The belts, briefly
Math Katas is built around a karate-themed practice session — that's why a session is called a Kata. The belts came along for the ride. Each belt is a time threshold for completing a Kata's 144 problems: White Belt is over an hour; Black Belt is under ten minutes; the intermediate belts step down in between. Black is the goal most parents and teachers care about, because it lines up neatly with an A on the timed multiplication test in school.
Scotty earned his Black Belt in a week and was the first kid in his class to finish the test with 100%. The whole story is on the Story page.
The kids who broke the timer
After a couple of years of the program running in Madelia, the lead teacher who ran it at Madelia Public Schools and I noticed something we couldn't explain. A handful of her students were finishing all 144 problems in under five minutes. Black Belt is under ten. These kids were blowing past it.
I asked her how. She said the kids were racing each other. They had even formed their own thing for it, informally, called the Math Facts club.
I believed her about the racing, but I still didn't understand the mechanism. You can want to be fast all you like; if your fingers and your recall don't keep up, you don't finish 144 problems in under five minutes. I needed to see if it was even possible.
Testing the belts on myself
I'm an engineer, so I tested every belt threshold myself before shipping it. The setup was a little ridiculous, but it worked.
I would start a Kata. I'd answer the first 140 problems quickly. Then I'd set a timer on my Apple Watch for whatever the remaining duration needed to be — say, fifty minutes — and just wait. When the timer went off, I'd answer the last four problems and finish. That gave me a Kata that took exactly the duration of the belt I was trying to validate.
I started at sixty minutes, then fifty, and worked my way down through the intermediate belts. By the time I reached Black Belt — the one with no waiting on the watch at all — I could finish a real Kata in under five minutes.
That answered the question. The kids weren't doing anything tricky. They had just put in enough reps that the answers had stopped feeling like math and started feeling like reflex.
See the problem. Instantly know the answer. Type the answer. Same every time.
That's the whole loop at this level. There's nothing else to it.
So I made Red Belt
Black Belt's threshold is under ten minutes, because that's what a school's timed test is asking for. But I'd now seen — in Madelia and in my own testing — that there was a clear next tier above it. Five minutes was a real line, not a stunt. So I added Red Belt: finish a Kata in under five minutes.
To be clear about what this is: Black Belt is the goal. Red Belt is extra. It isn't required to pass anything; it isn't on the school's report card; no kid should be aiming at it instead of Black Belt. It's there for kids who have already earned a Black Belt and want to push the fluency further, and for parents and teachers who want a cleanly defined target past the one the school is grading on.
How to earn it
I think any kid who can earn a Black Belt can earn a Red Belt. How many sessions it takes will vary by kid, but the protocol is simple, and it goes faster than the original Black Belt protocol because the kid is starting from a much higher baseline.
The version I'd recommend first:
- Pick a day when the kid has time — a Saturday is ideal.
- Do one Kata.
- Wait roughly an hour. Eat. Run around. Do whatever.
- Do another Kata.
- Repeat until they finish a Kata in under five minutes.
Because the kid is already at Black Belt, the first Kata of the day should take under ten minutes. Each subsequent Kata gets a little quicker as the answers warm up. Most kids should reach Red Belt inside an hour or two of total session time, which spread across a day comes out to a pretty light afternoon.
There's also a spread-out version — two or three Katas an evening over a few nights — that I think should work the same way, but I haven't verified it personally. If you try it that way, I'd be curious how it goes.
The honest caveat: like Black Belt, Red Belt depends on the kid having earned the foundation first. Don't try to rush a kid from scratch to Red Belt in a single sitting. The protocol that takes a kid from F to A — one Kata a day for a week — is the prerequisite. Red Belt is what comes after that's done.
Why this matters past fourth grade
Red Belt looks like a vanity badge if you stop at the fourth-grade timed test. It isn't. The reason it's worth pushing past Black Belt has to do with how the brain handles math under time pressure on every test that comes after — and the SAT and ACT are the cleanest examples.
On the SAT or ACT math section, a student has to do three things at once: remember formulas, choose a procedure, and run the arithmetic. Working memory — the small mental scratchpad the brain uses to hold things in mind for a few seconds — can only handle so much at the same time. If 9 × 7 takes a second or two of conscious effort, that's a second or two of working memory borrowed from whatever the actual problem is. Multiply that by forty problems and a hard time limit, and the gap shows up as questions left blank at the end.
The research community has been making this exact point for a long time. Math fact retrieval speed is a consistent predictor of performance on standardized tests, and the mechanism is the working-memory bottleneck: automatic recall of basic facts frees the scratchpad to do real problem-solving. Test-prep tutors say the same thing in plainer language — running out of time on the math section is one of the most common complaints they hear from students, and it's almost never the geometry that's slow. It's the arithmetic underneath the geometry.
Red Belt isn't aimed at SAT or ACT prep. But it is, in effect, what test-level fluency looks like for the multiplication table specifically: the answer arrives before the question finishes loading. If a kid has that, every later math test they take is a slightly easier test, because the cheap operations are no longer stealing from the budget for the expensive ones.
Bottom line
Focus on Black Belt. That's the one that gets the A, and that's the one every kid using Math Katas should be aiming at. Red Belt is extra — a next floor for kids who already have the A and want to push further. Earning it is mostly a question of finding an afternoon and stacking a few Katas, and the reason to bother is that the fluency it builds keeps paying off long after the fourth-grade timed test is in the rear-view.