Arrow properties — what the numbers measure, what the shot reveals

The unrestrained tip

At the moment the string moves, the nock is driven forward. The tip has no idea this is happening yet.

The nock is restrained — it is in contact with the string, and the string is applying force. The tip is unrestrained — there is nothing pulling or pushing it, only its own inertia holding it in place. For a brief but real period of time, the nock is trying to go somewhere the tip is not yet going. The shaft connecting them is caught in between.

That disagreement — driven nock, inertial tip — creates axial compression along the shaft. The nock end is being pushed. The tip end is resisting. The shaft between them is being compressed end to end. A slender column under end compression does one thing: it bends. The arrow flexes.

This is the root cause of everything that follows — the ‘archer’s paradox’ (the common term for the launch oscillation a compound arrow undergoes, borrowed from recurve but widely used for both), dynamic spine, oscillation, the entire field of spine selection. It all starts with the unrestrained tip.

What the spine number actually measures

Before getting to forces during the shot, it helps to know exactly what a spine number is — because it is not a strength measurement. It is a static deflection measurement.

The AMO standard is simple: support the shaft at two points 28 inches apart. Hang a 1.94 lb weight from the center. Measure how far the center deflects in thousandths of an inch. That number is the spine. A 400-spine arrow deflects 0.400 inches. A 1000-spine arrow deflects 1.000 inch.

The measurement describes bending stiffness — how much the shaft resists bending under a sideways load. Engineers call the relevant property EI: E is the material's elastic modulus (how stiff the material itself is), and I is the second moment of area (how the cross-section is shaped). A thicker wall produces a higher I. A stiffer material produces a higher E. Together they determine how much an arrow bends.

From the AMO measurement, EI can be calculated directly. For a simply-supported beam with a central point load:

The spine number is a proxy for EI. A lower number means higher EI means stiffer shaft. A 300-spine arrow is stiffer than a 400-spine arrow. The number is backwards from what the word "stiff" implies, which trips up everyone eventually.

The other static properties

Spine is one measurement. Three others matter alongside it — and together these four numbers give a complete picture of the arrow before it is shot.

GPI — mass per unit length

GPI is grains per inch: the bare shaft weight divided by the cut length. A 29-inch shaft that weighs 261 grains has a GPI of 9.0. A shaft that weighs 218 grains at the same length has a GPI of 7.5. Both can be labeled 400-spine. They are not the same arrow.

GPI appears on manufacturer spec sheets but almost never in spine selection discussions. It should. Natural oscillation frequency — the property that governs group consistency at distance — depends on stiffness divided by mass per unit length. Two shafts with identical spine but different GPIs oscillate at different frequencies. They will not group together at distance regardless of how well they are otherwise tuned.

Straightness tolerance

Commercial carbon arrows carry a stated straightness tolerance: .001″, .003″, or .006″ — the maximum allowed deviation from true straight, measured along the shaft at the factory. This is a manufacturing consistency specification, not a flight performance prediction. Whether the finished arrow — with point, insert, nock, and vanes installed — holds that tolerance depends entirely on the build process.

A shaft that is not perfectly straight has a small offset between its geometric axis and its mass center. As the arrow spins in flight, that offset creates a cyclic imbalance that perturbs the oscillation and produces a slightly different tip path on each shot. At 20 yards the effect is negligible. At 80 meters into a crosswind it compounds. Straightness tolerance is part of the consistency stack, not separate from it.

The spine plane — weak and stiff axes

Carbon shafts are not equally stiff in all directions. The fiber layup — the orientation of carbon strands within the resin matrix — creates directional stiffness variation in the cross-section. One plane through the shaft axis is slightly stiffer than the perpendicular plane. This is the spine plane.

The practical test: hold the nock end firmly — pinched at the very end, or in a vise with rubber pads — extend the arrow horizontally, and give the tip a sharp sideways flick. The arrow oscillates in its easiest plane: the weak axis. Rotate 90° and it resists more. The plane of free oscillation is where it wants to bend. Mark it.

For recurve archers, installing the nock so the weak axis aligns with the plunger deflection direction makes the arrow respond consistently to the same flex geometry each shot. For compound archers with a centered drop-away rest the alignment matters less mechanically, but rotating all arrows in a batch to the same spine plane orientation removes one source of inter-arrow variation. If every arrow in the dozen bends in the same plane relative to the rest, the oscillation behavior is at least consistent across the set even if the absolute frequency is not perfectly matched.

The column that should not survive

Here is where it gets interesting. The spine number describes how an arrow bends under a sideways load. But during the shot, the arrow is under an axial load — compression along its length. These are different problems.

