Muscle model for the compound bow: Ariffin 2016

This is a mathematical model, with no measurements on archers described in the paper. The authors represent both arms and the upper trunk as segments with six joints (wrist, elbow and shoulder on each side) and twelve muscles, six per arm, and look for the combination of muscle forces that produces the joint moments of each phase of a compound shot while keeping muscle effort to a minimum, expressed as the sum of the squared forces divided by each muscle's cross-sectional area.

Six phases are considered, from set-up to release. For the drawing arm the minimum sum is lowest at release and highest during set-up. In the model the right middle trapezius works only during set-up and then switches off, replaced by the biceps and middle deltoid, which the authors put down to the compound's cams carrying the load. For the bow arm the main contributors are the left middle trapezius, whose force goes from 4118 at set-up to 7882 at release (with no units given), and the triceps. The authors argue that this kind of calculation could help reduce the risk of shoulder injury.

The method is static optimisation using Lagrange multipliers. Each joint has a constraint: the sum of muscle forces multiplied by their moment arms must equal that phase's joint moment. Moment arms and muscle cross-sections come from the literature (for example 11.91 cm² for the biceps and 19.07 for the triceps). The version using cubed forces was dropped because it produced complex numbers, so the results concern only the squared version. Since the sign of the multipliers is not known in advance, the authors consider eight sign combinations and list for each which muscles come out active, inactive or undetermined.

The paper does not say where the joint moments for the six phases come from: no archers, recordings, bows or draw weights are described. The abstract mentions an amateur archer, without further detail.

Without knowing where the joint moments come from, the results cannot be assessed: they are the input on which everything depends. Muscle forces are given without units, and some are negative (for example −4128 for the right flexor carpi radialis during set-up) even though the model requires them to be zero or positive: the method used does not enforce this constraint. The model is highly simplified, with few muscles per joint and the scapula reduced to the middle trapezius, whereas many muscles work together in a shot.

The paper is not internally consistent: the conclusion says the drawing arm uses the middle trapezius, deltoid and biceps during the draw, but the table puts the middle trapezius at zero in every phase except set-up; and it refers to three joints where the method describes six. The link with injury is asserted, not measured.

For archers nothing changes. This is an exercise in applied mathematics on a model of the arms with a compound bow, with no experimental data described, with negative forces that the model itself should rule out and conclusions that do not match the tables. The idea that the compound's cams take load off the trapezius after set-up is intriguing, but this calculation cannot support it. On shoulder muscle activity in compound shooting, electromyographic measurements on real archers are a sounder starting point.