Running calculators · Steady State Health
Enter a marathon you ran and your finish time to see the equivalent time at Boston. The conversion comes from thousands of runners who ran both races.
Enter your race and finish time to see your Boston equivalent.
Runners in our data ran this race about % slower than they ran most other marathons. Hills, altitude, trail sections or a lot of people running it for fun can all cause that, so expect a less accurate conversion.
| Method | Boston time | Median error |
|---|
How it works
Most course converters estimate difficulty from a course’s hills and its typical weather. By that measure Boston should be fast, since it drops more than 400 feet from Hopkinton to Boston. Anyone who has run it knows the early downhills beat up your quads before the Newton hills, and the weather on race day changes a lot from year to year.
We built this calculator from race results. We compared the same runners across every marathon they ran between September 2022 and September 2026: finishes from runners who raced more than once. Because we compare each runner with themselves, we can see how much faster or slower each course and each race day was. There are two ways to use that data:
We also correct for qualifying races. Runners in our data ran their qualifying races about 2% faster than their usual times, so comparing a qualifier straight to Boston would make Boston look harder than it is.
Accuracy
To test the calculator, we set aside one in five Boston runners, rebuilt everything without them, and predicted each of their Boston times from one other race they ran. The table shows the median error: half of the predictions were closer to the runner’s actual Boston time than this, and half were further off.
| Method | Exact race and Boston year | Any year and Boston average |
|---|---|---|
| No conversion (assume the same time) | n/a | |
| Runners who ran both | ||
| Runner-ability model | ||
| Blend |
One runner’s marathon times vary by about 4 to 5% from race to race, because training, health, pacing and weather all change, so any conversion, however good, will still be off by about that much. That is why the calculator shows a range as well as one time. In the same test, 50% of runners finished within the narrower range and 80% finished within the wider one.
Choosing the exact race date and Boston year gives the closest estimates, because we know the weather on both days. Any year and the Boston average use more runners, but they average across different weather, so their errors are a little larger. Both are still closer than assuming your Boston time would match your other race.
Predicting a future Boston. The Boston average combines a warm year (2024), a mild one (2025) and an ideal one (2026), so it shows how your time compares with recent Boston conditions. We also tried predicting Boston 2026 from its weather alone, using only earlier races, and those predictions were no better than assuming the same time. Boston 2026 had a strong tailwind and ran about 1% faster than we predicted from its temperature, and three years of Boston results are too few for us to measure the effect of wind. We will test again after Boston 2027.
Credit
Race results come from Brian Rock’s Boston Marathon Cutoff Prediction Dataset (MIT License), the data behind his Boston cutoff tracker. Race-morning weather comes from NOAA’s hourly station records (public domain), and race locations come from GeoNames (CC BY 4.0).
Steady State Health is not affiliated with the Boston Athletic Association or the Boston Marathon. This page shows statistics for each race and no individual runner’s results. Use the estimates for training and goal setting. They have no bearing on Boston qualifying standards.
Steady State Health is a performance-based physical therapy and coaching clinic for runners. We’re based in Portland, Maine, and we work with runners around the world through virtual sessions. If an injury is getting in the way of your Boston training, we can help.
Virtual running physical therapy More running tools