My last major update on the Boston Marathon cutoff time projection came at the end of March. The projection was 5:20, and things had been moderating slightly for several weeks. But as I said then, April could prove to be an impactful month, with results from both Boston and London, along with many other races. And April did not disappoint.

Boston had a particularly fast year, and the number of qualifiers surged compared to last year. That was followed by a warm day at London, pushing things back down. And then there were 60,000 other results.

The other big change I've been hinting at for a while is that I finally introduced a new methodology into my tracker to account for runners who completed multiple marathons in a given qualifying period.

So, how have things changed in the last month? And what does the future of the Boston Marathon's cutoff time look like now?

Let's dive in and find out.

The Data and the Methodology

I've been tracking race results throughout the year and incorporating them into a dataset for this dashboard that projects an estimated cutoff time.

I include races in the United States and Canada with 200 or more finishers, along with the London and Berlin Marathons. To date, that includes 213 races with approximately 559,000 finishers in the qualifying period for the 2026 Boston Marathon. That corresponds to 218 races and 497,000 finishers from the same time period last year.

To group the results by runner, I started by estimating a potential range of birthdates for each result. Then, I used fuzzy matching to identify runners with the same (or similar) names, who were of the same gender, from the same country, and who had overlapping potential birthdates. For runners from the United States, I also checked to see if their state differed as a way to weed out false positives.

While this methodology will miss some actual duplicates and identify a handful of false positives, it's a good approximation. Most importantly, the same methodology was applied to both qualifying periods, and the biggest question we need to answer is the delta between last year's qualifying period and this year's qualifying period.

This took quite a bit of work to get set up. But now that I've created the workflow, it'll be easy to incorporate new results and continue to process them in the same way moving forward.

What Happened in April?

It's an understatement to say that the month of April, plus the first weekend in May, was busy. That time period included 50 races in my sample, totaling over 145,000 individual finishers.

The first big news came from Boston.

The weather last year was warm; this year, it was nearly ideal. The field was also slightly larger, and due to last year's cutoff time, runners were slightly faster.

Due to these combined factors, this year's Boston Marathon saw one of the highest levels of re-qualification in recent history, despite the new qualifying times. Vis-à-vis last year, which is the most important thing for determining this year's cutoff, there was a net increase of about 3,000 qualifiers.

You can read more about the results from Boston here. But as a result, the cutoff projection shot up to well over 6:00.

The following week was the London Marathon.

The field grew by a modest amount, but the weather this year was warm. Not terrible, but enough to nudge the qualification rate down several percent. Combined with the tougher qualifying times, the net result was that the number of qualifiers at London dropped by about 2,700, vis-à-vis last year.

You can read more about the results from London here. But as a result, the cutoff projection dropped back down below 6:00, but it remained well above where it had been at the end of March.

And then there was the rest of the busy slate of races. Many saw double-digit increases in finishers. Vancouver, Eugene, Jersey City, and REVEL Mt. Charleston all saw big increases. If you take out Boston and London, the remaining races saw an across-the-board increase of 6–7%, inclusive of the cancellation of the Carmel Marathon and the loss of 1,400 finishers.

When you combine everything — Boston, London, and the remaining races — this period saw a decrease of less than 500 qualifiers compared to last year. That's less than 2%.

Prior to April, the year-to-date drop had been almost 8%. Which means that the net result of April's races is to a) push the cutoff time higher and b) lock in the likelihood of a high cutoff time.

How Many Runners Complete Multiple Marathons?

Since I've been tracking results and projecting a cutoff time, people have asked whether I'm counting all finishers or only unique finishers.

Until this point, I've counted all individual finishers. For one thing, it's simpler. For another, it's safe to assume that a similar percentage of runners complete multiple marathons in a given year. So as long as we apply the same methodology to both years, it shouldn't matter much.

But some people are convinced that it makes a big difference. And it does represent a possible variable that we can minimize. Therefore, it makes sense to investigate.

So, how many people actually run multiple marathons in a given year?

The graph above shows the number of runners in each qualifying period who completed 1, 2, 3, 4, or 5 races in that span of time.

Note that for this specific visual, I've eliminated the London and Berlin Marathons. Since I don't track other European races, I likely missed more duplicates there. By focusing on North America, this ensures the greatest number of duplicate results are identified.

And the vast majority of runners complete only one race per year. I think it's easy to get caught up in what you see on social media. But when you look out in the real world, this passes the smell test for me — that most runners complete one marathon in a given year.

