Twenty-Seven Takes 111 Steps and Twenty-Six Takes Ten
Gepubliceerd op 26-10-2026 · 3 min leestijd · Dagelijkse calculators
Lena Hoffmann — Redacteur wetenschap en onderwijs bij OneKitly
Wiskunde · Natuurkunde
Getoetst aan 2 bronnen
The rule is as short as a rule gets: if the number is even, halve it; if it is odd, multiply by three and add one. Every starting value tried so far eventually reaches one, and nobody has been able to prove that every value must. The behaviour on the way is what makes it striking. Starting at 26 the sequence goes 26, 13, 40, 20, 10, 5, 16, 8, 4, 2, 1 and is finished in ten steps. Starting at 27 — one more — it wanders for 111 steps and climbs to 9,232 before coming down. Starting at 97 it takes 118 steps and peaks at exactly the same 9,232, because it joins the path 27 was already on.
Halve it if even, triple and add one if odd. Starting from 27 the sequence climbs to 9,232 before it falls; starting from 26 it is over in ten moves.
Why it climbs at all
An odd step multiplies by three and an even step divides by two, so a single odd move followed by a single even move leaves the number about one and a half times bigger. The sequence only shrinks because odd numbers become even immediately — 3n+1 is always even — and even numbers often halve several times in a row. On average the shrinking wins, which is a heuristic argument that the sequence should always come down, and heuristics are not proofs. Whether some starting value climbs for ever, or falls into a loop that never reaches one, is exactly what remains open.
Checked further than almost anything else
Computers have confirmed the conjecture for every starting value up to astronomically large bounds, and that is worth exactly as much as any other finite check: nothing at all as a proof. Mathematics has a long history of patterns that hold for enormous ranges and then fail, which is why the verification is reported as a bound rather than as an answer. What the computation does buy is confidence about where a counterexample could hide — it must be larger than everything tested, which constrains what any future proof or disproof has to look like.
| Start | Steps to 1 | Highest value reached |
|---|---|---|
| 26 | 10 | 40 |
| 27 | 111 | 9,232 |
| 97 | 118 | 9,232 |
Veelgestelde vragen
- Why do 27 and 97 reach the same peak?
- Because trajectories merge and never split. Once two starting values produce the same number at any point, everything after that is identical — the rule depends only on the current value and nothing else. The sequence from 97 arrives at 47 after fourteen steps, which is a number 27 also visits, and from there the two are the same journey. That merging is why the set of all trajectories forms a tree flowing into one, and why the conjecture is a claim about that tree having no separate pieces and no loops.
- Is a short rule a sign that a problem is easy?
- The opposite, often enough that it is worth expecting. The statement of a problem and the difficulty of settling it are unrelated quantities: this one fits in a sentence and has resisted every technique brought to it for the better part of a century, while plenty of results with pages of hypotheses fall to a page of argument. A short rule usually means few structural hooks — nothing to factorise, no symmetry to exploit, no invariant that survives a step — and the very sparseness that makes it easy to explain is what leaves a prover with nothing to grip.
Artikelen die je misschien interesseren
Alle gidsen →Vergelijkbare tools
Bronnen
Een fout in dit artikel gezien?