What’s the Impact of Downhill Races on the Boston Qualifying Cutoff Time?

If you want to start a heated discussion with a group of runners, ask them how they feel about downhill races. To some, they’re tantamount to cheating — an easy way to boost your time by 5 or 10 minutes (or more), claim a new PR, and qualify for Boston.

To others, they’re a perfectly legitimate way to achieve a goal time. There are lots of ways to gain an advantage. So if downhill courses are allowed by the rules — and they are certifiable by the USATF and allowed by BAA for Boston qualifying purposes — there’s nothing amiss.

I’m not here to wade into the arguments of right or wrong, “cheating” or not.

What I am interested in is how much of an impact these downhill races have on the Boston qualifying field. Whenever I share something about the cutoff time, somebody complains about downhill races, implying that if they weren’t allowed, the cutoff time would not be so bad.

I’ve written up a longer analysis on the topic in Runner’s Life on Medium. If you’re not a Medium subscriber, you can request a special link to read the full analysis here.

In the meantime, I’ll share the highlights below, including:

• Where specific races fall in the spectrum of flat to downhill
• How many runners BQ at downhill races – and end up running Boston
• How much net drop impacts race times

If you’ve run a downhill marathon, I’d love to hear your take in the comments. And if you’re left wanting more – I’d encourage you to read the full analysis on Medium.

How Steep Are Downhill Races?

On it’s face, it’s easy to define a “downhill race.” It’s one where the finish line has a lower elevation than the start line.

World Athletics rules allows a net drop of up to 1 m/km for record eligible courses – and anything below that is essentially flat. Anything beyond that is downhill.

But there are most definitely different shades to that definition. Boston is net downhill, but it’s not an easy course. Compared to the flatter majors, runners are consistently slower at Boston.

A race like REVEL Big Cottonwood, on the other hand, is unashamedly downhill. Of all the USATF certified courses, it has the largest net drop – 37.82 m/km.

To put that in perspective, I recently ran the United NYC Half Marathon. After climbing through Prospect Park, you’re rewarded with a steep decline in mile three. I felt like I was flying in that mile – effortlessly floating downhill.

Conner Mantz – who went on to finish second in the race and would have set a new American record had the course been record eligible – ran 4:30/mi for much of the race. He split 4:10 in that mile.

Over the course of that mile, Flatbush Ave descends about 124 feet – equivalent to 23.5 m/km. So REVEL Big Cottonwood is far steeper than that … over the course of the entire race.

To get a feel for just how far apart some of these races are, I took a look at the list of USATF certified courses. The USATF database includes a net drop for every certified course. I plotted the courses below, with the steepest courses to the right.

I also broke them out into categories, according to the following criteria:

• < 1 m/km — Flat
• 1 to 5 m/km — Minor
• 5 to 10 m/km — Moderate
• 10+ m/km — Steep

The vast majority of races are flat. I had to limit this visual to only including flat races with over 1,000 finishers, because otherwise the rest of the visual was unreadable.

One course on the list was net uphill – the Missoula Marathon.

There’s a sizable cluster of races with a slight net decline – including Boston and CIM. And there’s a smaller group with a moderate decline of 5-10 m/km.

There are relatively few races with a net drop in excess of 10 m/km. St. George is the biggest. There’s a group of races with 10-20 m/km drops, and then there’s a huge gap before you get to the steepest races – four REVEL races and Mt. Nebo.

When people complain about downhill races and suggest that BAA bans them for qualifying purposes, the common retort is, “Boston is downhill, too. Should we ban Boston?”

If you didn’t already know how disingenuous that response was, this visual makes it crystal clear. There are several obvious places you could draw a dividing line – 5 m/km, 10 m/km, 20 m/km – if you wanted to. And the REVEL races are so much steeper than the rest, that they are clearly outliers … even compared to other downhill races like St. George.

How Many People Qualify – and Run – Boston From Downhill Races?

Although the previous visual clearly establishes that there is a small subset of races that are very steep, the corollary to that is that there are also relatively few finishers at these races.

On the other hand, these races are heavily marketed to people aiming to qualify for Boston. So despite making up a relatively small portion of the running population, they could still produce a decent amount of Boston qualifiers.

So how many do they produce – and how many end up running Boston?

The full analysis goes into this in more detail. About 3.5% of all marathon finishes (in the United States) take place on a downhill course with 10+ m/km of net drop.

Since these courses are downhill, runners are much more likely to qualify. But their numbers are still small. When you look at the number of Boston qualifiers across the country, about 6.7% of them come from steep downhill races.

To fully play this out, I matched the results from the qualifying period for the 2025 Boston Marathon to the entry list that has been posted for the race. I was able to match 16,765 finishers from other races to the Boston entry list.

Of those 16,765 Boston runners, 1,919 of them qualified at steep downhill courses. That’s a much larger proportion than their share of the overall field, but it’s still only about 12%.

Left unmatched are around 7,000 other Boston runners. Many of these likely qualified at international races that I don’t track, while some are just unmatched because of irregularities in the data.