There are a decent number of runners who participate in multiple marathons per year. Around 20,000 to 25,000 run two races, and a few thousand more do three. From there, the numbers drop off sharply.

So, Does This Impact the Number of Qualifiers?

Given the fact that most runners only do one race in a year, I wouldn't expect there to be a huge impact from focusing only on a runner's best time.

But let's see just how much difference de-duplicating the results makes.

First, here are the stats for the total number of finishers.

When you look at all results, the number of finishers has increased from around 500k to 560k — roughly 12.5%.

When you zero in on unique results, the number of finishers last year drops to 450k. It goes down by about 10%. But the net increase from last year to this year is just under 60k — roughly 12.8%.

So the delta between last year and this year is consistent, whether you look at raw results or unique finishes.

It is possible that runners who are aiming to qualify for Boston are more likely to run multiple races. So let's take this a step further and drill down on the number of qualifiers.

Last year, there were 67,650 individual qualifying results, and that drops to 59,785 if you only count a runner's fastest time. That's a drop of 11.6%, so it is true that the drop-off among qualifiers is higher than among runners in general.

But when you look at the drop off among runners in the 2026 qualifying period, it's just about the same — 11.3%. Again, the delta is consistent … so the end result will also be consistent.

If you look at raw results, the net drop in qualifiers this year is 5.53%. If you focus on unique results, that drop is 5.22%.

No matter how you slice it, there are still a similar number of qualifiers this year, despite the new qualifying times. And that means the cutoff will remain high.

Are There Really That Many More Qualifiers?

At this point, it's looking incredibly likely that the cutoff time will be above 5:00. The data has been pointing that way since the fall, and with each passing month, there are fewer and fewer possibilities for that to change. Still, I frequently see people who are somewhere between skeptical and incredulous that things could change so much in a year.

It's not that they don't believe the data and the math, per se. They just find it hard to believe. A common line is: "So if the qualifying times didn't change, would there really have been a ten-minute cutoff?"

And, well … yeah. There would have.

The visual above slices the data up in three ways. In all cases, this is looking at unique qualifiers — based on their best qualifying time, not the raw number of results.

At the top, you've got the total number of qualifiers. If the qualifying times stayed the same, that number would go up 16%. If that change carried through to the number of applicants, there would be over 42,000 applicants this year.

That would require 18,000 people to get cut, and that means a huge cutoff. That's almost a 50% increase in the number of applicants who would need to get cut.

But then, some people still assume that there are a lot of people making it in just under their qualifying times. The second row in the data shows the number of qualifiers who met last year's qualifying time, plus the 6:51 cutoff.

The totals are (obviously) lower. But the increase from last year to this year would have been 21.5%. So if that cutoff reduced things to 24,000 applicants last year, it would have left the pool at 29,000 this year, requiring a deeper cut for those 5,000 additional applicants.

And finally, the last bar shows the number of qualifiers who were 5:00 under their qualifying time for older runners and 10:00 under for younger runners. Essentially, these are people who would meet the new qualifying times (which didn't change for runners 60+), plus a five-minute cutoff.

This cuts down the total numbers compared to the previous scenario. But the number of people meeting that bar is 20% higher this year than it was last year. And most importantly, the total for this year — 42.7k — is still higher than the number of qualifiers who met last year's qualifying times with the 6:51 cutoff.

In other words, this year's qualifying times with a 5-minute cutoff get you close, but it doesn't get you all of the way there.

So, yes. No matter how you cut it, the number of runners meeting their qualifying times is much higher this year than it was last year. And that can only mean one thing — a deep cut.

Can Things Still Change Before September?

How much of this is set in stone, and how much can still change?

Most of it. And not a lot. Or at least not because of new results.

Through this point last year, there were around 500,000 finishers in the sample. If you add up all of the races remaining through the end of the qualifying period, there are only about 55,000 finishers remaining.

In other words, about 90% of the results are in. Ten percent are unknown.

Absent any drastic changes — like the last-minute cancellation of several big races — there's just not enough wiggle room in there to change the trajectory of things. At this juncture, all signs are pointing to a high cutoff. The remaining races could nudge things up or down a little bit. But there aren't enough remaining results to meaningfully change the trajectory.

The fundamentals driving the change are sound, and they've been consistent all year. There are more finishers, and this is leading to more qualifiers. And that's not about to change.

The cutoff time may not be exactly 5:49 — the current projection. But I have little doubt that that's in the ballpark.

So What Is Still Uncertain?

There are still a few variables that could influence what the actual cutoff will be. The window is fixed, with a likely range of 5:00 to 7:00, but a few things could influence whether things end up on the high side or the low side. Some of those are just unknown.

