5K to Marathon Prediction Examples

The most-searched question: "What marathon time can I run based on my 5K?" Here are Riegel formula predictions for common 5K times:

Key benchmarks:

• Need a sub-4 hour marathon? → Aim for a 5K under ~27:15 (with proper marathon training)
• Need a sub-3:30 marathon? → Aim for a 5K under ~22:30
• Need a sub-3 hour marathon? → Aim for a 5K under ~19:00

To predict from a half marathon: double your half marathon time and add 5–15 minutes (lower end for high-mileage runners; upper end for 30–40 miles/week). Or enter your half marathon time directly in the calculator above.

How Race Time Prediction Works: Riegel's Formula

Race time prediction is based on the mathematical relationship between performance at different distances. The most widely used model is Peter Riegel's formula, published in American Scientist in 1977:

T2 = T1 × (D2/D1)^1.06

Where T1 is a known race time, D1 is that race's distance, D2 is the target distance, and T2 is the predicted finish time. The exponent 1.06 reflects the physiological fact that performance degrades faster than linearly as distance increases — longer races are proportionally harder than shorter ones.

Example: a runner with a 5K time of 22:00 predicts a marathon time of: 22:00 × (42.195/5)^1.06 = 22:00 × 9.12 = 200.6 minutes = 3:20:38.

Riegel's formula is remarkably accurate for trained runners racing across their usual distance range. The main limitation: it assumes equal preparation for both distances. If you've trained specifically for 5K but never done long runs, your marathon will be much slower than predicted.

Cross-Distance Prediction Table

Use this reference table to find predicted finish times across distances based on your known performance. Values use Riegel's formula (exponent 1.06):

Note that these predictions assume flat courses in moderate weather conditions (10–15°C) with appropriate race-specific preparation for both distances.

Why Predictions Miss and How to Adjust

Riegel's formula gives a statistically average prediction — but you are not a statistical average. Several systematic factors cause individual predictions to be off:

• Distance specialization: Pure speed runners (who train for 5K) typically perform better at short distances than predicted for marathons. Endurance specialists show the reverse — their marathons outperform their 5K predictions. If your training heavily favors one type of work, adjust predictions accordingly.
• Training volume: A runner with 30 miles/week cannot perform a marathon as predicted by their 5K time vs. a runner with 70 miles/week. Long run training is essential for marathon performance beyond pure aerobic capacity.
• Heat: For every 5°C above 15°C, expect 1–3% slower performance in longer races. This effect compounds — a 30°C marathon day can mean 8–15% slower.
• Course difficulty: Net elevation, total climb, and surface type all affect predictions. Most prediction formulas assume flat road courses.
• Aerobic maturity: Newer runners often see larger improvements between their base-distance races and longer predictions — their aerobic systems haven't yet reached their ceiling.

For most recreational runners, the most accurate predictor of marathon time is a recent half marathon race time (double it and add 10–15 minutes as a rough starting point, or use Riegel's formula).

Other Race Prediction Models

Several alternative models compete with Riegel's formula. Each has different strengths:

• Cameron model: Uses a more complex exponential equation. Generally more accurate at extreme distances (ultramarathons) where Riegel tends to over-predict performance.
• VDOT method (Jack Daniels): Convert your race time to a VDOT score, then look up equivalent times for other distances. Highly accurate for 1500m–marathon range when VDOT is derived from a true maximal effort.
• VO2max method: Estimate VO2max from your race time, then calculate expected time at other distances using the oxygen cost of running equations. Physiologically principled but requires more calculation.
• World Athletics scoring tables: While designed for scoring, these tables implicitly model equivalent performances across distances and can be used for rough prediction.

For practical race planning, Riegel's formula is the go-to for its simplicity and accuracy in the 5K–marathon range. Use multiple models and average them for important goal-setting decisions.

Using Race Predictions for Training

Race time predictions aren't just useful for setting race goals — they're a powerful training tool. Here's how coaches use predicted times:

• Confirm readiness: If your 5K time predicts a 3:30 marathon but you're targeting 3:15, you'll know you need either more training or a more conservative goal.
• Identify weaknesses: If your 10K time predicts a much faster marathon than your actual marathon performance, you likely have a training deficit in long-run endurance.
• Set intermediate goals: Use race prediction to set appropriate 5K and 10K race goals within a marathon training cycle. Racing shorter distances at predicted equivalent effort is a valuable training stimulus.
• Benchmark fitness: Run a time trial (5K or 10K) at race effort in week 8 of a 20-week plan. Use the prediction to verify whether your training is on track for your goal.

