ATC Games already kept score. Nails, Bails and Fails are how a run reads on the platform, and the leaderboards have been there since the games shipped. The Black Belt track puts an order on top of them.
A belt is earned across a spread of exercises rather than won in any single one, so no lucky run takes one.
What a belt asks for
The track is a sequence. Each belt names a set of things a student has to have done across the games, and the next one names more. A student can see where they stand and what the next belt is going to want from them.
Spreading the requirement is the entire design. A student who is excellent at one game and quietly avoids the rest does not move up, because a belt is not measuring a peak. It is measuring a floor.
Leaderboards, carefully
Leaderboards were already on the platform and students already watch them. The obvious risk in stacking belts on top is turning the whole thing into a contest between students, which suits the confident ones and nobody else.
A belt is not a ranking. Two students can hold the same one, and each earns theirs against a standard rather than against whoever is above them on a list. The leaderboard stays for the students who enjoy that. The belt is there for the ones who do not.
What a training manager gets
Nothing new to administer. The belts sit on runs the platform was already recording, so a student chasing one is practising, and that practice arrives in progress data exactly as any other run does.
That is worth saying plainly, because a badge system generating its own separate set of numbers would be one more thing to maintain and one more thing to distrust. This one does neither. It is a way of reading runs that already exist.
What it changes is when students practise. A student who has finished what was set and still has an hour spare does something with the hour. Whether any of it goes on the platform is a design question, not a training one.
Nails, Bails and Fails
The platform has never graded a run as a pass or a fail, and the belts do not change that. A run reads as a Nail, a Bail or a Fail, which are three outcomes rather than two, and the middle one is doing the real work.
A pass and a fail tell a student almost nothing. They are a verdict without a diagnosis, and a student who fails twice in a row has learned only that they failed twice in a row. Three outcomes at least distinguish between a run that fell apart and a run the student walked away from, and those call for different responses.
The belts are built on the same principle at a larger scale. Progress is described by what a student has actually done across the games rather than by a line they either cleared or did not, which is more informative to the student and considerably more informative to whoever is supporting them.
Why the games are games
Lektor builds games on purpose, and for some buyers that is the uncomfortable part, so it is worth answering head on. The training is not easier because it is playable. Nothing about the standard moves, and nothing about what a student has to demonstrate at the end moves either.
What playability buys is repetition, and repetition is what makes controllers. A student who wants another run takes another run without being told to. A belt is one more reason to take it.
More from Lektor
Practice does not keep office hours.
The T3 Simulator is open whenever the student is free, which is often not when the building is.
The Assessor refresher opens again.
It runs remotely with an instructor, and it starts when your assessor is ready rather than on a date somebody else picked.
The exercise gets a grade too.
Manager Tools now reads progress data by exercise as well as by student, so a wall a whole intake keeps hitting shows up as a wall.