Your site looks interesting but your terms of service seem really onerous. That stopped me from finishing the signup process. My running data is deeply personal; it's tied to my health metrics and contains years of location patterns around my daily life.
So when I see stuff like "irrevocable, sublicensable" rights to all of my running data...that's a lot to give up for a company and product I know very little about.
Capping your liability at $50 total for any harm and doing as much as you can to try and get me to lose access to legal protections such as class action and a jury trial is all a bit much.
Firstly, thanks for reading the ToS and for raising your concern. Not that it means much, but our goal is not to screw over our users. This is the first time someone has raised an issue with the ToS.
We understand how personal user fitness data is, which is why we’ve tried our best to leave data in the control of the users. A first step was to make sure that users actually see the ToS, and we require all users to scroll through them and accept before they link any data with us. But honestly, we know the ToS is daunting for many users, which is why we give clear switches for users to revoke our access to using their data (both for model training and for using their anonymized stats in aggregated statistics); we respect the user’s decision to revoke our access in these ways, even though our ToS doesn’t require us to. Your fitness data gets immediately deleted when you delete your account; we also allow specific sources (e.g., Garmin) to get deleted without deleting your entire account.
But we agree with you–the ToS are too aggressive. For context, we had them drafted by a well-respected firm in the Bay Area. As a startup, we didn’t have the budget to carefully, line-by-line, draft terms that perfectly fit our site. Instead, we gave the firm some ToS from large companies like Strava and Garmin, and asked them if they can draft something similar. We wanted to ensure that we were legally allowed to glean insights from the fitness data of our users, and when we read the terms, it looked like it provided that, which is why we approved it. We aren’t lawyers, so we didn’t understand the ramifications of the legalese, and we’ll make sure to emphasize that we respect the user’s decision to remove our access when we re-draft them. We’ll shop around again for other firms that specialize in this area.
There's a limit on how many people can run the Boston Marathon.
To qualify to "run Boston", you have to run another marathon in a qualifying time[1], prior to applying. For example, the qualifying time for a male 40-44 is 3h05m. For a female of the same age, 3h35. Non-binary, 3h35.
You submit your application and qualifying race and time, and then some time _later_, based on the number applications received that are within the cutoff (and it's always more than they can accept), they adjust the cutoff time downwards even further. That additional cutoff delta is the what's being calculated on the slider here. So if your published cutoff is 3h05, and the slider predicts a 6min delta, you need to have run 2h59, not 3h05.
Methodologically, why would you have one cutoff vs a different cutoff per group (as there are different qualifying times per group)?
I am not a marathoner, but I'd imagine that a 6 min decrease from the stated qualifying time cuts out a larger proportion of younger runners (i.e. decreasing the threshold from 2h55 to 2h49 for men 18-34 seems like a much sharper cut than decreasing 4h20 to 4h14 for women 60-64). I would have thought you'd want to pick the delta by looking at the distribution within each gender x age pool.
It could be intentional. For a lot of folks, running the Boston Marathon is a dream, so maybe the BAA wants to make that dream just slightly more attainable the older you get.
They can also improve that balance by adjusting the qualifying times from year to year, and they do.
They could even make a projection of future cutoff times and take that into account when setting the baseline qualifying times. In other words, be a little more generous with the 18-34 group initially knowing that you'll like penalize them more with your one-size-fits-all cutoff. I'm not sure if they do that.
Also, the current qualifying times are all multiples of 5 minutes. If they really want to improve balance between groups, the low-hanging fruit is to make those more granular.
Yeah, doing it by flat time delta rather than percent delta seems fundamentally flawed, but of course it makes it easier for the average person to understand.
I also don't understand what the motives are behind how the age/gender buckets are calculated in the first place. I'm not sure if it's public or not.
Are they:
* Trying to calculate based on an nth percentile finishing time across each bucket?
* Trying to ensure roughly equal percentages of applicants from each bucket get accepted?
Making sure they can accept a lot of high disposable income 40+ runners that will buy a lot of merch. The time cutoffs start to get much easier after 40.
Doesn't look true, I inputed the times in an age grading calculator and 39 cutoff time was easier than 44, even if it's not very accurate I doubt it's "much easier".