In 1744, Leonhard Euler derived the formula for how much axial load a slender column can carry before it buckles. It depends on EI and the length of the column:

The same calculation for every common spine at 29" arrow length:

Spine	Static buckling load	Context
150	69.4 lb	Stiff recurve / heavy hunting
170	61.3 lb
200	52.1 lb
250	41.6 lb
260	40.0 lb
300	34.7 lb	Recurve target
340	30.6 lb
350	29.7 lb	Compound hunting baseline
400	26.0 lb	Standard compound target
500	20.8 lb	Light compound target
600	17.3 lb
700	14.9 lb
1000	10.4 lb	Would buckle in your hand

Calculated at 29″ arrow length with pin-pin boundary conditions. Actual boundary conditions during the shot differ — see below.

A 70 lb bow. A standard compound target arrow with a 400 spine. The static buckling load is about 26 lb. The bow produces roughly 70 lb of peak compressive force on the nock. Statically, this arrow has no business surviving the shot. It should fail at roughly a third of the applied load.

It doesn't. The reason is fundamental to how arrows work — and it is the part nobody explains.

Dynamic loading — why arrows survive

Clarence Hickman at Bell Labs filmed arrow departure with high-speed cameras in 1929 and 1930. For the first time, it was possible to see what actually happens. The arrow bends into an S-curve as it leaves the bow — flexing around the riser on a recurve, oscillating on a compound — and then gradually recovers to straight flight. The bending is real and visible. The shaft is genuinely flexing well beyond what static analysis would predict is safe.

The reason it survives is dynamic buckling theory. A column that would fail under a sustained static load can survive the same load applied briefly, because the buckling mode does not have time to fully develop. Buckling is not instantaneous — the column has to bend, that bend has to amplify, and the amplification has to reach structural failure. Under a static load, you can wait indefinitely for this to happen. Under a brief impulse, the load may be gone before the amplification completes.

A typical compound shot takes approximately 12 to 15 milliseconds from string release to arrow departure. During that time the arrow flexes through its first half-cycle and is already leaving the bow before the bending can develop into structural failure. The arrow survives not because the forces are low — they are not — but because the forces are brief.

This is also why a grossly underspined arrow is dangerous rather than merely inaccurate. An arrow that flexes dramatically enough to contact the riser, cable, or rest during those 15 milliseconds receives a sharp lateral impact that carbon fiber is poorly equipped to handle. The failure is not axial crushing. It is side-load fracture from contact. Underspine by a small amount means inconsistent oscillation phase. Underspine by a large amount, on a poorly centered setup, means the arrow may physically hit part of the bow during the shot.

The force on the arrow during the shot

The compressive force on the arrow's nock end is not simply the bow's peak draw weight. It varies continuously through the power stroke, and the shape of that variation is different for compound and recurve.

On a recurve, the draw force curve is roughly triangular — high at full draw, dropping linearly toward brace height. The arrow starts under high string force and that force decreases as the arrow accelerates forward. Peak compressive load on the nock is early in the shot.

On a compound, the draw force curve has let-off. At full draw, the holding weight is a fraction of the peak — typically 10 to 20 lb on a 70 lb bow. When the string releases, the cams begin rotating back through their cycle. As they do, the string force rises from the holding weight back toward the peak draw weight, then drops back down as the arrow approaches brace height. The arrow actually experiences close to peak draw weight near the middle of the power stroke — briefly, but genuinely.

Here is the useful approximation: at any moment during the shot, the force at the nock equals the total arrow mass times the instantaneous acceleration. A 350-grain arrow on a 70 lb bow, reaching approximately 300 fps across a 26-inch power stroke, peaks at roughly 1,200 to 1,400 times the acceleration of gravity. At 350 grains total weight and peak acceleration:

The compressive load is not uniform along the shaft. The nock end carries the maximum compression — the full string force minus the small nock mass. At any cross-section along the shaft, the compressive force equals the mass of everything forward of that point times the instantaneous acceleration. At the tip, compression drops to zero — there is nothing forward of the tip to accelerate. The shaft behaves as a loaded column with a distributed mass, under a time-varying end force.

If your front weight is 120 grains out of 350 grains total — a typical target compound setup — roughly 34% of the string force at any moment shows up as compression near the tip end of the shaft, and approximately 100% shows up at the nock. This is why the nock end of an arrow is under the most structural stress during the shot, and why arrows, when they do fail, typically fail near the nock rather than at the tip.

What spine selection is actually about on compound

On a recurve, the primary job of spine selection is managing the archer's paradox. The arrow must flex enough to clear the riser and rest as it leaves the bow. Too stiff and it clips the riser. Too weak and it over-flexes and does not recover cleanly. The correct spine produces a specific amount of flex at a specific time in the shot, and the feedback — a directional miss, a paper tear, a bare shaft that angles off — is directly observable.

On a modern compound with a centered rest and a drop-away, the paradox is largely eliminated. The arrow does not need to navigate around a riser. The compressive load still flexes the shaft, but there is nothing for it to flex around. What spine selection governs instead is oscillation consistency.

