This year, there were a record number of qualifiers who applied to run the Boston Marathon.
That led to a steep cutoff time of 6:51 for the 2025 race and new qualifying times for the 2026 race and beyond.
A lot of people, myself included, hoped that would bring the cutoff time for future races back down to earth. The optimistic view was that a 1-2 minute cutoff would still be necessary, and the more pessimistic view was that it might be in the 3-5 minute range. But a cutoff time in excess of five minutes seemed pretty far fetched.
While it’s still too early to make a firm prediction about what the cutoff time will be, the fall marathon season has drawn to a close and the results of many races are now available to start analyzing the data for possible trends.
The cutoff time is essentially a function of three things:
- The number of people who qualify for the Boston Marathon
- The number of qualified runners who apply to run the Boston Marathon
- The number of applicants who are actually accepted
If we assume that #2 and #3 stay relatively constant – that a similar percentage of qualifiers actually apply and that a similar number of those applicants are accepted – the biggest variable factor is #1.
I’ve collected a ton of data and I’m ready to start looking at how that first factor is or isn’t changing.
I’ve written up a thorough analysis and published it in Runner’s Life on Medium. If you’re not a Medium subscriber, you can request a link to read the full analysis here.
Otherwise, I’ve shared a synopsis below of some of the key data points and conclusions.
Want the latest projection for the 2026 Boston Marathon cutoff time? Take a look at this dashboard, which is regularly updated with the latest race results.
What Data Are We Looking At?
Similar to last year, I’ve collected the results of marathons in the United States and Canada that had 200 or more finishers. I’ve also included the results of Berlin.
For now, I’ve limited this to races that have taken place from the beginning of the qualifying period through the Philadelphia Marathon. I narrowed this year’s results down to races that have taken place between 9/1/2024 and 11/24/2024, and then I matched that with last year’s races that have taken place between 9/1/2023 and 11/19/2023.
That’s about 100 races in each qualifying period. Note that a handful of smaller races did not have results available yet when I collected the results. But this represents a negligible amount of potential qualifiers.
For each individual, I used their age and finish time to calculate whether they met both the older (2020) qualifying standards and the newer (2026) qualifying standards. I also calculated the individual buffer for each runner.
The data was sourced from a combination of Athlinks, Marathon Guide, and individual race websites.
How Many Qualifiers Are There So Far this Year?
The visual below shows three things – the total number of finishers, the total number of runners who met the older qualifying times, and the total number of runners who met the newer qualifying times.
Before we talk about the number of qualifiers, look at the bar indicating the total number of finishers.
At this time last year, there were approximately 245,000 finishers. This year, there are 285,000. The total number of finishers – across all races – is up by a lot. The number of finishers in 2023 had already increased significantly over 2022, and this is one of the main drivers of the increase in qualifiers last year.
Now compare the two purple bars. At this point last year, there were about 33,000 qualifiers. If the qualifying times hadn’t changed, there would be 39,000 qualifiers so far this year. So if the qualifying times had not changed, the cutoff time would likely be getting much, much worse.
Finally, look at the blue bar in 2024. When you apply the new qualifying times, the number of qualifiers so far this year drops to just over 31,000.
So that is about a 5% drop in the actual number of qualifiers. But that decrease is much smaller than I had expected, and if things continue in this direction the cutoff time will be less than 6:51 … but not by a ton.
Which Races Are Growing?
A reasonable follow up question would be – which races are seeing increases in the number of finishers?
Most of the Majors have seen big increases, especially Berlin. But it goes beyond that.
Here’s a list of the top twenty races by the number of additional finishers.
Berlin is at the top of the list, with more than 10,000 additional finishers.
Twin Cities is next. It isn’t “growing” per se, but the race was cancelled last year. So there are an additional 6,700 finishers from that race in this year’s qualifying period.
Some of the other large races – New York, Chicago, Marine Corp, Philly – also saw hefty increases.
Notably, Las Vegas is a brand new race. This was its inaugural year, and it had over 2,000 finishers.