It’s a common problem. Time trialling uses Age Adjusted Time for events which uses a flat time reduction based on age. One guy pointed out the absurdity that he’s still extremely fast in his 50s, so his AAT end up impossibly fast, winning him a lot of events as a result!
>To qualify to "run Boston", you have to run another marathon in a qualifying time
just to add this to the mix: there are faster and slower marathon courses, so you can improve your qualifying time by running in one of them. "downhill" seems to be a promising factor.
You’re not running down Helvellyn, it’s more the cumulative downhill that qualifies a “downhill” marathon. Remember this is 1500 feet drop over the course of 26 miles, so you’ll get the benefit of gravity without the downsides of steep drops.
As someone with a 5:38 delta, I'm very anxiously waiting for BAA to announce the official cutoff.
In the meantime, if you're at all curious about the kinds of levels to which people go with trying to predict the cutoff check out this blog[1]. This is from Brian Rock [2], who every year collects data about a lot of marathons all over the world and then tries to guess the official cutoff for the Boston marathon. Very cool stuff!
Nice idea but I think the predictions are way off. I ran a 3:17 in spring and targeting a 3:10 marathon in a couple of weeks.
As unlikely as it sounds Strava predicts 3:13 (think they base it on my last marathon), Garmin is similar. Runalyze is about as off as you are.
Maybe you're putting too much emphasis on weekly volume. M35 and I can run this with 50k weekly and 75k peak volume. Relatively confident I'd be able to sub 3 with weekly volume of 80k/100k peak.
I have a friend who has run it for charity multiple times. They're a six-ish-plus hour marathoner. My partner has run them in the last 10 miles a couple of times pre-bombing. Her take is that it's pretty miserable if you're that slow because they start breaking down the course, the aid stations get light on supplies, and the porta-potties get unspeakable. So yeah, you can, but it would be a better experience if you're not quite so far behind a BQ time.
As for me, I don't run. Props to anyone who can run a BQ and enjoy it.
I was wondering if they would let a total slug run the marathon for money. It turns out that they expect you to run it in under 6 hours (which is a slow "recreational" pace). You're also expected to raise an average of $15,000, but it may be more depending on the charity. More details here: https://www.charityteams.com/boston-faqs
It's not about _qualified_ runners, it's about the size of the total field accepted.
Boston allows "roughly" 25k participants, but that number fluctuates somewhat every year. If they allow 30k, the cutoff delta goes down. If they allowed an unlimited field, the cutoff delta would be 0, and you'd only have to worry about your published qualifying time.
The slider adjusts the number allowed to run (apparently the pool of time qualified candidates can be assumed to be larger than the max value of the slider).
We have a friend who has qualified three times but has never run it due to the cutoff. Maybe the BAA should consider a separate bucket for these runners.
Your site looks interesting but your terms of service seem really onerous. That stopped me from finishing the signup process. My running data is deeply personal; it's tied to my health metrics and contains years of location patterns around my daily life.
So when I see stuff like "irrevocable, sublicensable" rights to all of my running data...that's a lot to give up for a company and product I know very little about.
Capping your liability at $50 total for any harm and doing as much as you can to try and get me to lose access to legal protections such as class action and a jury trial is all a bit much.
Firstly, thanks for reading the ToS and for raising your concern. Not that it means much, but our goal is not to screw over our users. This is the first time someone has raised an issue with the ToS.
We understand how personal user fitness data is, which is why we’ve tried our best to leave data in the control of the users. A first step was to make sure that users actually see the ToS, and we require all users to scroll through them and accept before they link any data with us. But honestly, we know the ToS is daunting for many users, which is why we give clear switches for users to revoke our access to using their data (both for model training and for using their anonymized stats in aggregated statistics); we respect the user’s decision to revoke our access in these ways, even though our ToS doesn’t require us to. Your fitness data gets immediately deleted when you delete your account; we also allow specific sources (e.g., Garmin) to get deleted without deleting your entire account.
