An attempt to predict the cutoff time for the 2025 Boston Marathon.
Like most runners aiming to qualify for the 2025 Boston Marathon, I would very much like to know what the cutoff time will be before toeing the line at my chosen qualification race.
But that’s impossible; no one will know the cutoff time until the BAA (Boston Athletic Association) announces it after the registration period once they know how many qualified applicants there are.
It is the conundrum of all but the fastest of Boston Marathon aspirants. Unlike many of the other major marathons, wherein the field size may grow depending on demand, year after year Boston’s field size is fixed at 30,000 runners. In actuality, only 22,000 to 24,000 of those entries are allocated to qualified runners.
If there are more applicants than available spots, the BAA imposes a cutoff time and denies entry to all applicants whose BQ* was not at least the cutoff time faster than their qualification standard.
We were lucky in 2022 and 2023. Because of lingering upheaval from the COVID-19 pandemic, the number of qualified applicants was less than the number of spots allocated to them. Hence there was no cutoff time imposed in either of those races and many, including myself, qualified handily.
That quickly changed. A record 33,058 runners applied for the 2024 race, a cutoff time of 5 minutes 29 seconds (5:29) was imposed, and 11,039 runners were denied entry. Only the fastest 22,019 of the qualified applicants will be racing on April 15, 2024.
What happened? Is the 2024 race an anomaly such that we can expect a shorter cutoff time to return in 2025?
Sadly, the answer is almost certainly “no”. The cutoff time for the 2025 race will get longer and more qualified applicants will be denied entry.
Indeed it is possible, in fact likely, that the BAA will make their standards, across all ages and genders, more aggressive by 5 minutes as they did after the 2019 race when rising demand for entry resulted in a similarly long cutoff time.
The following is my reasoning for these sobering predictions.
Any attempt to predict the state of the 2025 Boston Marathon entry process must first anticipate the number of qualified applicants there will be and then describe their mathematical distribution.
Qualified Applicants for the 2025 Boston Marathon:
We could have seen this coming. There are websites that track the results of all marathons and the number of BQ’s among all the finishers. One of them (marathonguide.com) each year tracks the top 50 marathons in North America (NA) for total number of BQ’s and percentage of BQ’s (BQ%) among its finishers.
I’ve looked at these data and noticed that, in years not preoccupied with pandemics, the total number of BQ’s from the top 50 NA marathons tracks closely to the number of Boston Marathon applicants. The number of Boston applicants is consistently on the order of 78–80% of the number of BQ’s in the top 50 marathons.
For the 2024 race, the top 50 NA marathons yielded 42,301 BQ’s and the number of Boston applicants was 33,058 (78.1%). That was a significant increase over 2023 when the top 50 NA marathons produced 29,533 BQ’s and there were a manageable 23,267 Boston applicants (78.8%).
The list of top 50 NA marathons changes slightly every year but the heavy hitters — such as Chicago, New York City, California International, Grandma’s, Philadelphia, Houston, and Boston itself — are always on the list.
Of course, the qualification period for 2025 is not over yet and 29 of the top feeder races from the 2024 qualification period have not yet happened for the 2025 qualification period.
Yet for those 21 marathons that have happened, 24,365 BQ’s were produced compared to the 18,679 BQ’s for the 2024 period — a 30% increase! And this is despite the fact that one of them, Twin Cities, which in the previous year had 513 BQ’s, was cancelled this year due to adverse weather.
It’s worth it to dive a bit deeper into these numbers.
One might think that the number of BQ’s is going up because there are more people running marathons now. That’s true in part.
But in those 21 aforementioned races there were 151,180 finishers in the 2024 qualification period and 169,434 for 2025, an increase of only 12% compared to the 30% increase in BQ’s.
We can thank the advent of the super shoe for this statistic. Before they arrived, about 11% of all runners in the top 50 NA marathons BQ’d. Now the number is over 15%.
Historically, the Boston Marathon, a race comprised of mostly qualified runners yielded 30–35% BQ’s. (It is a harder race to run than it is to enter.) However, in 2024 that percentage jumped to 52%.
That’s huge! It real terms over 8,000 more runners in the 2023 Boston Marathon BQ’d compared to 2022. True, the weather was conducive to a good run that day, but still, that’s a big change.
The effect of the super shoe is being felt throughout the running world. Sub 4-minute miles at the high school level have become commonplace and it won’t be long until the official world record for the marathon drops below 2 hours. I believe that will happen in Berlin or Chicago in 2024.
For the remainder of 2025 Boston Marathon qualification period, 29 of the top 50 races from the 2024 qualification have not yet happened. It’s anyone’s guess as to how they will go.