For example, how many of these qualifiers will actually apply? Data released by BAA in previous years about how many people fall into each bucket of buffers offers some insight.

But the dynamics could get shuffled this year. On the one hand, there's a perception that it's getting tougher to get in, which could encourage a greater conversion rate. There's also increasing demand across the board for the Majors, which would suggest more applicants than usual.

On the other hand, there are some rumblings that international runners — Canadians in particular — will sit this one out. Based on the results of this year's Boston, this doesn't seem likely to have a huge impact. But maybe it's enough to slightly soften demand.

There's also the possibility that people who have already qualified will pursue last-chance races, including downhill ones like the upcoming tunnel marathons, in an effort to increase their buffer.

That wouldn't change the total numbers much, but it could change the distribution of buffers. And if more of the qualifiers have large buffers, the race will fill up quicker.

One way to look at the cutoff time is from right to left. You get x qualifiers, and you push the cutoff time down to get to 24,000 seats. And here, each minute of cutoff represents a certain number of applicants eliminated.

But the other way to look at things is from left to right. You're starting with the fastest runners and filling up the applicant pool. If relatively more people have 10–15 minute buffers, then that pool will fill up more quickly than it has in the past.

Can the Model Be Improved?

In terms of the core methodology, the de-duplication of results that I incorporated into the dashboard this week is the last big change. There are, however, some minor improvements that could be made to narrow down the precision.

In particular, there are two related critiques that I often hear, and they have some merit. They don't undermine the fundamentals, of which I'm pretty confident, but they do offer an opportunity for refinement.

First, I primarily track North American races. In part, this is due to the availability of results. But it's also due to the likelihood of runners actually running Boston.

Incorporating the full results from races like Tokyo or Sydney (or some of the other large European races) could skew things because a small number of those qualifiers will actually convert. But some do convert, and that could be meaningful in narrowing things down.

Likewise, not everyone at different races throughout the United States is as likely to apply. A larger proportion of people who run downhill races are likely aiming for a BQ and are likely to convert their qualifying time into an application.

Both of these things can be addressed by looking at data over the past two years to identify who did and who didn't apply to Boston.

Before the summer is out, I plan to gather some additional data for large European races, match these results up to the runners at the 2023 and 2024 Boston marathons, and do some fancy statistical magic to determine which factors indicate whether someone is more or less likely to apply.

The result would be a system of weighting individual results to better predict whether someone would or wouldn't apply. And that would ultimately lead to a more precise prediction of the actual number of applicants and the cutoff time.

This will not be incorporated into the cutoff tracking dashboard, but I'll write up a more thorough analysis to share my findings. I aim to have that done sometime in July.

So, How Should You Be Thinking About the Cutoff Time for the 2026 Boston Marathon?

Put all the words and fancy stats aside. What's the most likely outcome — and how should you be thinking about the cutoff? Really, that's what matters to people. And the most common question I get is, "I have a x:yz buffer. Do I have a chance?"

At this point, I'd be very confident in saying that the cutoff time will be in the broader range of 5:00 to 7:00, with a likely outcome that's closer to 5:30 to 6:30.

There's a slim chance that it drops below 5:00, but almost no chance that it ends up below 4:00. If your buffer is lower than 5:00 and you really want to run Boston in 2026, I'd start looking for an early fall race.

On the other side of things, there's almost no chance that the cutoff time will be above 8:00. So 8+ is almost certainly safe. There's a slim chance — but not a likely one — that it slips above 7:00. But I'd feel pretty comfortable if I had a 7+ buffer.

If you're in the 5:00 to 7:00 range, then you're gambling. The higher you are, the better your chances. A cutoff time in the mid-6:00s will likely get in. But there's a meaningful chance that it's not quite enough.

As the summer progresses, I'll continue to upload results to the Boston Marathon Cutoff Time Tracker on a weekly basis. I plan to publish a more refined analysis, based on some more complex modeling in July. And once the results are locked in, come September, I'll publish a final look back over the full qualifying period.

A number of people have also expressed interest in the raw data to conduct their own analysis. I intend to share the full dataset, along with updates, on Kaggle. I hope to release that by the end of this month.

For more frequent updates, follow me on Threads. And if you don't want to miss that analysis, make sure you subscribe to email updates.

I'm an avid runner and a data nerd. I've qualified — but missed the cutoff — for Boston twice. I fell short at Jersey City, and I'm considering a return to Erie. Here's how you can follow along with what I'm up to: