The difference between those two children isn’t talent. It’s architecture. Imagine this. Two children — same age, same problem in front of them: 99 + 23. The first child picks up his pencil and begins to stack. Ones column: 9 + 3 = 12, write the 2, carry the 1. Tens column: 9 + 2 + 1 = 12, write the 2, carry the 1. Hundreds column: write the 1. He gets 122. He’s right. It takes him about eighteen seconds, and he needs the pencil the entire time. The second child pauses — maybe two seconds — and says the answer. Out loud. In her head. She didn’t touch a pencil. Both children got 122. But only one of them did mathematics. | | “99 is not a wall. It is a disguised 100, and the number itself is inviting you to simplify.” |
What did she do? She noticed something the stacking procedure is specifically designed to hide: 99 is one away from a round number. So she borrowed one from the 23, handed it to the 99, made 100, and then added the remaining 22. The whole computation collapsed into a single, elegant move. 100 + 22 = 122. This strategy — what we call working with “friendly numbers” (numbers that are easy to compose and decompose, especially multiples of 10 or 100) — is not a trick. It is the beginning of real mathematical thinking. It is the moment a child stops executing a procedure and starts reading the numbers in front of her. It is so easy to see why we, as parents, often lean back toward the stacking method. For most of us, that “column by column, carry the one” approach is written into our muscle memory. It’s the reliable tool we were handed as children, and it feels like coming home to a familiar neighborhood. It’s safe, it’s visible on the paper, and it has always worked for us. And here is what breaks my heart a little: he wasn’t wrong to want his child to get the right answer. He just didn’t know what was being built on the other side of the screen — inside our small group, with five other children who were all learning to see numbers rather than just process them. What the Pencil CostsThe stacking algorithm is not bad mathematics. It is correct, and it will produce the right answer reliably. But it is a procedure for paper. It has no life in the mind. Ask a child trained only on stacking to add 499 + 148 in their head, and they will either refuse or attempt to visualize the columns in the air — an enormous, exhausting cognitive task that produces errors and frustration. A child trained in mental math strategies sees 499 and thinks: that is one away from 500. She borrows one from 148, makes 500, adds 147. Done. In two seconds. In her head. Mid-conversation. Without stopping to find a pencil. The difference between these two children, compounded over twelve years of schooling, is not a gap in speed. It is a gap in mathematical vision. One child sees numbers as symbols to manipulate on a page. The other sees them as relationships to reason about in the world. That second child is the one who thrives in algebra. In calculus. In STEM. Not because she was smarter, but because she was taught to think — rather than to stack. On the blog, I go much deeper into how we teach this — the progression from number bonds to friendly-number mastery to multi-step mental strategies — and why this architecture is the very foundation that makes everything else in mathematics feel obvious rather than overwhelming. With enormous admiration for the children who pause before they compute, | 