The arrow exits the bow oscillating — bending and recovering through a standing wave pattern set by its mass and its EI. The frequency of that oscillation is fixed for a given arrow. What is not fixed is the phase at the moment of departure: depending on the timing of the shot, the arrow may leave the bow at a different point in its oscillation cycle from one shot to the next. If the oscillation period is long relative to the shot duration, the exit phase is sensitive to small variations in release timing. If the period is short — a stiffer arrow oscillating faster — the phase is more consistent across shots.

When spine is wrong on compound, this is the failure mode. The arrow is not visibly bent at the target. The paper tear looks acceptable. Groups are simply scattered — no directional pattern, no consistent offset, just larger-than-expected spread with no clear cause. Every arrow left the bow at a slightly different oscillation phase. The arrows disagree with each other, but not in any particular direction. This is why underspine on compound cannot be diagnosed the way underspine on recurve can. The evidence looks like random error rather than systematic error.

The X10 and the engineering of where an arrow bends

Knowing that spine selection determines oscillation behavior, and that oscillation behavior depends on EI — the bending stiffness — along the shaft, a question emerges: what if EI were not the same at every point along the arrow?

The Easton X10 is the most successful design in the history of Olympic outdoor target archery, and one of its distinguishing features is a non-uniform wall thickness. The nock end of the shaft is built thicker — heavier, stiffer — than the point end.

For a uniform shaft, the first bending mode under axial compression produces maximum deflection at the midpoint — a half sine wave symmetric around the center. For a shaft that is stiffer at the rear, the bending stiffness distribution is asymmetric. The mathematics of non-uniform beam columns show that stiffening the rear forces the region of maximum deflection to shift forward — away from the nock end, toward the point end.

For Olympic recurve archery, this is exactly what is wanted. The arrow must flex away from the riser as it departs, and that flex must happen in the forward half of the arrow to clear the rest and riser window cleanly. A stiffer rear section produces this behavior as a structural property of the shaft, not as a tuning artifact. The arrow bends where it is designed to bend, on every shot.

There is a secondary benefit. The nock end — where the string delivers its compressive impulse — is under the highest axial stress during the shot. Thicker walls at the nock provide structural margin precisely where it is needed. The engineering logic is layered: the rear is stiffer because that controls where the flex occurs, and the rear is heavier because that provides safety margin where the load is highest.

The aluminum core further separates the design variables. Carbon fiber is primarily responsible for axial stiffness. The aluminum core contributes to bending stiffness and adds predictable mass per unit length. By combining both materials in a single shaft, Easton can tune the ratio of axial stiffness to bending stiffness in ways that a pure carbon shaft cannot achieve. The result is an arrow that is narrow enough to minimize wind drift, stiff enough in bending to oscillate quickly, and controllable enough in its flex profile that the bending happens where the design calls for it.

The X10 is not the right arrow for most archers. At 3.2mm diameter, it is genuinely difficult to manufacture consistently, and the engineering advantage only translates into scoring advantage for an archer whose form and equipment are already at a level that the arrow's behavior is the limiting factor. For anyone else, the manufacturing risk at this diameter outweighs the design benefit. But as an example of what is possible when shaft design is treated as a structural engineering problem rather than a material selection problem, it has not been surpassed.

The dynamic load ratio

Knowing the static buckling load for any spine, and knowing the approximate compressive force during the shot, produces a number that no spine chart has ever shown an archer: the ratio of applied force to buckling force.

Every arrow in normal use is operating above its static buckling load. A ALR of 1.54 means the applied force is 54% above the level at which the arrow would collapse under a sustained static load. It survives because the load is brief. But the ALR is a direct measure of how dramatically the arrow flexes during the shot — and therefore how large the oscillation amplitude is, and how sensitive the exit phase is to small variations in release timing.

A lower ALR means smaller oscillation amplitude. Smaller oscillation amplitude means the arrow exits in a more consistent phase from shot to shot. Tighter groups. This is the physical quantity that spine selection is actually trying to minimize — and the spine chart never mentions it.

There is one more observation buried in this math. For a 70 lb compound, ALR = 1.0 — the exact boundary between statically stable and dynamically loaded — occurs at a spine of approximately 350 at 28 inches. Standard manufacturer charts recommend 340–400 for a 70 lb compound bow at 28 inches. The empirical chart builders, working from decades of practical observation, unknowingly converged on the stability boundary. Archers shooting the chart baseline are operating right at the edge of where columns stop behaving as elastic beams and start behaving as dynamically unstable structures. It is not accidental. It is where the physics says the transition happens.

What spine sorting is actually doing

Sorted arrows group better. Every competitive archer accepts this as fact. The physical reason is rarely stated clearly: sorting arrows by static spine is a proxy for sorting them by oscillation amplitude, which is what actually governs group consistency.

