Lonely Runner Conjecture Explained (2026 Update)

Quick answer: The lonely runner conjecture says that if n runners start together on a circular track of length 1 and run forever at different constant speeds, each runner will at some moment be at least 1/n of the track away from all the others. Jörg Wills (1960s) and Thomas Cusick (1973) posed it. It was proved for up to 7 runners by 2008; computer-assisted proofs for 8, 9 and 10 runners were published in 2026, and later 2026 preprints report 11 to 13 runners and beyond. The general conjecture remains open.

Picture runners on a circular track, all starting at the same line, each running forever at their own constant speed. Will every runner, at some moment, find themselves far from everyone else, “lonely”? The question sounds like a playground puzzle, yet it has stumped mathematicians for almost 60 years and is tied to number theory, geometry and even graph coloring.

The Precise Statement

Take \(n\) runners with distinct constant speeds on a circular track of length 1. A runner is lonely at time \(t\) if every other runner is at distance at least \(1/n\) from them along the track. The conjecture: every runner is lonely at some time.

Switch to a moving reference frame in which one runner stands still. After a further (nontrivial) reduction, the problem becomes: for any \(k = n – 1\) positive integer speeds \(v_1, \dots, v_k\), there is a time \(t\) with

\[ \lVert t\,v_i \rVert \ge \frac{1}{k+1} \quad \text{for every } i, \]

where \(\lVert x \rVert\) is the distance from \(x\) to the nearest integer.

A Worked Example: 3 Runners

With 3 runners, after shifting, the other two have speeds, say, 1 and 2 relative to the stationary runner, and we need a time when both are at least \(1/3\) of the track away. At \(t = 1/3\): the first is at position \(1/3\) and the second at \(2/3\). Both are exactly \(1/3\) from the start. So the bound \(1/n\) is achieved and cannot be improved: the speeds \(1, 2, \dots, k\) give a tight example.

How Far the Proof Has Come

Runners \(n\)Proved byYear
2, 3EasyClassical
4Ulrich Betke and Jörg Wills; also Cusick1972 to 1974
5Cusick and Pomerance (computer-assisted); simpler proof by Bienia and others1984; 1998
6Bohman, Holzman and Kleitman; simplified by Renault2001; 2004
7Barajas and Serra2008
8Matthieu Rosenfeld (computer-assisted)arXiv 2025; Mathematics of Computation 2026
9 and 10Tanupat Trakulthongchai (an Oxford undergraduate); 9 also independently by RosenfeldarXiv 2025; Electronic Journal of Combinatorics 2026
11, 12 and 13Touch Sungkawichai and Trakulthongchai2026 preprint
14 and beyondJaan AllikvereSeptember 2026 preprint (reported; awaiting confirmation)
Any numberOpenUnproved in general

Why It Took So Long, and What Changed

There are infinitely many possible speed sets, so you cannot just check them all. Earlier results showed that if the conjecture fails, there must be a counterexample with speeds below some bound. The trouble was that the bound was astronomically large. In September 2025, Matthieu Rosenfeld combined recent bounds on minimal counterexamples with a clever computer search that rules out whole families of speeds at once, settling 8 runners. Trakulthongchai added a mathematical “sieve” that pruned the search further, reaching 9 and 10 runners in November 2025; in April 2026 he and Touch Sungkawichai posted a preprint extending this to 13 runners. Quanta Magazine covered the progress in March 2026.

Progress has also come for any number of runners. An October 2026 preprint by Xiyu Hu and Yifan Jing reports that some runner always gets a gap larger than \(\frac{1 + c}{2k}\) for an absolute constant \(c > 0\), improving the easy bound of about \(\frac{1}{2k}\) but still short of the conjectured \(\frac{1}{k+1}\).

Why It Matters

The conjecture is equivalent to statements about how lines in high-dimensional space hit grids of cubes (view obstruction), and it has applications to scheduling and to coloring “distance graphs” of integers. It is a test case for how far clever computation can push a problem that has no obvious structure.

Common Misconceptions

“It only applies to integer speeds.” The general real-speed version reduces to the integer case, so integers are enough.

“Proved for 10 runners means almost done.” Each extra runner has so far needed new ideas and more computing. The general conjecture remains open.

“The shifted lonely runner counterexamples disproved it.” No. Those concern a different, stronger variant with different starting positions, not the standard conjecture.

Watch: The Lonely Runner Conjecture in Video

If you learn better by watching, these videos are worth your time.

The Lonely Runner: an unsolved mystery of mathematics (National Museum of Mathematics). Matthias Beck explains the conjecture for a general audience.
The lonely runner conjecture in a nutshell (Ishan Banerjee). A short, visual introduction (made before the 2025 to 2026 progress).

Check Your Understanding

Five quick questions to see how much of this has stuck. Pick an answer and press Submit; you will see a short explanation either way.

Answer each question and press Submit to see your result right away.

1. With n runners, how far apart must a lonely runner be from all others?
Found a bug? Let us know!
2. Who first posed the conjecture?
Found a bug? Let us know!
3. Up to how many runners was the conjecture proved by 2008?
Found a bug? Let us know!
4. Who proved the 8-runner case in 2025?
Found a bug? Let us know!
5. With relative speeds 1 and 2, at what time are both runners 1/3 of the track away?
Found a bug? Let us know!

Frequently Asked Questions

What is the lonely runner conjecture?

If n runners with distinct constant speeds start together on a circular track of length 1, then each runner is at some time at least 1/n away from every other runner.

Has the lonely runner conjecture been proved?

Not in general. It is proved for up to 10 runners (8 by Rosenfeld, 9 and 10 by Trakulthongchai, published in 2026), and 2026 preprints report 11 to 13 runners and more.

Why is the bound 1/n the best possible?

With relative speeds 1, 2, …, n − 1, no time does better than distance 1/n, so the conjecture would be sharp.

Who proved the lonely runner conjecture for 9 and 10 runners?

Tanupat Trakulthongchai, an undergraduate at the University of Oxford, in November 2025; Matthieu Rosenfeld independently proved 9 runners.

Keep Exploring

Sources and Further Reading

Related to This Article

What people say about "Lonely Runner Conjecture Explained (2026 Update) - Effortless Math"?

No one replied yet.

Leave a Reply