As you go down the list, the absolute numbers get smaller. But many races still saw 10-20% (or more) increases.
So a lot of the additional finishers are concentrated among the larger races. But across the board, races are increasing in size.
Has the Percentage of Runners Qualifying Changed?
There was a general assumption that the new qualifying times would reduce the percentage of runners who met their qualifying times. Is this, in fact, true?
The visual below divides the field into four broad age groups – under 35, 35-49, 50-59, and 60+. It’s also segmented by gender.
The left half of each visual indicates the percent of those runners who met the old qualifying times and the half on the right indicates the percent of runners who met the new qualifying times.
For the younger age groups, there is indeed a pretty large drop in qualification rates.
Young men drop about three percentage points, and at 7.36% they have the lowest qualification rate overall. Young women also drop a little over three percentage points, but their qualification is a little higher.
When you compare the other age groups, there’s a similar drop of around three percentage points between the old standards and the new ones.
The one exception is for runners 60+. Their qualifying times stayed the same – so the qualification rate remains unchanged. They continued to qualify at the highest rate – just under 20% for men and over 20% for women.
If you use the dropdown menu to toggle between 2023 and 2024, you can also see if the field as a whole is faster this year. And across the board, there is a small increase – but it’s about one percentage point (for the older age groups) or less (for the younger age groups).
Over time, this could change further. With the goal posts moved, people will train towards these new times, and the percent of people getting there could continue to increase. But at least in the short term, the distribution of finish times hasn’t changed drastically, and there hasn’t been a surge in the number of people hitting – for example – 2:55 vs 3:00.
So across the board, runners are slightly faster – but the tougher qualifying times are reducing the percentage of runners that actually qualify. If it weren’t for the large increase in the total number of runners, the number of qualifiers would have dropped much further.
What Does the Distribution of Buffers Look Like?
One final thing to look at is what the distribution of buffers looks like. This will become more relevant later when it’s time to estimate exactly how many applicants there are – and how many need to be cut.
The histogram shows the number of qualifiers, in one minute bins, from 20 minutes under their BQ through 5 minutes above their BQ.
The label for the bin indicates the left edge of the bin. So -20 includes anyone with a buffer of -19:01 to -20:00.
The bar labeled 0 is the dividing line between qualifiers and non-qualifiers. The zero bar and everything to the right is everyone who just missed out on qualifying. The remainder of the bars to the left is the number of people who did qualify.
There’s an interesting surge at -1 to -2 – likely people who were targeting a BQ time and came in just under it. Similarly, there’s another little surge at -6 and -7, likely people who were aiming for BQ-5.
Any time you look at distributions of finish times, there tend to be little clusters around specific goal times like this.
Otherwise, in general, there’s a fairly regular decline in the size of each bin – from the right end to the left end. It’s not the case that a ton of people just barely hit their BQ’s (BQ-1) and very few that had bigger buffers. Instead, it steadily and slowly declines as you go further to the left.
So What’s the Bottom Line?
Again, it’s early to give a prediction with any real confidence. But it’s not too early to point out the general trend and draw conclusions from that.
So far, we have data on a hundred races and ~285,000 finishers. That’s close to half of the races and finishers that will run a marathon between now and September 2025. So there’s a lot of hay in the barn, so to speak.
This data points to a) large increases in the number of finishers and b) moderate declines in the qualification rates that lead to c) minor reductions in the total number of qualifiers.
If you want to convert that general statement into a number, let’s start with these assumptions:
- A similar percentage of qualifiers apply to run Boston
- Therefore, the increase or decrease in qualifiers is equivalent to the increase or decrease in the number of applicants
- Based on last year’s data, each minute of the cutoff time is equivalent to ~1,800 applicants (from 0-10 minutes).
- The number of accepted applicants will once again be 24,000 runners
Now, plug in the numbers we know:
- Last year, there were 36,393 qualified applicants.
- The number of qualifiers (and therefore applicants) is on pace to decrease by about 5%, which would yield ~34,500 applicants.