If you assume all of them qualified at flat races, that would drop the share of downhill runners to 8%. So I think it’s a fair conclusion that between 8% and 12% of runners at Boston (in 2025) qualified via a steep downhill course. And the number of downhill runner at Boston is likely between 2,000 and 2,500.

The final step in this thought process, though, is how many of these runners would still qualify and make the cutoff if they ran a different race. About 40% of these runners had a buffer in excess of 15 minutes, so it’s likely that most of them could qualify elsewhere.

If you make generous assumptions, then, somewhere between 1,000 to 1,500 runners made it in to Boston through a downhill race – and wouldn’t qualify otherwise.

Is this significant? Yes.

But in the scheme of things, it likely wouldn’t move the needle on the cutoff time by more than a minute.

In the context of last year, that means the cutoff still would have been in the realm of 5:45 to 6:00. In the context of this year, the projected cutoff time would still be in excess of 4:00 – but probably below 5:00.

How Much Faster Are Downhill Races?

The last part of this question is thinking through just how much faster downhill races are than flat races.

I recently saw someone argue that these races aren’t inherently faster. And, well, that’s presposterous.

They cited an article in which the author compared some downhill races to some flat races – based on the average times and the number of runners who met specific benchmark times (i.e. 2:30, 2:45, 3:00).

That’s all well and good for trying to figure out which race attracts the fastest field. But it doesn’t tell you anything about the course.

To get a sense for how much faster one of these downhill courses is, you need to compare results from the same runner on multiple courses. Ideally, you could establish a baseline – and see how different courses compare to that baseline.

As luck would have it, we have a nice group of runners who have run multiple races in a short period time and who have all completed one race in common – the finishers of the 2024 Boston Marathon. I was able to match about 14,000 runners from that race to their qualifying times, and about 1,300 of those results were from steep downhill races.

The Boston Marathon course is tough, given its hills, so you’d expect to see a small difference between a runner’s qualifying time on a flat course and their performance at Boston. If steep downhill courses are easier – and help equivalent runners complete faster marathons – then you’d expect them to have a bigger difference between their qualifying times at Boston.

The visual below breaks the runners into six demographic groups – men, women; under 40, 40-59, 60+. The colors represent runners on six different types of courses. The Boston group qualified at the 2023 Boston Marathon, and I’ve added a “Very Steep” group to differentiate races with > 25 m/km of drop.

The dot represents the average (mean) difference for those runners between their qualifying time and their time at the 2024 Boston Marathon.

Keep in mind that the 2024 Boston Marathon was hot, so you’d expect everyone to have had a bad day. It’s not surprising, then, that every group saw an average 10+ minute difference between their qualifying times and Boston.

What we’re really interested is the difference between runners who qualified at flat races and the runners who qualified at steep downhill races.

And that difference is stark. The yellow dots – runners at REVEL races – are far to the right for every group. For younger runners, you’re talking about a 10-15 minute difference. For older runners, it’s closer to 20.

Runners at steep marathons (10 – 25 m/km drop) also tended to do worse than other runners.

Notice, though, that in most cases the minor downhill races aren’t very far removed from the flat races. So a small net drop might offer a minor benefit, but it’s nothing like the steeper races with 10+ m/km net drop.

I’d be hesitant to draw too many conclusions about the absolute time differences here, because the heat may have exacerbated that. I’ll do a similar analysis after the 2025 Boston Marathon – when there should be better conditions.

But at a minimum, this establishes that the very steep REVEL races are far faster than flat races, and they’re also far faster than other downhill races with smaller net drops.

So What’s the Bottom Line on Downhill Races?

At the end of the day, whether or not downhill races are “valid” and should be allowed for Boston qualifying purposes is a value judgement. That’s your opinion, you’re entitled to it, and I’m not here to argue one way or the other.

But I’m here to offer some underlying data to ground that discussion, because every time I see people talk about downhill races I see some really questionable assumptions and claims getting thrown around.

So here are some general conclusions:

1. The vast majority of races are flat or near flat. Only a small handful of races have steep downhills.
2. There is a big difference between a minor downhill race, like Boston or CIM, and a steep downhill race, like REVEL Big Cottonwood. If someone wanted to draw a distinction between “eligible” and “ineligible” courses, there are several logical places to make that division.
3. The vast majority of runners complete marathons on flat or nearly flat courses, so runners qualifying on steep downhill courses make up a relatively small group.
4. Although runners on downhill courses are a) more likely to qualify for Boston and b) more likely to apply for Boston, they still only make up ~10% of the field of qualifiers at the Boston Marathon.
5. A significant portion of these runners could get in by running a different course. So the net impact of additional runners at Boston solely due to downhill courses is in the realm of 1,000 to 1,500 runners.
6. This likely contributes no more than a minute to the cutoff time. And a minute may be a generously high estimate. It’s significant, but it’s small.
7. Eliminating downhill races from the Boston qualifying process would not eliminate the need for a cutoff.
8. Steep downhill races are unequivocally faster than flat and minor downhill races, to the tune of about 10 minutes – and possibly more – with greater benefits to older (slower) runners.

For more detail on the data, including additional visuals, I’d encourage you to check out the full analysis on Medium. Again, if you’re not a Medium subscriber, you can request a special link to read the article here.