Force and Mobility Ellipsoids in the Golf Swing: A Geometric Interpretation Using the Double Pendulum
Abstract
In a specified golf-swing model, the force ellipsoid and mobility ellipsoid provide a geometric description of how easily the club can move and how strongly the modeled linkage can resist or transmit force at an instant. These ellipsoids are derived from the Jacobian of the multibody system and identify model-dependent directions of high mobility, low mobility, high force capability, and low force capability.
This article presents a clear, self-contained explanation of these concepts using the classical double pendulum and the mechanical advantage of an inclined plane as intuition-building analogies.
Introduction
Think of a bicycle chain. When you push on the pedal, force travels through the chain to the wheel. But here’s the key question: how much of your pushing force actually moves the wheel forward? Some of it gets “lost” in friction, in the geometry of the system, in the mechanical advantage. The force mobility matrix tells you exactly this—it’s the “transfer function” of a mechanism, showing how forces at one point create motions at another.
The golf swing is the same. Your hands are one point, the clubhead is another. The geometry of your arms and the club acts like a transmission. At any instant during the swing, the configuration of these linked bodies (torso, arms, hands, club) imposes constraints on how the clubhead can move and how forces applied to the clubhead propagate back through the joints.
These instantaneous capabilities are captured compactly by two geometric objects:
the mobility ellipsoid, describing directions in which motion of the clubhead is easiest, and
the force ellipsoid, describing directions in which the golfer can generate or resist the greatest reaction forces.
Both ellipsoids arise directly from the Jacobian matrix of the system, and they are geometric duals of one another.
Understanding these ellipsoids helps explain why certain modeled configurations offer stronger mechanical leverage than others.
In the broader context of nonlinear control, these ellipsoids describe one part of the instantaneous “Effective Plant”: the modeled mechanical pathway through which intent is expressed as force or motion. The shape of the mobility ellipsoid can inform Drift-Control Ratio (DCR) analysis because high-mobility directions are often directions where small joint motion produces large task-space motion. Conversely, the force ellipsoid identifies directions where the linkage can transmit larger reaction forces in the model. Whether golfers exploit a specific ellipsoid alignment at impact is a hypothesis that requires measured swing data.
The Jacobian: The Mapping Between Joint Space and Club Motion
Let \(q \in \mathbb{R}^n\) denote the joint coordinates of the upper body, arms, wrists, and club.
Let \(x \in \mathbb{R}^6\) denote the instantaneous pose (position and orientation) of the clubhead.
The kinematic map is \[ x = f(q) , \] and the differential relationship is \[ \dot{x} = J(q)\, \dot{q}, \] where \(J(q)\) is the \(6 \times n\) Jacobian matrix.
Each column of \(J\) describes the infinitesimal motion of the clubhead resulting from a unit velocity in a single joint.
Thus, the geometry of \(J\) encodes the instantaneous mechanical leverage of the entire body at that moment.
The Mobility Ellipsoid
The mobility ellipsoid is formed by mapping a unit ball of joint velocities through the Jacobian: \[ \mathcal{E}_{\text{mob}} = \left\{\, \dot{x} = J\dot{q} \; \bigg| \; \|\dot{q}\| = 1 \,\right\}. \]
This produces an ellipsoid in task space whose principal axes represent: - directions of high mobility (large end-effector motion for a small joint motion),
- directions of low mobility (joint motion produces little club motion).
In the context of the control-affine derivation (\(\dot{x} = f(x) + G(x)u\)) presented in Part 1, this ellipsoid is a geometric representation of input efficacy in the chosen model. The principal axes of the mobility ellipsoid correspond to directions where the input map can produce larger task-space motion for a given joint-rate constraint. They should not be interpreted as a complete measure of biological agency, because muscle capacity, timing, noise, and inertia are not fully captured by the kinematic Jacobian alone.
These directions change continuously throughout the golf swing.
Near impact, the model predicts high tangential mobility, which is one reason small joint motions can have large task-space effects.
Synthesis: Drift Dominance This geometric mobility is one kinematic contributor to Drift Dominance in the model. As the mobility ellipsoid elongates along the tangential path, accumulated momentum can become large relative to available control. The resulting “steering” limit is a model prediction that should be separated from direct claims about a golfer’s perception or intent.
The Force Ellipsoid
Reaction forces at the clubhead map back to joint torques via the transpose of the Jacobian: \[ \tau = J^\top F. \]
The force ellipsoid is defined as \[ \mathcal{E}_{\text{force}} = \left\{\, F \in \mathbb{R}^6 \; \bigg| \|J^\top F\| = 1 \,\right\}, \] which describes all end-effector force directions that correspond to unit joint torque outputs.
