Welcome to our marathon race time calculator.
Simply enter how fast you’ve run one or two previous races and how much you’ve been training approximately per week, and this calculator will predict how fast you’d run a marathon.
Unlike most race time prediction tools, which tend to become inaccurate when predicting distances greater than a half-marathon, this marathon time calculator utilizes an advanced set of formulas developed specifically for predicting marathon race times for non-professional runners.
The formulas used in this marathon time calculator were developed by Andrew J. Vickers and Emily A. Vertosick at Slate.
For predicting a wider range of race distances, check out our classic race time predictor and our age-grade calculator.
Marathon Race Time Prediction Calculator
Previous Race 1
: :Previous Race 2 (Optional)
: :Looking to convert your race time goal into a pace target?
Then head to our Running Pace Calculator.
The Honest Truth About Marathon Time Predictors
Every marathon-time predictor is a statistical model fitted to someone else’s training and race data, projected onto you. The formulas work strikingly well for runners whose training profile matches the dataset the model was fitted on — and miss by 10–20 minutes for those who do not. Knowing which assumptions drive each formula, and where those assumptions break for your specific case, is the difference between a useful pace target and a race-day disaster.
What the main prediction formulas actually model
The Riegel formula (T2 = T1 × (D2/D1)^1.06) was derived from race-result datasets assuming a consistent fatigue exponent across distances 1Riegel PS. “Athletic records and human endurance.” American Scientist 69, no. 3 (1981): 285–290. The 1.06 fatigue exponent was fitted to pooled race-performance data across distances and is demonstrably too optimistic for untrained runners and too conservative for elites.. Daniels’ VDOT approach maps a race performance to an equivalent VO2max and then reads off expected times for other distances using oxygen-cost tables 2Daniels J. Daniels’ Running Formula. 3rd ed. Human Kinetics, 2014. VDOT equivalents assume stable running economy and fractional utilization across event durations; they underpredict marathon time for endurance-under-trained runners and overpredict for endurance-specialists.. Cameron’s formula and its derivatives correct Riegel for longer distances empirically, while Tanda’s model adds weekly-mileage and mean-training-pace terms that explain ~77 percent of variance in recreational marathoners 3Tanda G. “Prediction of marathon performance time on the basis of training indices.” Journal of Human Sport and Exercise 6, no. 3 (2011): 511–520. Weekly mileage and average training pace explained ~77 percent of the variance in marathon finish times in a recreational sample.. Each formula answers a different question — “what does your raw speed predict?” vs. “what does your training say you have earned?” — and the gap between the two is often where races go wrong.
Why VO2max-based predictions overestimate marathon pace
Marathon pace for trained runners sits at roughly 75–85 percent of VO2max for 3+ hours, and fractional utilization — how high a percentage of VO2max you can hold for marathon duration — is highly individual 4Joyner MJ, Coyle EF. “Endurance exercise performance: the physiology of champions.” Journal of Physiology 586, no. 1 (2008): 35–44. Marathon-pace fractional utilization of VO2max varies widely between trained runners and is strongly coupled to endurance-training volume rather than raw VO2max.. A runner with a 4:45 mile and a 17:30 5K might have the VO2max to run 3:00–3:10 for a marathon, but if their long run tops out at 14 miles their glycogen and connective-tissue systems will crack somewhere around mile 18–22. Running economy matters at least as much as VO2max at a given pace — two runners with identical VO2max can differ 15–20 percent in finish time depending on oxygen cost of gait and fractional utilization 5Jones AM. “The physiology of the world record holder for the women’s marathon.” International Journal of Sports Science and Coaching 1, no. 2 (2006): 101–116. Economy differences of 5–10 percent translate to minutes over a marathon independent of VO2max.. That is why a short-distance predictor based on 5K or 10K alone consistently overstates marathon readiness for under-mileaged runners.
The training-volume correction no calculator asks about
The correlation between weekly mileage and marathon finish time is one of the most robust findings in the literature — Tanda showed that weekly volume and average training pace explain roughly three-quarters of the variance in recreational marathon times, independent of body composition 6Tanda G, Knechtle B. “Marathon performance in relation to body fat percentage and training indices.” Open Access Journal of Sports Medicine 4 (2013): 141–149. Training-pace and weekly volume predicted marathon finish time better than body composition alone in recreational runners.. Most online predictors ignore this: they take a 10K time and extrapolate as if everyone has trained for the marathon. A defensible approach is to use the predictor’s output as an upper bound, then add 1–2 minutes per mile if weekly volume has been under 30 mpw, and another 30–60 seconds per mile if the longest run has been under 18 miles 7Midgley AW et al. “Training to enhance the physiological determinants of long-distance running performance.” Sports Medicine 37, no. 10 (2007): 857–880. Long-distance performance is strongly gated by sustained aerobic volume and specific long-run length, not by VO2max alone.. That correction is ugly, but it is closer to the real physiology than raw 10K extrapolation.