But we agree with you–the ToS are too aggressive. For context, we had them drafted by a well-respected firm in the Bay Area. As a startup, we didn’t have the budget to carefully, line-by-line, draft terms that perfectly fit our site. Instead, we gave the firm some ToS from large companies like Strava and Garmin, and asked them if they can draft something similar. We wanted to ensure that we were legally allowed to glean insights from the fitness data of our users, and when we read the terms, it looked like it provided that, which is why we approved it. We aren’t lawyers, so we didn’t understand the ramifications of the legalese, and we’ll make sure to emphasize that we respect the user’s decision to remove our access when we re-draft them. We’ll shop around again for other firms that specialize in this area.
I appreciate the reply. I'll try to remember to keep tabs on it!
OTOH, it's nice when companies that will screw you over, signal that, by having lawyers appear to carefully set up the cover for them to do so.
Same. Paused the signup process to take a look at how they were going to use my data, and then decided against it once I understood.
Context:
There's a limit on how many people can run the Boston Marathon.
To qualify to "run Boston", you have to run another marathon in a qualifying time[1], prior to applying. For example, the qualifying time for a male 40-44 is 3h05m. For a female of the same age, 3h35. Non-binary, 3h35.
You submit your application and qualifying race and time, and then some time _later_, based on the number applications received that are within the cutoff (and it's always more than they can accept), they adjust the cutoff time downwards even further. That additional cutoff delta is the what's being calculated on the slider here. So if your published cutoff is 3h05, and the slider predicts a 6min delta, you need to have run 2h59, not 3h05.
1. https://www.baa.org/races/boston-marathon/qualify
Methodologically, why would you have one cutoff vs a different cutoff per group (as there are different qualifying times per group)?
I am not a marathoner, but I'd imagine that a 6 min decrease from the stated qualifying time cuts out a larger proportion of younger runners (i.e. decreasing the threshold from 2h55 to 2h49 for men 18-34 seems like a much sharper cut than decreasing 4h20 to 4h14 for women 60-64). I would have thought you'd want to pick the delta by looking at the distribution within each gender x age pool.
It could be intentional. For a lot of folks, running the Boston Marathon is a dream, so maybe the BAA wants to make that dream just slightly more attainable the older you get.
As always it's probably because maths still is too hard for most people and keeping the rule simple won over fairness.
Ironman just switched to age-grading, which is just as simple and already used by many runner training plans and race calculators.
https://www.ironman.com/news/age-group-qualification-system
They can also improve that balance by adjusting the qualifying times from year to year, and they do.
They could even make a projection of future cutoff times and take that into account when setting the baseline qualifying times. In other words, be a little more generous with the 18-34 group initially knowing that you'll like penalize them more with your one-size-fits-all cutoff. I'm not sure if they do that.
Also, the current qualifying times are all multiples of 5 minutes. If they really want to improve balance between groups, the low-hanging fruit is to make those more granular.
Yeah, doing it by flat time delta rather than percent delta seems fundamentally flawed, but of course it makes it easier for the average person to understand.
I also don't understand what the motives are behind how the age/gender buckets are calculated in the first place. I'm not sure if it's public or not.
Are they:
* Trying to calculate based on an nth percentile finishing time across each bucket?
* Trying to ensure roughly equal percentages of applicants from each bucket get accepted?
* Something else?
Making sure they can accept a lot of high disposable income 40+ runners that will buy a lot of merch. The time cutoffs start to get much easier after 40.
Doesn't look true, I inputed the times in an age grading calculator and 39 cutoff time was easier than 44, even if it's not very accurate I doubt it's "much easier".
It’s a common problem. Time trialling uses Age Adjusted Time for events which uses a flat time reduction based on age. One guy pointed out the absurdity that he’s still extremely fast in his 50s, so his AAT end up impossibly fast, winning him a lot of events as a result!
>To qualify to "run Boston", you have to run another marathon in a qualifying time
just to add this to the mix: there are faster and slower marathon courses, so you can improve your qualifying time by running in one of them. "downhill" seems to be a promising factor.
https://findmymarathon.com/fastestmarathoncourses-state.php?...
there is now a time adjustment for downhill, and extreme downhills are no longer qualifying
https://runtothefinish.com/downhill-boston-qualifiers/
Dunno man, running downhill is murder for knees, I usually had to slow down instead of speeding up on down slopes.
You’re not running down Helvellyn, it’s more the cumulative downhill that qualifies a “downhill” marathon. Remember this is 1500 feet drop over the course of 26 miles, so you’ll get the benefit of gravity without the downsides of steep drops.
