre is a terrain outcome, not a universal command.
Observational studies often associate steadier pacing with better results, but “steady” here means controlled expenditure, not identical speed on every gradient.
Step 3
One time looks reassuring but hides uncertainty. Use three scenarios and widen the window as the race progresses.
Dry trail, mild weather, smooth stops. Not a target to chase.
Most likely assumptions, calibrated on comparable efforts.
Fatigue, heat, technicality or longer stops. Adapt to actual uncertainty.
Use comparable races or outings, not your best road pace.
Movement time and stopped time are separate visible costs.
Later, longer, night or exposed segments need wider uncertainty.
Compare cut-offs with the cautious scenario, not only the central one.
These percentages and times illustrate a method. They are not universal Trailboard engine parameters or a prediction for your race.
Step 4
An aid station reduces carrying only when its location and supplies are confirmed. A map icon or last year’s memory is not enough.
Segment need combines estimated duration and hourly target. Margin follows uncertainty, heat and actual resupply options.
What target was tolerated at a comparable duration and intensity?
Compute segment total and distribute intake without long gaps.What intake was observed in similar conditions?
Plan carrying capacity without forcing drinking beyond need or thirst.What is known about sweat rate and concentration?
Keep the assumption visible and never treat sodium as insurance against all hyponatraemia.What if water is unavailable or one food becomes intolerable?
Carry a simple, tested alternative compatible with the pack.Carried quantity must be explainable per segment. Unknown needs remain ranges and unconfirmed resources remain outside the calculation.
Step 5
A robust plan cannot predict everything. It prepares simple responses to common deviations and defines when safety replaces performance.
Passage leaves the central window but still clears the cut-off.
Identify the cause: terrain, stop, effort or physical issue.
Recalculate from actual time without punitive acceleration.
Confusion, collapse, chest pain, persistent vomiting, inability to drink or pain changing gait are not plan deviations to optimise: stop and seek assistance.
Final deliverable
On race day nobody rereads the model. The final sheet must expose facts, windows, supply and decisions without ambiguity.
When data is not confirmed, write “unknown” or “assumption”. An empty cell must never look like absence of risk.
Bibliography
Pacing studies are mostly observational. They describe associations and trends; they do not prescribe an individual pace.
Shows why distance, grade, terrain, physiology and technique must be considered together.
↗Measures the nonlinear change in energy cost with grade; one kilometre never has a constant cost.
↗Describes speed loss by gradient category and trail-specific pacing variability.
↗Associates steadier pacing with faster results across several UTMB editions.
↗Highlights the importance of preserving downhill speed in later race stages.
↗Connects nutrition, aid-station logistics and ultramarathon-specific demands.
↗Frames race intake, individual tolerance and gut training.
↗Safety consensus on excessive fluid intake and exercise-associated hyponatraemia.
↗Frames how heat changes sustainable intensity, cooling and hydration.
↗Documents sleep management, alertness and reported consequences in mountain ultras.
↗