CurlTi.me

CurlTi.me/Learn/Ice speed

Ice speed, and what to expect from a sheet

No two sheets are the same, and no sheet is the same at the end of a game as it was at the start. Reading the ice is not mysticism — it is arithmetic done with a stopwatch, and this is the arithmetic.

Last updated 2026-08-16

Keen and heavy#

Ice speed has one meaning: how far a stone runs for a given delivery. Ice that lets it run a long way is keen, or fast. Ice that stops it short is heavy, or slow. That is the whole vocabulary, and it is worth being precise about because the stopwatch runs the other way round from the intuition:

The bit that catches everyone

On keen ice a draw to the button takes a longer hog-to-hog time, not a shorter one. The stone has to be thrown more gently to finish in the same place, so it spends longer crossing the 72 feet between the lines. Fast ice, slow watch.

How much difference does it make?#

A great deal more than people expect. Here is a single delivery — identical release speed, identical everything — on three sheets:

The same delivery on three different sheets. On club ice it stops well short of the house; on the calibration sheet it stops on the button; on championship ice it runs sixteen feet past.Tee lineBack lineClub iceNormal iceChampionship ice24.9 ft between the same stone on two sheetsone delivery, released at exactly the same speed
One delivery, three sheets. The gap between the slowest and the fastest is nearly twenty-five feet, which is twice the width of the house.
SheetFrictionThat delivery stops atIts own draw to the buttonWhat it feels like
Club×1.10117.5 ft12.72 sHeavy. Everything falls short until you adjust, and guards are easier to reach than the house.
Normal×1.00126.0 ft13.31 sThe calibration sheet: a draw to the button is 13.3 seconds hog to hog.
Championship×0.85142.4 ft14.37 sKeen and swingy. Big curl, delicate weight, and a stone that keeps going.

Ten per cent more friction — the gap between club ice and a well-kept sheet — costs eight and a half feet of draw. That is the difference between sitting on the button and sitting on the hog line side of the rings, from a delivery that felt identical in the hand.

A useful identity

Friction and draw distance are exact reciprocals. Scale the friction by k and the stone travels exactly 1/k as far — the whole slide becomes a time-rescaling of the original one. So a sheet 3% keener draws 3% further, and nothing has to be measured to know it.

A sheet is not even the same as itself#

Two separate things move, and they are worth keeping apart because they feel different from the house.

Break-in: the first few ends#

Fresh pebble is sharp. The first stones and brushes of a game polish the tips down and knock the sharpest ones off, and the sheet gets keener as it happens. A newly pebbled sheet starts roughly 3.5% slower than it will settle at — about three feet of draw — and picks that up over the first two to four ends. Ice technicians describe a sheet "picking up a foot or two" early on; this is that.

Practically: your first end is on different ice from your third end, and a team that keeps throwing the weight that worked in the first end will be through the back of the house by the fourth.

And the closing ends#

Late in a game the pebble can be worn down far enough that the stone's running band rides on more of it, which slows the sheet again. In an arena, the humidity a crowd brings can put frost down, which does the same. It is a smaller effect than the break-in and it does not happen every game — but when it does, it happens in the ends that matter.

Reading a sheet from one stone#

Here is the actual technique, and it is simpler than it is usually made to sound. Somebody times the stone between the two hog lines. You know what weight was called, and you know what split that weight should produce. The gap between the two is the ice.

  1. The skip calls a weight — say a 7, a draw to the button.
  2. The chart says a 7 is a 13.31 second split.
  3. The watch reads 13.9.
  4. The stone took longer, so it met less friction than the chart assumed. The sheet is keen, and the stone will run long.

The conversion from that gap to feet is not the obvious one, and getting it wrong is a common way to over-correct. The tempting arithmetic — "what release speed would have produced that split, and where would that stone stop?" — is wrong by roughly a factor of two. A keener sheet helps the stone over its whole slide, whereas a heavier release only buys it a speed surplus that friction then eats at the usual rate. What the split is evidence of is the friction the stone actually met, and that is what has to be solved for.

CurlTi.me does this arithmetic for you and reports the answer as feet of extra draw, because that is the number a curler wants. It also cannot tell the ice apart from the thrower's hands — and neither can a stopwatch on real ice. One stone is evidence, not an answer. Two or three agreeing is an answer.

How much a sheet curls#

Speed is only half of what a sheet does. The other half is how far a stone bends, and this one is genuinely unsettled.

These do not agree, and the reason is that no published study has ever measured a championship-prepared sheet. Rather than pick a side, CurlTi.me exposes it as a setting: the calibration sheet curls the measured amount, and the championship preset curls 60% more.

Ice that curls a lot is called swingy; ice that curls little is straight. It changes which shots exist. On swingy ice a come-around is a routine call and guards genuinely protect what is behind them. On straight ice guards are much less useful, because a stone can be driven almost straight past them, and the game turns into hitting. See strategy.

When a sheet is not flat#

Ice is made by people, in a building, and it has character. The two names worth knowing:

Both are read the same way: by watching what stones actually do, rather than by arguing about what they should do. Good teams build a picture of the sheet over the first two ends and then trust it.

Ice on CurlTi.me#

Three presets — club, normal and championship — and two switches. One rolls each game's sheet to settle somewhere within about ±3% of its preset, which is ±2.9 feet on a draw to the button: enough that the first draw of a game is a real question, not enough that it is a lottery. The other turns on the break-in and the late slowdown described above.

Both are seeded and deterministic, so both players are on the same sheet, and a replay is played on the ice the game was played on rather than on a fresh roll of the dice.