A child failing long division is rarely failing long division.

You know the scene, because you were in it last night. It is 9 p.m. The homework should have taken twenty minutes; it has taken ninety. There have been tears — possibly yours. And the sentence that keeps circling the kitchen table is the one that makes no sense at all: "But we practised this yesterday."

You did practise it yesterday. That is exactly the clue. When practice at today's level doesn't stick, the problem is almost never today's level. It is a smaller, quieter skill from a year or two earlier that never quite landed — and nobody noticed, because at the time, nothing looked wrong.

Maths is the subject that remembers

Maths is the most vertical subject your child studies. History forgives a missed week; you rejoin the story at the next chapter. Maths does not. Every skill stands directly on an earlier one, and every new topic quietly assumes the whole tower beneath it is solid.

Here is the uncomfortable property of that tower: a gap does not announce itself. A child can wobble through a shaky year on effort and partial understanding, pass the tests, and move up. The gap goes underground. Then, one or two grades later, a new topic lands its full weight on the missing brick — and what surfaces is not "she never secured regrouping in Grade 2 (Class 2 in India, Year 3 in England)". What surfaces is "she's struggling with maths."

This is a global pattern, not a family failing, and the attainment data says so plainly. India's ASER 2022 survey found that only roughly 44% of rural Grade 8 students could solve a division problem pitched at Grade 4 level.

In the United States, NAEP results have repeatedly shown large shares of students performing below proficiency in maths, and international assessments like TIMSS and PISA find, in nearly every participating country, substantial groups of students working years below the expectations for their age.

Whether the school is in Rajasthan, Rotterdam or Rhode Island, an enormous number of children are carrying invisible gaps up the maths staircase. Yours is not behind because of anything you did. Yours is behind a specific brick.

Three children, three hidden gaps

The pattern is easiest to see in the flesh. Here are the three versions of it we meet most often.

1. "He fails long division"

Long division is not one skill; it is a loop of four — estimate, multiply, subtract, bring down — repeated three or four times per question. Now suppose the subtraction step is not automatic. Suppose 42 − 36 still requires a small internal expedition.

Every pass through the loop, that expedition burns working memory — the small mental workspace where the child is also trying to hold the algorithm, the running answer, and which digit comes down next. Research by Mark Ashcraft and others on working memory in arithmetic points to exactly this mechanism: when basic facts are not automatic, multi-step procedures exceed a child's working-memory capacity, and the procedure collapses.

The child is not careless. The child is overloaded. The fix is not more long division; it is subtraction facts, until they cost nothing. (If those facts are still arriving via fingers at age nine, that is the same signal wearing a different coat.)

2. "She just doesn't get fractions"

A child who reads ½ as "a 1 and a 2 with a line between them" is not confused about fractions. She is missing the idea underneath them: division as sharing. If "divide 12 sweets among 4 friends" never became a picture in her head, then ½ has nothing to be a picture of. It is just typography.

Teachers can present fractions beautifully for a term and it will slide straight off, because the lesson assumes a foundation that was never poured. Go back to sharing — real objects, real arguments over who got more — and fractions stop being mysterious rather quickly.

3. "He was fine until this year" (age 11–13)

The pre-algebra years are where old debts get called in. "The x stands for a number" is only meaningful to a child for whom numbers themselves are structured objects — tens, ones, a place-value system with rules. If place value never truly landed back in Grade 2 or 3, then x has nothing to attach to, and algebra-readiness work feels like being asked to translate a language he never learned to a language he cannot hear.

Parents in this stage often blame puberty, the new school, the phone. Sometimes fairly. But check place value first. It is cheaper than blaming the phone.

Why "more practice" makes it worse

The instinct when a child struggles is beautifully logical and completely wrong: more practice at today's level. More long division worksheets. Another tutor for fractions.

But if the gap is two grades down, today's practice is drilling the symptom. Worse, every failed session teaches the child a lesson you never intended: I practise and I still fail, so I must be the kind of person who fails at this. That belief calcifies into maths anxiety, and anxiety then eats the very working memory the maths needed. You end up further back than you started, with a child who now flinches at the workbook.

Governments have noticed that gaps compound this way, which is why so many run foundational-numeracy campaigns — India's NIPUN Bharat mission, England's Maths Hubs programme, numeracy acts in several US states. Entire education systems are built around going back to secure the foundations.

Your kitchen table can run the same policy, at rather lower cost.

Your child doesn't have a maths problem. They have a specific maths problem — and it's probably two grades younger than they are.

The 15-minute kitchen-table diagnostic

You can find the brick tonight. Take fifteen minutes, keep it light — this is a treasure hunt, not a test — and ask five questions from two grades below your child's current grade. No timer, no marks, and you say so out loud.

For a child struggling in Grade 5, that means Grade 3 material, roughly:

  1. A subtraction with regrouping: 62 − 38, done however they like.
  2. Two or three quick facts: 7 × 6, 13 − 7, 8 + 5. Watch how the answer arrives — instantly, or after a visible journey.
  3. A place-value probe: "What is 10 more than 397?"
  4. A sharing question: "12 chocolates, 4 people — how many each? How do you know?"
  5. One question from their current topic, for contrast.

Then just watch. You are not scoring answers; you are timing hesitations. Anything from two grades down that produces a long pause, finger-counting, or a wrong answer delivered confidently — that is your brick. (A milestone-by-milestone map of what should be automatic when is in our grade-by-grade mental maths guide.)

Two warnings. First, one wobbly answer is data, not a diagnosis; look for the pattern. Second, whatever you find, your face stays cheerful. You have just located the problem. That is the best news this house has had all term.

One father we heard from spent three months on Grade 5 division worksheets with his daughter. Every evening, the same worksheets, the same wall; the scores did not move by a single mark. It was her teacher who finally suggested, almost as an aside, checking her subtraction. Ten flashcards later he watched his daughter finger-count her way through 13 − 7 — a fact she was supposed to have owned for three years. They shelved the division worksheets and spent two weeks on subtraction facts, ten minutes a day. Two weeks of the right practice did what three months of the wrong practice could not, and the division, when they returned to it, mostly taught itself.

Go backwards on purpose

So the plan is simple, and it is the opposite of the instinct: stop pushing forwards, go back and pour the foundation, then let the tower rebuild itself. It usually rebuilds fast — the child was never short of ability, only short of one brick.

The hard part, honestly, is doing this every day: finding today's exact edge, noticing which wrong answers point backwards, and adjusting tomorrow accordingly. That diagnosis-first loop is precisely what CREST Champs' free daily adaptive practice does automatically — a few minutes a day, pitched at the skill your child actually needs next, even when that skill lives two grades down. It goes back and finds it. That is the whole idea.