Its principal axes represent: - directions in which the system can generate large force or resist external load (high constraint),
- directions in which very little force can be produced or resisted (low constraint).
Synthesis: Intentional Constraint Collapse The duality of these ellipsoids offers one possible mechanical interpretation of Intentional Constraint Collapse. If a body-club configuration aligns high force-transmission directions with impact-relevant loads, the model predicts greater passive resistance in those directions. Whether elite golfers reliably create this alignment, and whether it stabilizes the clubface in measured swings, remains an empirical question.
Algorithm: Computing the Ellipsoids
Given a Jacobian \(J \in \mathbb{R}^{m \times n}\) (task-space dimension \(m\), joint-space dimension \(n\)), the singular-value decomposition \[ J = U\,\Sigma\,V^\top, \qquad \Sigma = \operatorname{diag}(\sigma_1, \sigma_2, \dots, \sigma_r), \quad \sigma_1 \geq \sigma_2 \geq \dots \geq \sigma_r > 0, \] provides the principal-axis decomposition for both ellipsoids:
| Ellipsoid | Defining set | Principal axes | Semi-axes |
|---|---|---|---|
| Mobility \(\{\dot{x} = J\dot{q} : \lVert\dot{q}\rVert \leq 1\}\) | image of joint unit ball | columns of \(U\) | \(\sigma_i\) |
| Force \(\{F : \lVert J^\top F\rVert \leq 1\}\) | dual via \(J^\top\) | columns of \(U\) | \(1/\sigma_i\) |
The two ellipsoids share the same principal axes (columns of \(U\)) but their semi-axes are reciprocals — this is the origin of the mobility-vs-force trade-off. (Strictly, the reciprocal-semi-axis picture describes a bounded force ellipsoid only when \(J\) has full row rank. If \(J\) is rank-deficient or has more task dimensions than independent joint contributions, \(J^\top F\) has a nontrivial null space and the set \(\{F : \lVert J^\top F\rVert \le 1\}\) is an unbounded cylinder along those uncontrollable force directions; the \(1/\sigma_i\) description then applies only within the range of \(U\).) A direction with a large \(\sigma_i\) is a direction of easy motion and weak force resistance; a direction with a small \(\sigma_i\) is the reverse. At a singular configuration (\(\sigma_r \to 0\)), the mobility ellipsoid flattens (motion in that direction becomes unachievable) and the force ellipsoid blows up (infinite reaction force required to maintain joint-torque bound).
import numpy as np
def ellipsoid_axes(J):
"""Compute principal axes and semi-axes of mobility / force ellipsoids.
Returns (U, sigma, mobility_semiaxes, force_semiaxes).
force_semiaxes is np.inf at singular directions (sigma == 0).
"""
U, sigma, _Vt = np.linalg.svd(J, full_matrices=False)
with np.errstate(divide="ignore"):
force_semiaxes = np.where(sigma > 0, 1.0 / sigma, np.inf)
return U, sigma, sigma.copy(), force_semiaxes
# Example: 2-link planar arm, L1 = L2 = 1, elbow angle theta2.
def planar_2link_J(theta1, theta2, L1=1.0, L2=1.0):
c1, s1 = np.cos(theta1), np.sin(theta1)
c12, s12 = np.cos(theta1 + theta2), np.sin(theta1 + theta2)
return np.array([
[-L1*s1 - L2*s12, -L2*s12],
[ L1*c1 + L2*c12, L2*c12],
])
J = planar_2link_J(theta1=0.3, theta2=1.2)
U, sigma, mob, force = ellipsoid_axes(J)
# sigma ~ [1.86, 0.50] at this posture: mobility ellipsoid stretches ~1.86 along u_1
# and shrinks to 0.50 along u_2; the force ellipsoid is the reciprocal.Near the fully extended configuration (\(\theta_2 \to 0\)), \(\sigma_2 \to 0\): the arm becomes singular and the force ellipsoid degenerates to an infinite strip along \(u_2\) (the axis of the straightened links). This recovers the planar intuition from the next section.
The Double Pendulum Analogy
The golf swing is often modeled as a double pendulum: an upper arm segment of length \(L_1\) and a club segment of length \(L_2\) connected by hinged joints.
When the two links are nearly collinear, the system exhibits characteristic features:
High Mobility Perpendicular to the Links
A small rotation at the proximal joint produces a large transverse motion of the clubhead.
