The Number That Always Comes Back Number Play

Line up several children of different heights, and ask each one to call out how many of their neighbours (looking both left and right, all the way to the ends of the line) are taller than they are. Two children standing right next to each other can genuinely call out the same number, and a clever arrangement can even make both children standing at the very ends of the line say '2' at once. What this puzzle really tests is whether a single number, reported by each person, can carry enough information to describe something about their entire surroundings -- exactly the spirit the rest of this chapter runs on: numbers as carriers of pattern and structure, not just amounts to be added up.

In a grid of numbers, a supercell is any cell whose value beats every one of its direct up/down/left/right neighbours (diagonal neighbours don't count). Colouring in every supercell across a full grid of unique numbers reveals that the biggest number in the whole grid is always automatically a supercell (nothing anywhere can beat it), while the smallest number can never be one (something adjacent will always beat it). Beyond these two extremes, though, exactly how many supercells a grid contains, and where they end up, depends entirely on how the numbers are arranged -- rearranging the very same set of numbers differently can create far more supercells, or far fewer, making this a genuine design puzzle rather than a fixed fact about the numbers themselves.

Adding up a number's individual digits (its digit sum) turns out to be a surprisingly useful shortcut: every number sharing a particular digit sum, like 14, forms a whole family (248, 356, 653, and endlessly more, since extra zeroes or well-chosen extra digits can always be added without disturbing the total). Digit sums of a run of consecutive numbers, like every whole number from 40 to 70, reveal a repeating rhythm rather than a random scatter. A related but different kind of question -- how many times does the single digit 7 actually appear, written out, somewhere among all the numbers from 1 to 1000 -- needs a completely different kind of careful counting, position by position, rather than simply adding digits within one number at a time.

A palindrome reads identically forwards and backwards, like 4884 or 15251. Taking almost any number, reversing its digits, and adding the reversal back to the original, then repeating this process on the result, eventually produces a palindrome for every single 2-digit number -- but for 3-digit numbers, this is only mostly true. Starting from 196 specifically and repeating reverse-and-add, no palindrome has ever been found, no matter how many digits mathematicians have let the computation run to (well beyond a hundred million digits, using modern computers); it remains a genuinely open, unsolved question whether 196 ever reaches one at all, or never will.

D.R. Kaprekar, a mathematician working in Devlali, Maharashtra, discovered a genuinely startling routine in 1949. Take any 4-digit number whose digits are not all identical, arrange its digits to form the largest possible number and the smallest possible number, subtract the smaller from the larger, and repeat this exact process on the result. Starting from 6382: 8632-2368=6264, then 6642-2466=4176, then 7641-1467=6174 -- and repeating the routine on 6174 itself gives 7641-1467=6174 right back again. Astonishingly, every single starting number (except those with all identical digits) reaches exactly 6174 this way, though different starting numbers take a different number of rounds to get there.

Dates and times can be palindromes too: 11/02/2011 (in day/month/year form) reads the same forwards and backwards, and a digital clock showing 10:01 is itself a tiny palindrome. Meanwhile, an entirely different kind of number puzzle, the Collatz Conjecture (proposed by the German mathematician Lothar Collatz in 1937), asks something almost embarrassingly simple: starting from any whole number, if it's even halve it, if it's odd triple it and add 1, then repeat forever -- does this process always eventually reach 1, whatever number it started from? Every single number ever tried, across untold billions of trials on modern computers, has eventually reached 1. Yet nobody has ever proved this must always happen for every possible starting number, making it one of the most famous unsolved problems in all of mathematics, simple enough for a child to fully understand and attempt themselves.

Not every number question needs an exact answer -- estimating how many steps it takes to walk somewhere, or how many hours a student has genuinely spent in school, is its own real skill, checking a claimed number against reasonable common sense rather than calculating it precisely. A completely different kind of number thinking closes this chapter: in the game '21', two players alternately add 1, 2, or 3 to a running total starting from 0, and whoever says '21' first wins. Working backwards from the target reveals a fixed winning strategy: always aim to be the one who says a number one more than a multiple of 4 (1, 5, 9, 13, 17, 21), and the game is won before it even properly begins, regardless of what the other player tries.