The 2025 Tokyo Marathon: Three Interesting Data Insights (Plus More)

Feature photo by nakashi on Wikimedia Commons — CC BY-SA 2.0. Scenes from the 2024 Tokyo Marathon

This past weekend was the Tokyo Marathon — the first Abbott World Marathon Major of 2025.

Last year was an exciting one for the series:

• London, Berlin, Chicago, and New York all set record levels of participation.
• Boston set a new record for applicants, and both Boston and Chicago announced new qualifying times
• New York set a record for the largest marathon in the world.
• Ruth Chepngetich set a new world record in the women’s marathon at Chicago
• Sydney hosted the Age Group World Championships and officially joined the ranks of the Majors

So what does this year have in store us?

The 2024 Tokyo Marathon was exciting in its own right, with the winners of both the men’s and women’s races setting new course records. This year’s field was hyped up, and both of last year’s champions returned. Could they repeat?

And of course the weather wasn’t too cooperative. The weather at Tokyo is typically fairly mild – neither all that cold or all that hot – but this year was on the warmer side. I saw a lot of complaints about it on social media. Did it have an impact on finish times?

I collected the preliminary results from the 2025 Tokyo Marathon – which included the name, gender, nationality, finish time, and splits for each runner – and combined it with data that I’ve previously collected for earlier Tokyo Marathons going back to 2015.

I wrote up a full analysis in Runner’s Life on Medium. If you’re not a subscriber, you can go here to request a special link to get behind their paywall.

But in the meantime, let me share a couple insights below – but know that there is some more analysis and additional visuals in the full write-up.

Tokyo Is Increasingly an International Race

When the Tokyo Marathon joined the Abbott World Marathon Majors back in 2013, it was largely a regional race. Japanese runners made up the lion’s share of the field, along with other nearby Asian countries.

There were a smattering of other international runners, but they made up a small part of the 35,000 person field. Since then, however, things have been changing.

The visual below segments the field into three groups – Japanese runners, American runners, and everyone else. The United States has, pretty consistently, been the #2 source of runners at the race over the past ten years.

Back in 2015 – the earliest year I could get detailed demographic data – about 80% of the field came from Japan. There were only about 500 Americans, and a few thousand runners from the rest of the world.

Those numbers slowly shifted from 2015 to 2019. But the dynamics in 2019 weren’t all that different.

Then came COVID. When the dust cleared in 2023 and the race returned to its full glory, there were far more American runners and other international runners. And far fewer Japanese runners. That trend continued in 2024.

So what happened this year?

The number of Japanese runners dropped to about 19,000 – just over 52%. The number of American runners increased from 3,669 to 4,388 – a 20% increase. And the share from the rest of the world also grew by over 20%.

It probably doesn’t help that Japan’s population is aging and participation among younger runners in the United States is booming. But Japan isn’t alone in this. The share of German runners at Berlin has also been on the decline, and Australian runners will surely be outnumbered at Sydney this year.

For better or worse, this is a consequence of the growing interest in the Abbott World Marathon Majors.

The Gender Distribution Varies By Nationality

When you compare the Majors, Tokyo has by far the most male-dominated field. Only about 25% of the runners are women. That’s a much smaller share than the other races.

But when you look under the covers, there are significant differences in gender distribution depending on which group of runners you look at.

Here’s a visual that breaks the data down into the same three groups as above – Japanese runners, American runners, and other runners – and compares the number of men and women from last year to this year.

Year to year, the distribution is stable. There’s hardly any difference from 2024 to 2025.

But between the three groups, Japanese runners are about 80-20 men, while American runners are closer to 52-48. The rest of the world is in the middle, with a 70-30 split.

This is consistent with other data that I’ve looked at which shows that American runners and races tend to have a more balanced gender distribution than the rest of the world. But Japan sits at the other end of the extreme.

Did the Weather Impact Finish Times?

When I saw some of the first anecdotes on Threads and Facebook, it made it sound like the weather was terrible. Apparently they ran out of water towards the end – and that is definitely a problem.

But the conditions weren’t all that extreme, with temperatures only rising to the mid-60’s later in the day. The back of the pack likely suffered a bit more, but the first waves would have finished before the heat set in.

The winners did fairly well. They didn’t set new course records, but they ran quick times, as did the rest of the elite field. More on that in the full analysis.

But here’s a visual with the overall distribution of times for all runners – split by gender.

The field is divided into five minute buckets, with the label marking the left end of the bucket. So the line labeled 2:55 includes all runners who finished between 2:55:00 and 2:59:59.

You can use the dropdown menu to switch between men and women, and you can hover over an individual to see the cumulative total number of runners who finished faster than a specific time.

You can take some time to explore, but here’s the bottom line. The distribution of times between 2024 and 2025 were not very different. There are some minor differences here and there, but the heat did not have a significant impact on finish times across the board.

How Fast Is Harry Styles?

And now for the most important question that everyone has been talking about – is Harry Styles fast?

Apparently, he ran the race and finished in 3:24:07. For a few days, it’s all social media could talk about.

Some people hyped it up, shocked and surprised he was that fast. Add in the detail that his splits were quite even. And then there were the haters, who claimed this was no big deal.

I’m not here to wade into the debate – other than to present some cold, hard facts.

Using my age grade calculator for marathon performances, you can see how this performance ranks against other men in the 30-34 age group. It’s based on data from American races throughout 2024.

His time would put him at about the 80th percentile – better than 79.2% of other runners in his age group.

Objectively, that’s faster than average – which is around 4:05 for his age group. It’s also significantly slower than BQ – which would be 2:55 (plus the cutoff) for his age group.

You can decide whether you consider that fast or not. But as usual, I’m more interested in presenting the data.

Also, while his first and second half are only a second apart, his pace fluctuated a bit in the middle of the race. He sped up after the halfway point – splitting 23:26 and 23:16 for 25-30k and 30-35k respectively. And he slowed to 25:47 (his slowest split) for 35-40k.

Still … quite a respectable effort!

What Did You Think About the 2025 Tokyo Marathon?

Did you run the Tokyo Marathon this year? If so, what was your experience like? Or maybe you’ve run it before — what are you curious to know more about with this year’s race?

At some point, the Tokyo Marathon will release the official results with the ages of runners. When that happens, I’ll likely write a follow-up analysis to more closely examine the distribution of finish times — and the trends for runners who BQ.