Practice, training and getting better
The useful range of draw weight is about seven centimetres per second of release speed. Everything worth practising follows from that one number.
Last updated 2026-08-18
The number everything follows from#
On the calibration sheet, a called 4 — the top of the house — leaves the hand at 2.126 m/s. A called 10 — the back line — leaves at 2.26 m/s.
- 0.13 m/scovers the entire house, front to back
- ~0.02 m/sis about a foot of draw
- 7 cm/sfrom the top of the house to the button
That is the sport. Not power, not aim — the ability to reproduce a slide within a couple of centimetres per second, repeatedly, on ice that is changing while you stand on it.
Two things follow immediately, and they are the whole of this page:
- Consistency beats range. A curler who throws the same slightly-wrong weight every time is far more useful than one whose average is perfect and whose spread is four feet. The first can be corrected in one sentence; the second cannot be corrected at all.
- Unmeasured practice is nearly worthless. At this precision you cannot feel the difference between a good session and a lucky one. You have to write something down.
Practice that records something#
The single change that improves most curlers' practice is bringing a stopwatch and a piece of paper.
Time every stone between the hog lines. Then, before looking at where it finished, compare the split to what the called weight should have read. That comparison separates the two failures that look identical from the house:
| Split | Finish | What actually happened |
|---|---|---|
| Right for the call | Right | A good stone. |
| Right for the call | Wrong | Your delivery was fine. The ice is not what you thought it was — go and read the sheet again. |
| Wrong | Wrong | Your delivery. This is the one to take to practice. |
| Wrong | Right | Two errors cancelling. The most dangerous outcome, because it feels like success. |
Most curlers spend years treating every miss as a delivery problem. Roughly half of them are not.
Six drills worth the ice time#
Each of these produces a number. That is the criterion for being on the list.
| Drill | What it is | Why |
|---|---|---|
| Draws to the button, in sets of ten | Both turns. Count how many finish within six feet of the tee, and write the number down. The number is the point — a session where you did not record anything is a session you cannot compare to the next one. | Measures the only thing that matters: repeatability. |
| Hog-to-hog timing on every stone | Somebody times each delivery between the hog lines. Compare the split to what the called weight should read, before looking at where the stone finished. | Separates a bad delivery from a misread sheet. Without it you cannot tell which one you just did. |
| The no-broom slide | Slide out of the hack with no stone and no brush. If you cannot do it in balance, you have been leaning on something. | The most honest diagnostic in curling, and it takes ten seconds. |
| Weight ladder | Throw a called 4, then a 7, then a 10, then back down. Not for accuracy — for whether you can feel the difference between them on demand. | Weight control is a discrimination skill, not a strength one. |
| Takeouts at one weight | Fifteen hits, all at control weight, all at the same target. Judge them on how consistent the shooter’s roll is, not on whether the target left. | Anybody can hit. Hitting the same way twice is the skill. |
| Freeze attempts | Ten stones, each trying to stop dead against a stationary stone. Expect to make very few. | The hardest shot in the game, and the one that most rewards practice nobody does. |
Eight out of ten inside six feet is a fact. "I threw well tonight" is a mood. A season of the first is a training record; a season of the second is nothing at all.
Practising the thing that is not delivery#
Delivery gets all the practice and roughly half the misses. The other half is reading the sheet, and it is practisable.
The arithmetic is on the ice speed page: a stopwatch split tells you the friction the stone actually met, and the gap between that and what the called weight should have produced converts into feet of extra or missing draw. The conversion is not the obvious one — the tempting version is wrong by roughly a factor of two — which is exactly why it repays practice rather than intuition.
Practically: time the first two or three stones of every end for a whole game and predict, out loud, where the next one will finish. Being wrong is the useful part.
The draw simulator is a reasonable way to build the intuition off the ice, because it runs the same friction model as the game and lets you see the ice-speed effect isolated from your own delivery.
Off-ice, honestly#
This section is short on purpose. Curling's off-ice needs are not exotic, and most content that says otherwise is selling something.
What plausibly helps#
- Single-leg balance. The delivery is a slow balance task on one leg with a 19 kg weight in one hand. Standing on one foot, eyes closed, is a more curling-specific exercise than most things in a gym.
- Hip and ankle mobility. The crouch is deep, and most people cannot reach it comfortably. This is usually the limiting factor rather than strength.
- Core stability. Planks, side planks, bird dogs — the standard list, and it is the standard list because it works for anything that involves holding a position while something else moves.
- An aerobic base. Sweeping is by a distance the hardest physical work in curling, and it is cardiovascular. Anything that raises your heart rate for twenty minutes counts.
- Slide practice on a floor. A slider on a smooth surface, no stone, working on the position. It transfers.
What is oversold#
Heavy strength work for its own sake. Nothing in curling is limited by how much you can lift — a stone is 19 kg and you are pushing it along ice, not raising it. Strength that supports balance and endurance is useful; strength beyond that is just strength.
Knees, backs and shoulders#
Curling looks gentle and produces a great deal of soreness. The research is small but consistent, and worth knowing before you conclude that your knee is unusual.
- 79%of competitive curlers report curling-related musculoskeletal pain
- 54%knee
- 33%back
- 20%shoulder
Sweeping and delivering are the two activities most likely to provoke it. Time-loss injuries, by contrast, are rare — about two per thousand athlete exposures — which is the signature of accumulated load rather than accidents.
Two specific findings worth acting on:
- Toe sliding loads the knee more. A pilot biomechanical study found a larger moment arm at the knee for toe sliding than for flat-foot sliding. That is not a reason never to toe-slide; it is a reason to know which one you are doing and why.
- Warm up before the first stone. Almost nobody does, and the first delivery of the night is a deep crouch on a cold knee on a cold floor.
And the structural answer: if the crouch is what hurts, a delivery stick removes it entirely and is welcome in ordinary club play. It is how a great many people keep curling for another twenty years.
The off-season#
Most clubs are dark from April to September, which is a long time to lose a slide you spent a winter building. Three things carry over:
- Floor sliding. A slider and a smooth floor keeps the position alive without ice.
- Mobility. The crouch is the first thing to go and the slowest to come back.
- Strategy. The tactical half of curling does not need ice at all. Strategy, the hammer and the numbers are all readable in July, and playing games — here or anywhere — keeps the decision-making warm even when the delivery cannot be.
Elsewhere#
- Skills and drills — CurlTechA coaching resource with more drills than are worth listing here, from people who coach.
Sources#
- Primary Self-reported injury patterns among competitive curlers in the United States British Journal of Sports Medicine, 2004The 79% figure, the knee/back/shoulder breakdown and the time-loss injury rate.
- Primary A pilot biomechanical assessment of curling deliveries: is toe sliding more likely to cause knee injury than flatfoot sliding? Sports Biomechanics, 2017The moment-arm comparison between delivery styles.
- Primary The sports science of curling: a practical review Journal of Sports Science and Medicine, 2009Open access, and the clearest account of the physiological demands of sweeping.
- An examination of studies related to the sport of curling: a scoping review Frontiers in Sports and Active Living, 2024The map of what has and has not been studied, if you want to check whether a training claim has any evidence behind it.