Heat, terrain, and pacing discipline: the 5–15 percent penalties
No marathon predictor accounts for race-day conditions, but their effect is measurable and large. Each 5°C rise above ~10–12°C (50–55°F) slows recreational marathon times by roughly 1–3 percent, with the penalty growing steeply above 20°C 8Ely MR et al. “Impact of weather on marathon-running performance.” Medicine and Science in Sports and Exercise 39, no. 3 (2007): 487–493. Marathon performance declined systematically with rising wet-bulb globe temperature across multiple races, with recreational runners more affected than elites.. Elevation gain adds roughly 10–15 seconds per mile for every 50 feet of net gain over the course, and altitude above ~1500 m costs additional seconds per mile in aerobic work 9Peronnet F et al. “Correlating the performance of elite marathon runners to altitude and terrain.” Journal of Applied Physiology 70, no. 1 (1991): 399–404. Marathon performance at altitude and on hilly courses slows in line with decreased oxygen availability and mechanical grade costs, with effects of 1–5 percent commonly reported.. Pacing discipline is the final lever — analyses of large marathon datasets show that runners with a second half slower than first by more than 8 percent systematically underperform their predicted time, while those within 4 percent overperform 10Angus SD. “Statistical timing analysis of running.” Journal of Quantitative Analysis in Sports 10, no. 4 (2014): 395–409. Positive-split patterns greater than ~8 percent in recreational marathons correlate with significantly worse finish times than even or slight-negative splits.. Any calculator output should be treated as a best-case estimate under even splits in cool weather on a flat course.
When the predictor is not the right tool at all
For first-time marathoners, the predictor’s output is almost always too optimistic because the underlying models are fitted to runners who have already completed at least one marathon. The honest advice for a debut is to set an A goal based on training long-run pace, a B goal 5–8 minutes slower, and a C goal at finishing, rather than reading a 3:28 number off a calculator because the 10K was 44:00 11Nielsen RO et al. “Training errors and running related injuries: a systematic review.” International Journal of Sports Physical Therapy 7, no. 1 (2012): 58–75. Novice marathoners display substantially higher variance in finish-time outcomes relative to predictions, driven by load-management and pacing errors more than raw fitness.. For masters runners, the predictor should be corrected downward by roughly 5–8 percent per decade past 40, reflecting declines in VO2max and tendon remodeling capacity 12Ganse B et al. “Endurance performance in masters runners: an update.” International Journal of Sports Medicine 42, no. 10 (2021): 889–895. Masters runners lose approximately 5–10 percent of VO2max per decade past 40, with tendon remodeling slowing and economy partially compensating.. And for anyone returning from illness, injury, or a multi-month layoff, a predictor built on recent short-distance racing is meaningless until the long-run/weekly-volume base has been rebuilt for at least 6–8 weeks.
How Does The Calculator Work?
Limitations of Traditional Prediction Methods
One big issue with the traditional race time prediction formulas, such as the Riegel formula, Cameron formula, and age-graded method, is that they tend to be based on the race times of professional runners, meaning they are often inaccurate when used to predict race times of recreational distance runners, especially when predicting marathon times.
For example, as of 2025, the world record time for the half-marathon is 56:42, and the world record for the marathon is 2:00:35, meaning that both races are run at a similar pace.
Yet, the average runner, who hasn’t had access to professional training, is more likely to run a marathon at a notably slower pace than a half-marathon.
Therefore, a formula developed from the race times of professional runners isn’t going to be particularly useful for predicting the marathon distance for the average runner, and formulas such as the Riegel formula tend to underpredict race times beyond the half-marathon distance by ten minutes or more.
Some data even suggests that only 5% of runners may actually beat the prediction given by the Reigel formula when predicting a marathon finish time based on a half-marathon time.
This is an issue if recreational runners base their training goals and pace targets on inaccurate predictions, resulting in potential overtraining, missed goals, and race-day disappointment.
This is where the more advanced marathon race time prediction method steps in.
The Vickers-Vertosick Method
Vickers and Vertosick utilized data from over 2,000 recreational runners to determine the most effective predictors of endurance running performance based on training and runner characteristics. It is unique in the fact that it focused on non-professional runners, whereas past studies tended to focus on elite runners.
They found that weekly training mileage was strongly associated with race time, and by incorporating this as a predictor, they were able to develop a set of equations for predicting race time significantly more accurately than the classic Riegel formula, particularly for longer distances.
The equations developed by Vickers and Vertosick are used in the above marathon time calculator.
Other Factors Affecting Marathon Race Time
No prediction calculator can be 100% accurate, so it’s important to consider that any of the following factors can also affect your race time:
- Weather Conditions – Running a marathon in hot conditions or other challenging weather conditions will likely result in a slower finish time.
- Cross-Training – Cross-training can help improve running power and reduce the risk of injury, resulting in better performance on race day.
- Interval Training – Runners who properly incorporate interval training into their routine can expect their run time to be approximately 3% faster.
Predicting your marathon time is never an exact science, but using a tool designed specifically for recreational runners gives you a far more realistic benchmark than traditional formulas.
Use this calculator as a guide to set smarter training goals, pace your long runs more effectively, and line up on race day with confidence. Remember—your training, recovery, and mindset will all play just as big a role as the numbers.
If you are looking for a training plan for your next marathon, check out our database:
