Race Time Predictor

About this tool

Riegel formula across distances

Predict race times across distances using the Riegel formula. Enter your known race, get predictions for 1 mile, 5K, 10K, half, marathon, 50K, 100K. Used by Strava + Garmin.

RT-HLT-012 · Health & Fitness

Known race distance (km)

Known race time (h : m : s)

: :

Fatigue exponent

Distance	Predicted Time	Pace
Enter your known race time + distance above

Enter a known race to predict times at other distances

How to use the Race Time Predictor

Enter a recent race result

Use your best race time from the last 6 months. Ideal source distances: 5K, 10K, or half marathon — these are the most accurate baselines. Time-trial efforts on a track work too, but should be true-max effort with reasonable pacing.

Read the predicted times across distances

The table shows predicted times for 1 mile, 5K, 10K, 15K, half marathon, marathon, 50K, 100K. Your input row is highlighted. Use predictions for race-day pacing strategy + goal-setting. Caveat: predictions assume similar training base + race-day effort. A 5K-trained runner won\'t hit Riegel-predicted marathon time without endurance training.

Adjust the fatigue exponent if needed

Default 1.06 is the original Riegel value, accurate for trained runners. Try: 1.07-1.08 for less-trained runners or first-time marathoners (more slowdown at distance); 1.05 for elite-trained runners (less slowdown); 1.10 for tropical heat conditions or first marathon ever.

Use predictions for pacing + planning

The pace column tells you per-km pace for each distance. Use this to: (1) plan race-day pacing strategy; (2) calibrate training pace targets (Daniels-style training paces tied to race performance); (3) set realistic goal times for upcoming races. Predicted time + 5-10% buffer is a sensible first-attempt goal for any new distance.

Riegel formula — how Strava + Garmin predict your race times

Pete Riegel\'s 1981 paper "Athletic Records and Human Endurance" introduced the formula T₂ = T₁ × (D₂ / D₁)^1.06 for predicting race times across distances. The exponent 1.06 represents the "fatigue cost" of running further — a 10K takes ~2.07× your 5K time (not 2×) because endurance fatigue accumulates. Riegel validated the formula empirically against thousands of race results from amateur to elite athletes. Forty-plus years later, the formula remains the most widely-used race-prediction tool: Strava, Garmin Connect, COROS, Polar, TrainingPeaks, and most coaching software embed Riegel or a variant. This calculator implements the classic version with an adjustable exponent for different runner types + conditions.

Why the exponent 1.06 works

If running were purely linear (no fatigue), the exponent would be 1.0 — a 10K would take exactly 2× your 5K time. Reality: longer races involve more aerobic vs anaerobic energy, more accumulated fatigue, more pacing challenges, more mental load. The 1.06 exponent is the empirical best-fit across data: 5K → 10K ratio is 2^1.06 ≈ 2.085 (instead of 2.0); 10K → marathon ratio is 4.2^1.06 ≈ 4.54 (instead of 4.2); half → marathon ratio is 2^1.06 ≈ 2.085 (instead of 2.0). For most trained amateur runners, these ratios match real-race performance within 3-5%. Elite athletes often run sub-1.06 (smaller slowdown) because they\'ve trained the endurance ceiling higher; first-time marathoners often run super-1.06 (bigger slowdown) due to wall + pacing inexperience.

Riegel's formula predicts within 3-5% for trained runners. It's used inside Strava, Garmin Connect, COROS, Polar, TrainingPeaks — basically every endurance platform.

Where Riegel breaks down

The formula has known limitations. (1) Distance ratio beyond 3×: predicting marathon time from a 5K is less accurate because the energy systems differ too much (5K is heavily aerobic-glycolytic; marathon is primarily aerobic-oxidative). (2) Untrained athletes: someone who can do a 22-min 5K but has never run more than 10K will likely run substantially slower than Riegel predicts for marathon — they lack endurance base. (3) Heat + humidity: tropical conditions penalise longer races more than shorter (heat-stress accumulates over time). ASEAN-marathon Riegel predictions should bump the exponent to 1.07-1.08. (4) Trail + hills: Riegel assumes flat road; trail races + significant elevation gain require different prediction models. (5) Ultra distances: 50K+ ultras involve nutrition, sleep, terrain — Riegel underestimates how slow these can get. (6) Elites and recreational runners differ: elites typically run sub-1.06; recreational runners often super-1.06. Adjust the exponent if you know which camp you\'re in.

The ASEAN race-time reality

Tropical race conditions in ASEAN markets significantly shift race-time predictions. Singapore Marathon (December, ~25-28°C, 80%+ humidity), KL Marathon (September, 28-32°C), Bangkok Marathon (November, 26-30°C), Jakarta Marathon (October, 28-32°C) all run measurably slower than equivalent-effort cool-condition races. Heat penalty estimates: ~3-5% time loss at 25°C vs 15°C ideal; 7-12% at 30°C; 12-20%+ at 32-35°C. For Riegel predictions: use exponent 1.07-1.08 for ASEAN long-distance races; goal times should bake in the heat penalty. Early-morning starts (3-5am) help significantly — Singapore Marathon and most regional events use early starts for this reason. Trail + altitude: ASEAN trail races (Cameron Highlands, Da Lat, Bali) at altitude or with elevation gain require additional adjustments — Riegel doesn\'t capture climb effects. For long ASEAN ultras, factor in nutrition + heat strategy heavily.