This corresponds to a large principal axis of the mobility ellipsoid pointing perpendicular to the aligned links.
Mathematically, when the links are collinear, the moment arm between joint rotation and distal motion is maximized: \[ v_\perp \propto (L_1 + L_2)\, \dot{\theta}. \]
Low Force Capability in the Same Perpendicular Direction
However, the ability to resist external force in that same perpendicular direction is very poor.
Because joint torques must react forces through long, nearly straight lever arms, a modest force at the clubhead creates a large torque at the joint: \[
\tau \approx (L_1 + L_2)\, F_\perp.
\]
Thus, the system has: - high mobility transverse to the links, and
- low force capability in that very same direction.
This is exactly what the mobility and force ellipsoids reveal.
High Force Capability Along the Axis of the Links
Conversely, when an external force is applied along the axis of the aligned links, the moment arm is tiny.
A substantial axial force produces only a small joint torque.
This direction becomes a principal axis of the force ellipsoid.
Thus: - motions along the axis have low mobility, but
- the system can generate or resist large forces along that axis.
The Key Insight
The directions of high mobility and high force resistance are orthogonal:
what the system can move easily, it cannot push against strongly;
what the system can push against strongly, it cannot move easily.
Connection to the Inclined Plane
The mechanical advantage of an inclined plane provides an even simpler analogy.
Consider pushing a mass up a ramp of angle \(\theta\):
\[ F_{\parallel} = mg \sin\theta, \qquad F_{\perp} = mg \cos\theta. \]
As the ramp becomes shallow (\(\theta \to 0\)): - motion (mobility) along the ramp becomes easy,
required force along the ramp becomes small,
resistance perpendicular to the ramp becomes large.
The ramp is simply a constraint surface, and its geometry dictates both the ease of motion and the availability of resistive reaction forces.
Likewise, the kinematic configuration of the golf swing imposes geometric constraints that shape the mobility and force ellipsoids.
Why Force and Mobility Ellipsoids Are Duals
The force and mobility ellipsoids are duals because the Jacobian and its transpose are linked by instantaneous power conservation: \[ F^\top \dot{x} = \tau^\top \dot{q}. \]
Substituting \(\dot{x} = J\dot{q}\) gives \[ F^\top J \dot{q} = (J^\top F)^\top \dot{q}. \]
Since this holds for all \(\dot{q}\), we obtain \(\tau = J^\top F\) and the fundamental duality between motion and force mappings.
This duality guarantees that:
directions of high mobility are directions of low force capability,
directions of low mobility are directions of high force capability.
This relationship is not an empirical observation but a geometric consequence of the system constraints.
Synthesis: Null Space and Parallel Mechanisms
While the double pendulum models the arm-club system as a serial chain, the two hands on the grip technically form a Parallel Mechanism (a closed loop). This creates an “over-constrained” system where internal forces can exist in the Null Space of the Jacobian. These internal forces (co-contraction between the hands) do not produce motion (mobility) but do reshape the stiffness of the joint, effectively inflating the Force Ellipsoid in specific directions without changing the kinematic configuration. This allows the golfer to modulate the “Effective Plant” impedance independent of the trajectory.
Video Demonstration
The following video provides a visual demonstration of mobility and force ellipses in action:
Connection to Drift-Control Decomposition
The force and mobility ellipsoids are more than geometric curiosities: they describe configuration-dependent mappings between generalized and task-space quantities. Mobility alone does not establish that drift dominates a direction. That comparison also requires the current state, applied-input bounds, reference point, and selected norm.
Near impact, a model may combine high task-space mobility with large velocity-dependent drift. Whether the available control is small or large relative to that drift is an empirical model result, not a consequence of the ellipsoid alone. The hand-path attribution study defines the same-state comparison needed to test it.
Conversely, in low-mobility directions, the force ellipsoid bulges outward: the model predicts greater structural resistance but less ease of motion. This is where passive stability may arise. The stronger claim that elite golfers deliberately exploit a particular ellipsoid alignment should be treated as a testable hypothesis, not as a conclusion established by this article.
Conclusion
The mobility and force ellipsoids provide a precise geometric language for understanding the instantaneous capabilities of the golf swing. Using the double pendulum analogy, we see that when the club and lead arm are aligned, the system moves easily in a direction perpendicular to the links but resists force most strongly along the links. This illustrates the universal rule: mobility and force capabilities are geometric duals arising from the Jacobian and its constraints.