- That would require the elimination of ~10,500 applicants
- That would require a cutoff time of about 5:50
There are still some unknowns and some other variables, so things could change between now and September 2026. So I wouldn’t fixate on the actual time.
You could also quibble with some of these assumptions, as they do over simplify things. But they’re sound enough to get us in the ballpark.
I think this does suggest that the over/under on a likely cutoff time is around five minutes. It’s possible that it’s under five minutes, but probably not by much. It’s also possible that it could be over five minutes, but probably not by much.
And the optimistic view – that the new cutoff time will only be 1-2 minutes – is becoming increasingly unlikely.
What Is Still Uncertain?
Then, of course, there are some things that we don’t know – and that could change.
One variable is the total number of finishers. I think there is a long term, sustained increase going on, and the data shows this surge is particularly concentrated among younger runners. But at some point, it’s got to level out … right? So maybe the spring races continue to be much larger than last year’s – or maybe they start to flatten out a bit.
Another variable is race day conditions. There are a couple very large races yet – including Boston and London – which each contribute a large portion of the qualifying field. The weather on those two days could have a big impact, one way or another, on the number of qualifiers. Although it’s worth keeping in mind that last year’s Boston had bad weather – and a low rate of re-qualification. So it’s unlikely things get worse than last year.
A third variable is the number of accepted applicants. Last year saw an increase to 24,000, which is among the highest amount of accepted applicants in the last decade. Unless BAA increases the overall field size – which they planned on doing in 2020 – it’s unlikely this increases further. But it could.
Finally, I’m not sure if I should include the results of the Sydney Marathon. Both the 2024 and 2025 Sydney Marathons are in the qualifying period for Boston 2026, while the 2024 race fell just outside the qualifying period for 2025. It has a large (and growing) number of finishers, and they likely have an interest in the Majors. The 2024 race also included the Abbott Age Group World Championships.
I think it’s best to keep that separate – but at the end of the day to factor it in as a potential wildcard.
What Other Data Is In the Analysis?
Remember, this is just part of the full analysis. Some other questions I explored with the data:
- Which races saw the most additional qualifiers?
- Which age groups saw the greatest increase in finishers?
- Did the new qualifying times change the age distribution of the qualifying field?
If you’re not a Medium subscriber, you can request a link to the full analysis here.
Over the next few months, additional data will make the situation more clear. I plan to publish an update to this article every month or two through next September.
To keep up to date with how the qualifying field is shaping up, fill out the form below to subscribe to my weekly newsletter.
Is any of your data able to incorporate how many runners have multiple qualifying times within a given qualifying year?
For example, the same runner runs two qualifying times in a year. Or perhaps the number of these runners is insignificant to the overall data.
Yes, I can screen out some of the duplicates using a runner’s name, age, and, gender. It’s not perfect, but it identifies a good amount of the duplicates. At this point, I haven’t done that, because the number of people that have run two marathons within the months of September, October, and November is negligible. But when spring comes around, I’ll deduplicate the data and see how that effects that prediction.
I’m surprised you didn’t wait until after CIM to do your analysis but after seeing the fast times at CIM, it only would have made your hypothesis stronger.
I had actually planned on getting this initial analysis out a little earlier – closer to the Philly weekend. But life got in the way for a couple of weeks.
I’m working on a follow up that should be out this weekend that includes CIM and the other December races.
Do you take into account runners who are aging up? For example I was 68 when I ran Wineglass in 2024 but qualified as a 70 year old for 2026.
Not currently, no.
The data on this gets a little messy. For some runners (those who complete a marathon more than a year before Boston), it’s possible to know for certain that they’ll age up. But for those who run a race later in the year (say May), it’s impossible to know for sure whether they’ll age up or not.
And then there are some races (Chicago is a big one) that only release age group data and not individual ages. So in this case, there’s no way to know if someone is 65 or 69.
So there’s a general assumption that there will be more qualifiers than I’ve counted in the analysis – but as long as the methodology is the same for each year that should wash out.
Down the road, I’m going to work on an algorithm that can identify the people who can be positively aged up, and age them up. But I’ll never be able to account for 100% of them.