The Jack Daniels VDOT approach is especially powerful here: once you calculate your VDOT from any race, you have prescribed training paces AND predicted times for all standard distances simultaneously.

Age-Graded Race Performance

Raw race times naturally decline with age due to physiological changes — reduced VO2max, lower maximal heart rate, slower recovery, decreased muscle mass. Age-graded performance tables adjust for these changes, allowing fair comparison of performances across different ages.

World Athletics (formerly IAAF) maintains age-grading tables. A 60-year-old running 4:30 in a marathon might receive an age-graded score of 72%, meaning their performance equals 72% of the world-record level for their age and gender. A typical competitive recreational runner at any age scores 55–65%; elite age-groupers score 75–85%.

Age-graded times can also be used for prediction: if you know your age-graded factor, you can estimate how much faster you would have run the same race 10 or 20 years ago — or project how your absolute time goal should adjust as you age.

Rule of thumb: VO2max declines approximately 10% per decade after age 25 for sedentary individuals — but only 5–7% per decade for those who maintain consistent aerobic training. Staying active dramatically slows the age-related decline in running performance.

Daniels' VDOT as a Race Prediction System

While Riegel's formula is a pure mathematical model, Jack Daniels' VDOT system takes a physiologically grounded approach to race time prediction. In Daniels' Running Formula, he developed VDOT tables from decades of coaching data that map any race performance to equivalent times at all standard distances from 1500m to the marathon.

The key advantage of VDOT over Riegel: Daniels' tables are empirically derived from actual athlete performances — not just a mathematical extrapolation. They account for the non-linear relationship between distance and performance more precisely, particularly at the extremes (very short and very long races).

VDOT prediction examples compared to Riegel:

For most practical purposes, the difference between Riegel and VDOT predictions is small — typically 1–2 minutes for a marathon. However, VDOT has the significant advantage of also prescribing your training paces, making it a more complete system for both prediction and preparation.

Daniels cautions that VDOT equivalences assume equal training specificity for both distances. A pure track runner may not achieve their VDOT-predicted marathon time without specific marathon preparation. Conversely, a runner who exclusively trains for the marathon may underperform their VDOT equivalent at 5K due to lack of speed work.

Pfitzinger's Approach to Race Time Equivalencies

Pete Pfitzinger, in both Advanced Marathoning and Faster Road Racing, provides practical guidance on using race time predictions within a training context. His approach emphasizes the importance of specificity of preparation as the primary modifier of any mathematical prediction.

Pfitzinger identifies three tiers of prediction reliability:

• High reliability (adjacent distances): Predicting 10K from a recent 5K, or half marathon from a recent 10K. These distances are close enough that training overlap is substantial. Accuracy: within 1–2%.
• Moderate reliability (2–4x distance ratio): Predicting marathon from a half marathon, or half marathon from a 5K. Requires adequate distance-specific training. Accuracy: within 3–5% with proper preparation.
• Low reliability (extreme ratios): Predicting marathon from a 1500m or mile time. The physiological demands are so different that mathematical prediction alone is unreliable. Can be off by 10%+ without marathon-specific training.

Pfitzinger's practical rule for marathon prediction from a half marathon: double the half marathon time and add 5–15 minutes, with the lower end for well-trained runners (60+ miles/week) and the higher end for lower-mileage runners (30–40 miles/week). This "doubling plus" rule accounts for the exponential fatigue of the marathon's second half.

He also emphasizes that prediction accuracy is seasonal. A race time from 6 months ago is a less reliable predictor than one from 4–6 weeks ago, because fitness changes — both positive and negative — affect prediction validity significantly.

The Hansons Cumulative Fatigue Model and Race Prediction

The Hansons method offers a unique perspective on race prediction: rather than relying solely on mathematical formulas, they use training performance under fatigue as a direct predictor of race capability.

In the Hansons system, key training benchmarks predict race readiness:

The Hansons approach to prediction is fundamentally practical: if you can hit your training benchmarks on fatigued legs, your race goal is achievable. If you're consistently missing pace targets in training, the mathematical prediction is irrelevant — your body is telling you the goal needs adjustment.

This training-based prediction approach complements Riegel and VDOT by adding a real-world validation layer. The best race predictions combine: (1) a mathematical model from a recent race time, (2) confirmation from training performance at race-specific efforts, and (3) adjustment for race day conditions (weather, course, altitude).