The builder answers one question — what are we doing at the next practice. This page is the rest: why the rungs are shaped the way they are, what the map knows, and what it doesn't.
Asking a new head coach to type in their trailheads is asking them to do work the map can do. So the team's location goes to OpenStreetMap, trail segments come back, and they get clustered into networks ranked by distance and by how much the map already knows.
It works about as well as the underlying data allows, which is the honest result. Bowie Bulldogs gets Bauerle Ranch, 3.8 km from the school — and that one has now been all the way through the engine. Three hundred and seventeen ways came back; pruning to the connected network that carries the graded trail left 182 segments over 15.8 miles, and the search returned loops for Bronze 3, 4 and 5. It refused Bronze 1 and 2, because every graded segment at Bauerle is blue and not one is green. Pick Bowie and you can read the refusal in the coach's own words.
Hays County gets a network 11.5 km out that the engine first named "Tonkawa · Blue Stem" from its own trails. Checking it against OpenStreetMap's park boundaries resolved it to Spring Lake Preserve in San Marcos — fourteen of seventeen ways graded, which is the best-mapped venue found anywhere in this exercise.
Steiner Ranch is the interesting one. Their own trails are in the map — West Ridge Canyon, Lakewood Hills, Bright Sky Overlook, Summer Vista, 1.2 km from the school — and not one metre of it carries a difficulty grade. That is the common case in Texas, and it is the whole argument for coach-entered ratings outranking the map. Four trails is five minutes of a coach's time, and it unlocks the venue.
Every network is named from its own trails, because OpenStreetMap often has the trails and not the venue — there is no "Canyonlands" in the map either. So the engine proposes a name and the coach confirms or renames it. Guessing silently would be worse than asking.
Steiner Ranch publishes the numbers: 60+ riders and 30+ coaches, pods of 8–10 with a minimum of two NICA-certified coaches each, riders reassigned as skills change. With that bench they can run five bands and never think about it.
A twelve-rider team with two coaches can run two. Not because their riders are less varied, but because a third group would put kids on trail without a licensed coach. That's why the ladder ships as an N-rung system rather than a five-rung one — and why the roster question comes before the plan rather than after it.
Groups are capped by the coaching bench. A team with eight coaches can field eight groups; a team with two can field two, however many rungs the ladder has on paper. That single rule explains most of why small teams struggle to differentiate — and it's a fact the product can surface in October instead of letting it emerge in February.
A practice is not a loop. It starts and ends at the same place — the meeting point for that practice, which changes with the venue — and the coach strings trail together until the group has covered the mileage the stripe asks for. Riding a section twice is normal. So the engine's question is not "does a circuit close here", it is "can this ground make ten miles and get us back".
| Venue | What the map holds | What a 10 mi fourth-stripe night looks like |
|---|---|---|
| Pace Bend | 29.7 mi, 54 green / 11 blue / 0 black | 2 laps of a 5.12 mi blue lap |
| Canyonlands · Lakeway | 26.9 mi, 9 named trails, 0 graded | 4 laps of 2.73 mi — mileage solid, difficulty a guess |
| Marion Sansom · Fort Worth | 16.1 mi, 56% graded, well connected | 3 laps of 3.12 mi |
| Bauerle Ranch · Austin | 15.8 mi connected, 80 segments graded, all blue | 3 laps of 3.45 mi |
| Walnut Creek · Austin | 22.9 mi, half graded, fragmented | 5 laps of 1.83 mi |
| Gateway Park · Fort Worth | 13.2 mi of Trinity Trails, linear, 0 graded | Out 5 mi, turn, come back |
| Hamilton Greenbelt | 6.9 mi in disconnected pieces, longest 1.2 mi | 9 times through the 1.2 mi piece |
Hamilton Greenbelt and Gateway Park are the two worth pausing on, because an earlier version of this tool refused both of them outright — "nothing closes into a loop." That was the wrong question. Gateway is a river corridor, so an out-and-back is not a fallback, it is simply how you ride a corridor. Hamilton is fragmented, so a ten-mile night there really is nine passes through the same 1.2 miles — which tells a coach something true and useful: that is a skills venue, not an endurance one, and it should be scheduled as one.
Difficulty, not topology. Bauerle Ranch has eighty graded segments and every one is blue, so the first two stripes get nothing there however the mileage is assembled. Marion Sansom has 0.02 mi of green. Those are real refusals — no amount of route-stringing turns blue trail green, and a stripe that has not earned blue should not be on it.
OpenStreetMap can tell you that Powerline is 0.46 miles and rated blue. It cannot tell you that Powerline is where you teach a standing climb, or that the third switchback on Paleface is the one worth stopping at. No map carries that. A coach who has ridden it does, and at the moment that knowledge leaves with them when they stop coaching.