I like to bracket my expectations. At the low end, Boston will yield its historical 30–35% BQ% and the rest of the top 50 NA races will not keep up with the blistering pace we have seen to date. These assumptions suggest 33,990 applicants for the 2025 Boston Marathon, a modest increase over the 33,028 applicants for the 2024 race.
At the high end, the 30% increase in BQ rate will carry through the rest of the qualification period on the remaining top 50 NA races and there will be an astonishing 43,121 applicants for 2025.
I can also think of a midpoint compromise — Boston stays at the 51% BQ% but the rest of the feeder races perform more like they did the last time around. These assumptions yield 37,502 applicants for 2025, still a very big number.
The BAA most likely is looking at these data also and may be loath to announce another huge cutoff time. Therefore, I expect that there will be a 5 minute adjustment to the qualification standards before the 2025 qualification period is over.
However, for the modeling that follows, I will continue to use the standards that are currently in place.
2025 BQ Distribution:
This next part is trickier. I have my guesses as to the number of applicants and I am going to use the number of acceptances to the 2024 race (22,019) as the target for 2025 acceptances. But predicting the cutoff time requires an idea of how the BQ times are distributed mathematically.
The BAA gave us a hint in their announcement of the 2024 cutoff time:
During the registration period, the breakdown of accepted Qualifiers was as follows:
- 6,182 Qualifiers met their qualifying time by 20 minutes, 00 seconds or more.
- 8,858 Qualifiers met their qualifying time by between 10 minutes, 00 seconds and 19 minutes, 59 seconds.
- 6,979 Qualifiers met their qualifying time by between 05 minutes, 29 seconds and 9 minutes, 59 seconds.
- 11,039 qualifier applicants were unable to be accepted into the 128th Boston Marathon.
These numbers can be used to generate a model for the BQ time distribution. Although presented in a stairstep manner, the actual data is undoubtedly smoother and most likely normal (i.e. Gaussian).
In the below chart, I have fitted the 2024 BAA data to a normal distribution. As plotted, with “Runners per second” on the vertical axis and “Cutoff Margin” in seconds on the horizontal axis, the area under any portion of the curve is the number of runners for that particular range of cutoff times. (Note that the 2024 cutoff time of 5:29 appears as 329 on the horizontal axis.)
The fitting parameters for the distribution are amplitude, mean, and standard deviation. The parameters I have chosen match the BAA-supplied data buckets to within 1% up to 20 minutes (1200 seconds) of cutoff margin. I have not attempted to match the BAA data for cutoff margins greater than 20 minutes. Those fast runners aren’t the issue here.
How should I adjust the fitting parameters to account for the additional expected applicants to the 2025 race?
I think it unlikely that the spread (standard deviation) of the data will change much year over year although one could argue that the mean might as more runners plunk down the big bucks for super shoes. However, to keep it simple I only adjusted the amplitude to account for the percentage increase in expected applicants. Below is the corresponding model for the 2025 race.
Prediction the 2025 Boston Marathon Cutoff Time:
As previously reported, the area under the curves describes the number of runners in any given range of cutoff margins. I have tabulated the various cases in the chart below. The cutoff time is determined by the area required to match the number of denied applicants.
Given everything I have seen thus far in the data, I expect that the 2025 Boston Marathon will tend towards the higher end of my bracketed range. Thus, I think that the 2025 cutoff time will be between 7:04 and 8:43.
Obviously, there are opportunities for error in this prediction. I welcome debate and suggestions for improving the model.
I would love to see how well the model fits to the 2022 and 2023 qualification data but, for those zero cutoff time years, the BAA did not release any BQ distribution data. I will continue to track the data on the remaining feeder marathons and, of course, will review whatever data the BAA releases when they announce the 2025 cutoff time just to see how well the model performed and to determine if it can be used for 2026.
After the 2024 Boston Marathon concludes in April, I will revisit this model to include the data from the top qualifiers to date, update my estimate, and post the result to this blog.
Remember, this started with my desire to get into the 2025 Boston Marathon. It is clear to me that the cutoff time will be longer than the 5:29 of 2024 and, I believe, considerably longer.
I will train with the intent of finishing in 3:40:00, which gives 10 minutes of cutoff margin for my 3:50:00 qualification standard. That would be a PR for me and at age 63, PR’s don’t come easy. I have to include a lot more strength training in my workouts.
Failing this, I get some relief in 2026 when my qualification standard jumps up 15 minutes to 4:05:00. But that’s as it stands now. Super shoes will only get better. I expect that the “aging up” advantage will likely shrink as further adjustments are made to the standards.
Click here for an update to this cutoff prediction.
*BQ: A marathon finish time that meets the BAA’s qualification standard for a runner’s age and gender. As a male runner who will be 64 years old at the time of the 2025 Boston Marathon, my current qualification standard is 3 hours 50 minutes (3:50:00).
