I think route acceptance should work like this: five requirements that decide whether something is a valid FKT route, and a five-part score that rates how well the submission is documented.
The split matters. The requirements are facts about the route, and a route either meets them or it does not. The score is about the submission, and every point a submitter loses there tells them exactly what to add. Nothing in either half asks an editor to guess whether anyone will want to run the route, because that is not a question anyone can answer, and a standard built on it cannot be applied evenly.
A submitter can work out their own result before submitting. That is the test I set myself. If you cannot score your own submission from the published criteria, the criteria are not doing their job.
Work out whether a route would be accepted under the standard we propose, and what your submission is missing. Nothing you enter leaves your browser.
Part one: the five requirements
Facts about the route. Every one has to be met.
1 Could another athlete attempt the same thing and know whether they finished it?
A fixed route supplies a GPX track, or names a trail travelled between stated points. An open challenge names the full set of objectives and any rules on order, timeframe, or self-powered links. A single summit names the peak, the start point, and any style constraint.
2 Are the start and finish determinable by someone who has never been there?
Named features or stated coordinates. Where the athlete picks their own start, as on a peak-bagging challenge, say when the clock starts and stops instead.
3 Is travel along the whole route legal for anyone?
No trespass. Any permit is one any member of the public can obtain. No permanent closure crossed. Rights of way across private land are fine.
4 Is it at least 5 miles (8 km) long, or does it climb at least 500 feet (150 m)?
Either one clears it, not both. For an open challenge, measure the shortest legal completion. This is the platform's existing published minimum, unchanged.
5 Is it absent from the site already?
Fails only if an existing route shares both endpoints and covers the same objective. Sharing ground along the way does not count: a section of a long trail overlaps its parent completely and is still a different route.
Part two: how good is the submission
This scores your write-up, not your route. 9 of 15 to pass, and anything you lose here you can go and fix.
Endpoint precision –
How exactly the two ends are pinned down.
Coherence –
Whether the route or objective set hangs together as one thing.
Description completeness –
Whether the next athlete could follow it from what you wrote.
Access information –
What a stranger needs in order to legally start.
Hazard disclosure –
Not a difficulty rating. What the next person needs to know before setting out.
Answer the questions above and the result appears here.
The five requirements
Every one must be met. A refusal names the requirement that was not.
1. The objective is repeatable. Another athlete could attempt the same thing and know whether they finished it. A fixed route supplies a GPX track or names a trail travelled between stated points. An open challenge names the full set of objectives and any rules about order, timeframe, or whether links must be self-powered. A single summit names the peak, the start point, and any style constraint.
2. The start and finish are determinable. Both are named features or stated coordinates a stranger could find. Where the athlete chooses their own start, as on a peak-bagging challenge, the rule says when the clock starts and stops.
3. Travel along it is legal. No trespass. Any permit is one that any member of the public can obtain. No permanent closure is crossed. Rights of way across private land are fine, which matters everywhere outside the United States and in much of it.
4. It meets the published minimum. At least 5 miles (8 km), or at least 500 feet (150 m) of climbing. Either figure clears it. For an open challenge, measure the shortest legal completion. This is the platform’s existing rule.
5. It is not already listed. An existing route does not already share both endpoints and cover the same objective.
Sharing ground with another route is not a reason to refuse. A section of a long trail overlaps its parent completely and is still a different route, because it starts and finishes somewhere else. The site already carries 22 Pacific Crest Trail routes and 15 Appalachian Trail routes on exactly that basis.
The score, out of fifteen
Nine points to pass. Every item is something the submitter can improve and resubmit, which is the point of scoring it rather than gating on it.
Endpoint precision, 0 to 3. Three if both ends are named features or coordinates. Two if one is named and the other described. One if both are described but a stranger could still find them.
Coherence, 0 to 3. Three if the route follows a single named trail or natural feature between its endpoints, or the objective set is a recognised complete group. Two if it links named trails or features logically, or is a defined subset with a stated reason. One if there is a defined line or set with a stated reason.
Description completeness, 0 to 3. Three if the distance and climbing are given and match the track, the surface and terrain are described, and the route is described well enough to follow. Deduct for each missing piece.
Access information, 0 to 3. Three if the land manager, any permits and how to get them, parking or trailhead access, and any seasonal closure are all stated. Deduct for each one an athlete would have to find out for themselves.
Hazard disclosure, 0 to 3. Three if technical sections, exposure, water carries, river crossings and navigation difficulty are described, or the route has none and says so. This is not a difficulty rating. It is what the next person needs to know before they go.
What is deliberately absent
No points for scenery, for a route being notable, or for anyone predicting that others will want to repeat it. Those cannot be scored the same way twice by two different people, which is what makes them subjective.
No points for guidebook coverage, media attention, social media presence, or an existing race on the course. Those measure how well documented a place already is, which is a fact about local publishing rather than about the route, and it falls hardest on routes outside wealthy English-speaking countries.
No points for who submitted it, how many routes they have submitted before, whether a local club supports it, or whether locals have already run it. None of those are properties of the route.
No penalty for a route being in a busy area. The current guidelines raise the bar where routes already exist, and that gets it backwards: the places with the most routes are the places where new routes actually get run.
Why the score is not a verdict
A low score means the submission needs more work, and it says which part. A submitter who scores six can go and add the permit information and the hazard notes and come back with eleven. Nobody is told their route is not interesting enough, because nobody is in a position to know that.
The five requirements are different. Those are facts, and a route that fails one fails it for a reason the submitter can read and check for themselves.