10 Things to Know About Race Prediction

Riegel formula (1981): T₂ = T₁ × (D₂/D₁)^1.06. The most-used race prediction formula in endurance sports.

Used inside Strava, Garmin, COROS, Polar, TrainingPeaks. Standard tool across the endurance platform ecosystem.

Accuracy: ±3-5% for trained runners within 3× distance ratio (5K→10K, half→marathon).

5K → 10K = ~2.07× (not 2× — endurance fatigue accumulates).

Half → Marathon = ~2.10×. Most amateurs add 5-15% buffer due to "the wall" at mile 18-22.

Exponent 1.05 for elites (less slowdown); 1.08-1.10 for first-time marathoners (more slowdown).

Heat + humidity add ~3-5%/5°C above 20°C. ASEAN marathon predictions need exponent 1.07-1.08.

Predicting marathon from a 5K is less accurate than predicting from half marathon — energy systems differ.

Formula doesn\'t capture elevation, trail, ultra-specific challenges. Use only for road races.

Other prediction models exist (Cameron, Vickers + Vertosick) but Riegel remains the simplest + most widely-validated.

Frequently Asked Questions

Within 3-5% for trained runners predicting across distances within a 3× ratio (e.g., 5K predicting 10K, half predicting marathon). Less accurate at the extremes: predicting marathon from 5K can be off by 10-15% if you don\'t have endurance base; predicting 5K from marathon often underestimates short-distance speed by 5-10%. The formula assumes consistent training across distances — gaps in your training will show up as bigger prediction errors.
Common causes: (1) The wall: first-time + under-prepared marathoners hit glycogen depletion around mile 18-22 and slow dramatically. Riegel doesn\'t model the wall. (2) Pacing errors: starting too fast based on training paces, blowing up later. (3) Heat: 25°C+ races cost 3-10% vs cool conditions. (4) Insufficient long runs: marathon prediction assumes 16-20mi long runs in training; if your longest run was 14mi, expect to run slower than Riegel predicts. (5) Fueling: under-fueled marathons hit the wall earlier. Re-run with exponent 1.08 for realistic first-marathon expectations.
For predictions, 10K is usually best — it\'s long enough to reflect endurance + short enough to be a true effort. 5K predictions for marathon often overestimate (you\'re using a heavily anaerobic baseline to predict an aerobic event). Half-marathon time is the gold standard for marathon prediction — almost the same energy system, just twice the distance. Use 5K only when you don\'t have a recent 10K or half-marathon result.
Riegel doesn\'t handle them well. Trail races involve technical terrain, elevation gain, and pacing strategies that differ from road. Ultras (50K+) involve nutrition, sleep, fatigue management — all variables Riegel ignores. For ultras, use ultra-specific calculators or estimate as: flat-road Riegel prediction × 1.15-1.30 depending on elevation gain. For 100K + 100mi: prediction is highly individual — depends on nutrition + pacing + experience. Use only as a rough planning baseline.
Yes for half marathons + marathons. Tropical heat (25-32°C) penalises long races more than short. Suggested exponents: cool (under 18°C) — use 1.05-1.06; temperate (18-24°C) — use 1.06; warm (24-28°C) — use 1.06-1.07; hot (28°C+) — use 1.07-1.08. Singapore Marathon, KL Marathon, Bangkok Marathon, Jakarta Marathon all benefit from heat-adjusted predictions. The short-distance races (5K, 10K) are less affected — exponent stays close to 1.06 even in heat.
Riegel assumes your training base supports the predicted distance. A fast 5K runner without long-run training won\'t hit predicted marathon time. Calibration check: do your weekly long runs match what marathon training requires (16-20mi peak)? Does your weekly mileage match the goal distance (40-60mi for sub-3:30 marathon)? If training doesn\'t match the distance, increase the exponent to 1.07-1.08 for realistic predictions.
Several exist with marginal accuracy gains. Cameron formula: uses different exponent depending on race time; slightly better for sub-elite athletes. Vickers + Vertosick (Slate Runners): more sophisticated regression model published 2016; slightly better accuracy across distance range. Jack Daniels VDOT: alternative system mapping race times to a "VDOT" number, then to predicted paces across distances. All are within ±1-2% of Riegel for most amateur runners; Riegel\'s simplicity + 40-year validation keep it dominant.
Not directly — cycling has different fatigue dynamics (more aerodynamic drag, less impact load, drafting effects in groups). For cycling, use power-based prediction: FTP (Functional Threshold Power) and intensity-duration curves. Try our Cycling FTP Calculator (RT-HLT-015) for cycling-specific metrics.
No. All calculations run in your browser via JavaScript. Open DevTools → Network and confirm zero outbound requests. Race time + distance stay on your device. Safe for personal training logs + race planning.
Pair with: VO2 Max Calculator (RT-HLT-011) for fitness-level baseline; Running Pace Calculator (RT-HLT-013) for goal-pace splits; Sweat Rate Calculator (RT-HLT-014) for race-day hydration; Heart Rate Calculator (RT-HLT-009) for training zones. External: Daniels Running Formula book; McMillan Running Calculator (paid version with more features); Jack Daniels VDOT tables for training-pace prescription.

Race Time Predictor