CurlTi.me

CurlTi.me/Learn/Research library

The curling research library

Curling research is scattered across physics, tribology, sports medicine and operations research, and those literatures barely cite each other. This is all of it in one place, grouped by what it is about and annotated with what it actually found.

Last updated 2026-08-18

This page is a bibliography with opinions. Every entry says what the work claims, not only what it is called, so it can be read straight through by somebody who wants to know what is known and has no intention of opening a journal. Where something is open access, that is noted, because it is usually the deciding factor.

If you only read four things#

These are the entries that change what you think rather than adding to it.

Why stones curl: the physics literature#

This is the largest and the least conclusive of the four. It runs from 1924 to the present without converging, and the disagreement is genuine rather than merely unresolved — competent people have looked at the same stone and reached incompatible conclusions. The argument itself is narrated on why a curling stone curls; what follows is the paper trail, grouped by the mechanism each camp argues for.

The first look, 1924–1981#

The problem is stated, and immediately gets several different wrong answers at once.

Front–back asymmetry#

The stone leans on its trailing edge, so the back of the running band grips harder than the front. Elegant, and the direction comes out right — but the amount of asymmetry it needs is very large.

Pivot–slide#

The running band does not slide smoothly: it catches on a high point, pivots about that point for an instant, then releases. Repeat a few thousand times and the pivots add up to a sideways drift.

Scratch-guide#

The rough underside of the stone scores fine grooves into the pebble tops, and the next part of the stone to arrive is steered along them, like a skate edge running in its own track.

Measurement, and what any theory has to match#

The experimental papers. Several of these are the reason particular theories are no longer tenable.

Ice, sweeping and prediction#

Work that takes the mechanism as given and asks what follows: how ice quality can be measured, and how well a trajectory can be predicted.

The curler: physiology, biomechanics and injury#

Much smaller, much more recent, and much more likely to be immediately useful. If you have ever wondered whether the ache in your knee after a bonspiel is normal, this literature has measured it.

The curler: physiology, biomechanics and injury#

Much smaller than the physics literature, and much more directly useful. If you want to know whether your knee is supposed to hurt, this is the section.

Strategy and analytics#

The newest of the four literatures and the one most likely to tell you something a good skip would dispute. Several of these results contradict conventional wisdom, and the disagreements are worth taking seriously in both directions: the models are built on real games, and the skips have played them.

Strategy and analytics#

Win probability, when to blank, and what the numbers say about decisions curlers make by instinct. Several of these disagree with conventional wisdom.

Official documents#

The primary sources. Anything on any curling website — including this one — that disagrees with these is wrong.

Data and statistics#

Where the numbers come from, if you want to compute rather than to cite.

Elsewhere#

Not research, but primary in their own way: the people and institutions that hold the objects and the records.

Getting hold of papers you cannot open#

A good deal of the physics literature sits behind paywalls, and a curler with a question should not have to pay a journal ninety dollars to learn why their stone bends. Three things that legitimately work:

Something missing?

This list is maintained by hand and is certainly incomplete, particularly outside English. If you know of work that belongs here — especially the Japanese glaciology literature, which is substantial and under-cited in the English-language debate — it is worth adding.

The argument these papers are having Back to the guides