So every named trail in the next practice's assembly can be tagged with the skills it actually teaches — and the tags use NICA's On-the-Bike 101 skill names: Neutral and Ready Position, Braking, Bike Body Separation forward-and-back and side-to-side, Introduction to Cornering, Shifting, Climbing Dismount, and the Seated, Crouched and Standing climbs.
OTB 101 is required for Level 2 and 3 coaches, so the coaches leading groups have already been taught these exact names. The tags are not new vocabulary anyone has to learn — they are the words already in the room. Past 101, NICA does not publish a skill list, so drops, switchbacks, rock gardens and log-overs are team tags and are marked as team tags rather than passed off as NICA's.
The teaching itself stays with NICA. This tags a trail with a skill's name and links out to their courses; it does not reproduce their instruction, and a coach's own notes on a trail are their own words.
Tags pool per venue, across the league. Six Texas teams ride Pace Bend. The first one to tag Powerline as a standing-climb trail has done that work for all of them, and the seventh team to arrive starts with a venue that already knows what it teaches. That is the one dataset in this whole product that compounds — trail geometry is a commodity, difficulty grades are thin, but what a piece of ground teaches a twelve-year-old exists nowhere else and gets better every season somebody rides it.
It runs backwards too, which is the part that changes a practice. Right now the engine plans by distance and difficulty. With trails tagged it can plan a session: the next practice is cornering, so here is the assembly that puts your group on the three segments tagged for it, three times. And it can answer the question a coach actually asks in September — my Stripe 3 riders need all three climb variants, which of our venues has them?
This group is mostly late fifth grade through ninth — riders roughly ten to fifteen. That band is the whole reason grouping is hard rather than clerical. It is the stretch where two riders the same age can be a year apart in size, strength and nerve, and where one rider can change more between October and April than an adult changes in five years.
Look at how the tiers above it actually sit:
| Tier | Typical grades | What decides it |
|---|---|---|
| Foundation My team calls it Bronze | late 5th – 9th | Ability. Five stripes inside it. |
| Silver | 8th – 10th | Ability |
| Gold | 10th and up | Ability |
| Platinum | any | Ability only — age never enters it |
The grades overlap on purpose. An eighth grader could sit in Foundation, in Silver, or — if they are one of the strongest riders on the team — in Platinum next to a senior. Grade does not predict what a rider can ride. Neither does age. That is not a quirk of one team; it is what a ten-to-fifteen age range does to any sport that groups by ability.
Now add NICA's model: no selections and no sidelines, no try-outs, no bench-warmers, and an explicit rejection of the up-or-out approach most youth sport runs on. Every kid who signs up rides.
Put those together and the entire weight of the season lands on one thing — putting each rider in the right group, and moving them when they outgrow it, which at this age happens mid-season. A twelve-year-old in a group that is too fast gets dropped in front of their friends, which is a different and worse experience at twelve than it would be at twenty, and the sweep coach then rides the whole session at someone else's pace. A rider in a group that is too slow is bored. Both roads end in the same place: a kid who quietly stops signing up. The thing NICA exists to prevent.
Most teams sort riders into ability pods and reassign them as the season goes on. That is the recommended baseline and it is what the coach education library teaches. The gap at Lake Travis was never a missing hierarchy. It was that once everyone was on the trail, nobody could tell which group a rider belonged in.
The fix was orange reflective duct tape, cut into strips by hand, on the handlebars. One stripe per level. Any coach can now read a rider's level across a car park — the sweep above all, because the sweep is the one who needs to know who is likely to be at the back and why. Before that, the wrong kids ended up on the wrong trails, in the wrong groups, and nobody found out until somebody was already having a bad afternoon.
Lake Travis used to run seven levels counting down: a rider started at 7 and worked toward 1. It now runs five counting up. Not so riders can be ranked against each other — so that progress reads as something gained rather than a deficit slowly shrinking. At eleven, that difference is not subtle. Anyone who has earned a stripe on a belt, or a stripe for a rank, knows which of those two feels like getting somewhere.
The timed benchmark and the feature checklist exist to answer one question, and it is not who is fastest. It is whether this rider is ready for that ground. "Challenge by choice" only works if the challenge on offer actually fits the rider — which is a measurement problem, and measuring it is the coach's job, not the twelve-year-old's.
So pod position and trend stay coach-only. Riders and families see their own stripe, their own goals and their own progress, never a ranked list. A leaderboard at this age would produce precisely the outcome this is built to avoid.
A rider can sit on the same stripe for years. Some are here to race and will climb fast; plenty are here because their friends are here and the trail is fun, and the evaluation asks which before it asks anything about ability. A tool that treated time-on-a-stripe as a problem would be quietly pushing every rider toward a start line they never asked for — which is the opposite of challenge by choice. So the coach view flags riders who have met the bar and not been moved, and riders the group is leaving behind. It does not flag contentment.
None of it is about who is best. It is about an eleven-year-old riding with riders who ride like they do, often enough and safely enough that they are still riding bikes at thirty.