Within a batch labeled “400 spine,” actual spine values range across some spread — perhaps 390 to 410 for a quality shaft, wider for a commodity shaft. Each spine value produces a specific P_cr. Each P_cr produces a specific ALR. Each ALR determines how large the arrow oscillates on that shot. Arrows with different DLRs exit the bow with different oscillation amplitudes and therefore different tip positions at departure. That spread in tip position at departure is part of what makes the group.

Expressed as buckling force rather than spine units, a “400 spine” batch with a ±10 spine spread (390–410) has a P_cr spread of 25.4 to 26.7 lb — about 1.3 lb. A batch sorted to ±3 spine (397–403) has a P_cr spread of 25.8 to 26.2 lb — about 0.4 lb. Expressed in pounds, the benefit of sorting becomes tangible rather than abstract. You are narrowing the range of compressive forces at which your arrows buckle, which narrows the range of amplitudes at which they oscillate, which narrows the spread of tip positions at departure.

Why GPI belongs in spine selection

The natural frequency of a vibrating beam depends on two things: stiffness and mass. EI is the stiffness. Mass per unit length — what the arrow industry calls GPI, grains per inch — is the mass. Natural frequency is proportional to the square root of their ratio:

In a 15ms shot, a 43 Hz arrow completes approximately 0.65 oscillation cycles before it clears the bow. A 47 Hz arrow completes approximately 0.71 cycles. Those two arrows depart at different phases of their oscillation — the tip is pointing in different directions at the moment the nock leaves the string. They will not group together at distance, even though their spine numbers are identical.

This is the variable that every spine chart ignores. The chart knows your draw weight. It knows your arrow length. It knows your point weight. It does not know your shaft GPI. It assumes all arrows of a given spine are the same mass, which they are not. Two archers with identical setups but different shaft models — same spine, different GPI — will need different spines to achieve the same oscillation behavior.

It also means that sorting arrows by spine alone is incomplete. True oscillation sorting requires matching both EI and GPI. The quantity that governs natural frequency is proportional to spine × shaft weight (for arrows of the same cut length). Match that number across your batch, and you are matching actual oscillation frequency rather than the static proxy for it. Two arrows with identical spine × weight products will oscillate at the same frequency and exit the bow in the same phase, regardless of whether their individual spine numbers are exactly matched.

The practical method requires nothing more than a spine tester and a precision scale. Measure both for each bare shaft. Multiply. Sort by the product. The arrows that cluster together are your matched set.

Oscillation frequency — what it is and why it matters

Natural oscillation frequency is the rate at which the arrow completes one full flex cycle when disturbed. It depends on both stiffness (EI, from spine) and mass per unit length (GPI). Two shafts with identical spine but different GPIs oscillate at different frequencies. In a 15ms shot, a 1 Hz difference between two arrows produces roughly 0.015 cycles of phase offset at departure. A 3 Hz difference produces 0.045 cycles — enough to put the tip in a measurably different position at the moment the nock clears the string. At 60 meters that position difference is group spread.

Direct measurement of arrow oscillation frequency requires equipment most archers do not own. The fundamental for a 400-spine shaft at 29″ falls around 43–48 Hz — deep bass, below the useful range of consumer phone microphones and below most clip-on tuner pickups. Researchers measure it with contact accelerometers and data acquisition hardware. In a production environment it is practical. At home it is not.

What is practical: a spine tester and a scale. Spine × bare shaft weight is a reliable proxy for oscillation frequency within a batch of the same shaft model cut to the same length. Lower product = higher frequency = stiffer oscillation. Sort ascending and pull the tightest cluster.

What the charts do not know

The spine selection charts used by every manufacturer today were developed primarily around recurve and traditional archery, where the physics are more legible. The equivalent draw weight adjustments — add five pounds per inch over 28 inches, add five pounds per 25 grains of front weight — are empirical best-fit values from decades of observing what worked. They were not derived from the oscillation frequency analysis described above. They do not distinguish between a drop-away compound and a recurve with a blade rest. They do not know what the cam profile does to the force curve, or how long the power stroke is, or where the oscillation node lands.

What the charts give you is a starting point that works reliably enough across a wide range of setups to be useful. That is genuinely valuable. What they cannot give you is a confirmed answer for a specific bow, because the physics involved are too dependent on variables the chart does not have.

The confirmed answer comes from the arrow itself. A bare shaft — no vanes, no fletching, just the raw arrow — shot at 20 yards on a well-tuned bow shows you the oscillation behavior directly. On recurve, the bare shaft's direction at the target tells you which way the flex is biased. On compound, the bare shaft group size tells you how consistent the exit phase is. The spine that produces the smallest, most consistent bare shaft group is the spine that matches the oscillation timing of your specific bow. No chart calculates that. The arrow does.