I ran a downhill marathon that was on a rail trail through the mountains. It was a very very gradual downhill the whole way.
My friend that ran it was mad at me when we finished because he thought parts of it were uphill (even though it wasn't.)
That race won't be a Boston qualifier next year.
You can also run for a charity by raising sufficient money for a cause that sponsors you.
As someone with a 5:38 delta, I'm very anxiously waiting for BAA to announce the official cutoff.
In the meantime, if you're at all curious about the kinds of levels to which people go with trying to predict the cutoff check out this blog[1]. This is from Brian Rock [2], who every year collects data about a lot of marathons all over the world and then tries to guess the official cutoff for the Boston marathon. Very cool stuff!
[1]: https://runningwithrock.com/boston-marathon-cutoff-time-trac... [2]: https://runningwithrock.com/about-me/
Brian Rock's tracker is great. It's a ton of work to collect and maintain that estimate throughout the year, so we hope he keeps it up!
According to Boston Marathon 2025 statistics, about 9% of entrants don't start the race.
Number of male and female entrants in the 18-39 age group are almost even, but the proportion of female entrants drops in the older age groups.
This pattern appears in 2024 and 2023 as well.
see https://registration.baa.org/2025/cf/Public/iframe_Statistic...
Nice idea but I think the predictions are way off. I ran a 3:17 in spring and targeting a 3:10 marathon in a couple of weeks.
As unlikely as it sounds Strava predicts 3:13 (think they base it on my last marathon), Garmin is similar. Runalyze is about as off as you are.
Maybe you're putting too much emphasis on weekly volume. M35 and I can run this with 50k weekly and 75k peak volume. Relatively confident I'd be able to sub 3 with weekly volume of 80k/100k peak.
If you're not fast enough to qualify then you can also run Boston 2026 by donating or raising about $10K for charity.
https://www.baa.org/races/boston-marathon/charity-program
I have a friend who has run it for charity multiple times. They're a six-ish-plus hour marathoner. My partner has run them in the last 10 miles a couple of times pre-bombing. Her take is that it's pretty miserable if you're that slow because they start breaking down the course, the aid stations get light on supplies, and the porta-potties get unspeakable. So yeah, you can, but it would be a better experience if you're not quite so far behind a BQ time.
As for me, I don't run. Props to anyone who can run a BQ and enjoy it.
I was wondering if they would let a total slug run the marathon for money. It turns out that they expect you to run it in under 6 hours (which is a slow "recreational" pace). You're also expected to raise an average of $15,000, but it may be more depending on the charity. More details here: https://www.charityteams.com/boston-faqs
According to https://sportstoursinternational.co.uk/blog/how-to-get-entry..., you can also gain entry via a tour operator for non-USA entrants or if you've already run several of the Abbott World Marathon Majors, there's a draw program.
Boston Marathon tour operators: https://www.baa.org/races/boston-marathon/plan/international... and https://www.baa.org/boston-marathon-international-tour-progr...
Abbott World Marathon Majors draw program: https://www.worldmarathonmajors.com/content-hub/majors-draw-...
Is this usage of "cutoff" normal?
I run a lot of races and cutoff time has always meant the amount of time you have to finish the race.
it's nice to see so many people running marathons now
y'all better safety-qualify at least -6 minutes tho, -7 if you can
Why does the cutoff get smaller with increasing qualified runners?
I would have expected the opposite.
It's not about _qualified_ runners, it's about the size of the total field accepted.
Boston allows "roughly" 25k participants, but that number fluctuates somewhat every year. If they allow 30k, the cutoff delta goes down. If they allowed an unlimited field, the cutoff delta would be 0, and you'd only have to worry about your published qualifying time.
It's a cutoff delta, you have to subtract it from the "preliminary" qualifying time.
The slider adjusts the number allowed to run (apparently the pool of time qualified candidates can be assumed to be larger than the max value of the slider).
Why are they using a linear model for cutoff time vs qualified runners?
No. The answer is no.
We have a friend who has qualified three times but has never run it due to the cutoff. Maybe the BAA should consider a separate bucket for these runners.
Betteridge's law of Boston qualifying times.
Nerds will do everything but put in more mileage to get fast /s