Elite curling in biomechanical terms: dorsiflexion, contact area and slide-phase muscle synergy, timing diagnostics and load management via session RPE.
Elite curlers differ from sub-elite not through strength but through measurable control of the slide phase: greater ankle dorsiflexion, therefore a larger contact area of the sliding foot on the ice, and a positive rather than negative moment of the centre of mass [1]. That is the finding of a biomechanical comparison between elite and sub-elite curlers — and it is uncomfortable, because it leaves nothing to hide behind.
Plain speech: your scatter is not a talent problem. It is a coordination and diagnostics problem. Both are addressable.
How fine the resolution is at this level shows in the timing layer: a difference of 0.1 seconds in split time moves the stone about six feet, nearly two metres, at the far end [4]. Anyone who wants to play at elite level works inside a window of a few hundredths of a second — and measures it instead of feeling it.
And the bar sits high. In competitive analytics, 88 to 90 per cent shot percentage — the share of stones executed as called — counts as the elite band; a skip at 92 per cent is outstanding, and teams below 88 per cent as a unit essentially do not medal at major events [7]. Two percentage points of execution separate the semi-final from the stands here.
Those numbers come from competition analytics, not from a controlled study — they are an order of magnitude, not a law of nature. The biomechanics underneath, by contrast, are measured.
The elite marker is mechanically simple: more ankle flexion extends the metatarsophalangeal joint further, the slider lies flatter on the ice, the glide steadies and the centre of mass works with you instead of against you [1]. Measure that in side-on video as an angle and as slider contact length — not as a feeling. If your dorsiflexion collapses under fatigue, your consistency collapses with it.
A study of 16 professional curlers using 3D motion analysis and wireless muscle-activity recording from eight sliding-leg muscles compared three propulsion and sliding variants — flat full-foot contact, outward-toed full-foot contact and toe contact. The result: each variant forces its own joint kinetics and its own muscle synergy patterns, which feeds straight through to accuracy and stability [2]. Translated into a training decision: pick one variant, automate it across thousands of repetitions and do not switch it mid-season.
The arithmetic is known: in a toe slide the knee joint force is more than twice that of a flat slide, and it correlates strongly with the lever arm of the ground reaction force — measured via in-shoe pressure distribution and sagittal video [3]. Anyone running the more aggressive variant has to offset the hip and knee load in strength work and volume management. Not a ban — a bill.
At elite level the clock is an instrument. Draws typically sit at 13.0 to 14.5 seconds hog to hog, takeouts depending on weight at around 9 to 11 seconds [4]. Log target time, actual time and deviation per session — and then the standard deviation across 20 stones. The mean lies, the scatter does not.
A finding with real leverage at this level: in curlers, the accuracy of duration judgment relates to delivery performance [5]. So train blind: call the time before the watch shows it. Work on stroboscopic visual training in elite curlers points in the same direction, stressing visual information processing under degraded conditions [5].
Session RPE — session duration times perceived exertion — is validated in Olympic curling: the index correlated significantly with external load measures (p < 0.01) and with cortisol and creatine kinase (p < 0.05) [6]. In follow-up work, a metric derived from it (an exponentially weighted moving average) classified injury risk with 90.0 per cent accuracy (AUC 0.899) [6]. — here that becomes measurable.
At this level, injury prevention is part of performance planning, not an appendix to it. The delivery markedly raises peak hip abduction, knee flexion and the associated joint moments and demands coordinated activation of hip abductors, knee flexors and extensors and external rotators [2] — which is why high stone volumes without accompanying hip and knee strength work end in overuse.
Use the diagnostics for that: an injury-risk classification at 90.0 per cent accuracy from session-RPE trajectories [6] is only worth something if you actually react to a rise — cut volume rather than grit your teeth.
And despite all the numbers: this guide provides metrics and principles but never replaces coaches, performance diagnostics and sports-medical support. Anyone changing technique variants under high load does it supervised.
Build a fatigue diagnostic into your technique training: 20 stones fresh, then three minutes of hard sweeping, then 20 stones immediately. Compare not the hit rate but the standard deviation of your split times before and after. Exactly that rise in scatter is why games tip in the eighth end — and it is trainable, because it traces back to balance control and ankle work [1]. Anything that shrinks that difference is genuine elite training. Everything else is occupation.
Above all control of the slide phase. In the comparison study, elite curlers showed greater ankle dorsiflexion, a larger sliding-foot contact area and a positive moment of the centre of mass, while sub-elite showed a negative one [1]. The result is better control of speed and balance — not more strength.
That is a cost-benefit decision, not an article of faith. The toe slide produces more than double the knee joint force of a flat slide, strongly correlated with the lever arm of the ground reaction force [3]. And each sliding variant forces its own joint kinetics and muscle synergy patterns [2]. Choose, automate, and offset the load in strength training.
Four series: split and hog-to-hog times with standard deviation [4], shot percentage per game [7], dorsiflexion angle and slider contact from side-on video [1], and session RPE as a load measure [6]. The mean shows your level, the scatter shows your potential.
In competition analytics the elite band sits at 88 to 90 per cent, a skip at 92 per cent counts as outstanding, and teams below 88 per cent as a unit essentially do not medal at major events [7]. Important context: these are analytics-derived orders of magnitude from competition data, not controlled study values.
Your Delivery on Video: The Three Elite Markers You Can Actually See