These ellipsoids help explain why release, impact loading, and mechanically strong or weak positions can appear in a modeled swing. They do not replace physiological analysis: muscle capacity, tissue compliance, timing, and measurement uncertainty still determine what motions and forces are possible for a real golfer.
This analysis assumes rigid body dynamics. The Jacobian \(J(q)\) describes how rigid segments connected by ideal joints transform motion. But real bodies are not rigid: the shaft flexes, the wrist ligaments stretch, bone deforms under load. These compliances shift the effective “point of rotation” and couple dynamics between segments in ways the rigid-body Jacobian misses. For high-speed impacts, compliance becomes substantial, and the elegant geometry of ellipsoids becomes muddied by elastic effects we have ignored.
Mobility ellipsoid conditioning degrades near singularities. When the links approach a singular configuration (nearly straight line for a 2-link system), the Jacobian’s numerical conditioning worsens and the ellipsoid becomes arbitrarily thin. Mathematically this is correct—mobility vanishes in singular directions. But numerically and biologically, systems near singularity are unstable and difficult to control, so the prediction that “you can move freely perpendicular to the singular direction” becomes practically irrelevant. The mathematics is honest; the practical implication is that singular-adjacent regions are simply fragile.
Static analysis does not capture dynamic coupling effects. The force and mobility ellipsoids are based on the Jacobian alone, which encodes instantaneous kinematics. But in a dynamic system, inertia (\(M(q)\)) and Coriolis forces (\(C(q,\dot{q})\)) create coupling that the kinematic ellipsoids do not show. A direction might have high kinematic mobility but extreme inertial resistance—you can move there kinematically but your inertia won’t let you accelerate in that direction quickly. The ellipsoids tell you the skeleton’s geometric possibilities, not the body’s dynamic capabilities. We call this the “effective plant,” but we measure only half of it.
The golf swing is a trade-off between moving fast and resisting force. This article explains the geometry of that trade-off.
Mobility (The Speed Zone)
"Mobility" describes the directions where a small movement of your joints creates a huge movement of the clubhead. This is what you want for speed.
Force (The Power Zone)
"Force" describes the directions where your body is stiff and strong, capable of withstanding the violence of impact.
The Great Trade-off
You can't be fast and strong in the same direction at the same time. Math shows these two qualities are opposites. If you are set up to move the club fast (high mobility), you are weak against recoil (low force).
Real-World Analogies: Configuration-Dependent Behavior
The force-mobility matrix reveals something profound: the relationship between input (force) and output (motion) depends entirely on your current configuration. Here's how this appears in everyday life:
- Drift-Control Ratio (DCR): How mobility ellipsoids quantify the battle between passive inertia and active control. (See:
articles/controllability-drift-ratio.qmd) - Intentional Constraint Collapse: The strategy of aligning the force ellipsoid’s major axis with impact loads. (See:
articles/intentional-constraint-collapse.qmd) - Null Space: The mathematical home of internal forces in parallel mechanisms. (See:
articles/null-space-constraint-jacobian.qmd) - The Effective Plant: How geometry dictates the baseline mechanical impedance.
Every theory faces scrutiny. Here's what skeptics and alternative perspectives say:
The Static Fallacy (Kinematic vs. Dynamic)
Skeptics argue that force and mobility ellipsoids are derived solely from the kinematic Jacobian (J), which describes velocity mappings. However, in a high-speed ballistic motion like the golf swing, dynamic manipulability (which accounts for the Mass Matrix M and inertia) is far more relevant. A direction might have high kinematic mobility but extreme inertial resistance. Relying on J alone risks confusing "geometric ease of motion" with "dynamic ease of acceleration."
The Biological Actuator Problem
The duality relation τ = JTF assumes ideal torque generators that can produce force equally well in any configuration. Muscles are not ideal motors; they have non-linear force-length and force-velocity properties. Critics point out that a "strong" direction in the theoretical Force Ellipsoid might be biologically weak if the primary driving muscles are at a mechanical disadvantage or are contracting at high speeds where their force output drops precipitously.
Planar Reductionism vs. 3D Reality
The article relies heavily on the double-pendulum analogy, which is a planar (2D) system. Critics argue that the most critical stability issues in golf—such as "coming over the top," shaft droop, and face closure—are inherently 3D phenomena involving out-of-plane forces. Planar ellipsoids fail to capture the cross-coupling terms (e.g., how arm elevation affects wrist radial/ulnar deviation leverage) that are essential for a complete understanding of the swing.