What a ballistic coefficient really is

“Ballistic coefficient” is the most quoted and least understood number in shooting. It gets treated as a single quality score — bigger is better, buy the high-BC bullet — when what it actually is, is a scaling factor that maps your bullet onto a standard shape whose drag someone already measured. Understand that, and the whole subject stops being folklore and becomes physics you can reason about. This piece demystifies BC honestly: what it is, what it hides, and what a real measured drag curve buys you that a BC cannot.

Drag is not one number — it changes with speed

The force slowing a bullet is aerodynamic drag, and the crucial fact is that drag does not scale simply with speed — it depends on how fast the bullet is going relative to the speed of sound. That ratio is the Mach number:

$M = \frac{v}{a}$

where $v$ is the bullet’s speed and $a$ is the local speed of sound (about 340 m/s at sea level, and — importantly — it changes with temperature and altitude). A bullet leaving the muzzle at Mach 2.7 and arriving at the target at Mach 1.1 has experienced very different drag physics along the way.

The drag itself is summarized by a dimensionless drag coefficient $C_d$, and $C_d$ is a function of Mach number, written $C_d(M)$. Its shape tells the whole story:

• Well below the speed of sound (subsonic), $C_d$ is relatively low and fairly flat.
• Approaching Mach 1 (transonic), $C_d$ rises steeply and behaves badly — shock waves form and shift around the bullet, and this is the region every model handles worst.
• Well above Mach 1 (supersonic), $C_d$ settles down and slowly declines as Mach climbs.

So “drag” is really this whole curve, $C_d(M)$, not a constant. Any honest trajectory model needs the curve, not a number.

The reference-projectile trick: G1 and G7

Measuring the full $C_d(M)$ curve for every bullet on the market would be enormous work. The industry’s clever shortcut — over a century old — is to measure the curve once, carefully, for a standard reference projectile, and then describe every real bullet by how it compares to that standard.

There are two standard shapes in common use:

• G1 — a flat-based reference projectile with a short, blunt nose. It resembles old round-nosed and flat-based bullets. It is the number most manufacturers still print, mostly for historical reasons.
• G7 — a long, boat-tailed, sharply pointed reference projectile. It resembles modern long-range match bullets far more closely.

Each reference shape comes with its own published, standard $C_d(M)$ table (the BRL / Mayevski drag functions). Which reference you pick matters enormously: a modern VLD bullet described against G1 will have a BC that drifts with speed, because the bullet’s real drag curve and the blunt G1 curve don’t have the same shape — so the single number is only “right” over part of the trajectory. Described against G7, the same bullet’s BC is far more stable, because the shapes match. This is why matching the bullet to the right reference is the first honest step, and why a G7 BC on a modern bullet is worth more than a G1 BC on the same bullet.

What the BC actually does

The ballistic coefficient is the scaling factor that stretches the reference curve onto your bullet. Conceptually it rolls together two things:

• How aerodynamic the bullet is compared to the reference shape (its form factor), and
• How much mass is packed behind that frontal area — its sectional density, the bullet’s weight divided by the square of its diameter. A heavier bullet of the same caliber carries more momentum per unit of drag, so it holds velocity better.

In the trajectory equation, the reference drag curve $C_d(M)$ is scaled by the BC to give the actual retardation your bullet feels. The full equation of motion for the point-mass model is:

$\frac{d\vec{v}}{dt} = -\,k\,\rho\,C_d(M)\,\lvert\vec{v}\rvert\,\vec{v} \;+\; \vec{g}$

Reading it in words: the bullet’s acceleration, $\tfrac{d\vec v}{dt}$, is the sum of two vectors. The first term is drag — it points opposite to the velocity $\vec v$ (that’s the minus sign and the $\vec v$ direction), and its size grows with the air density $\rho$, the drag coefficient $C_d(M)$ at the current Mach number, and the square of the speed (that’s $\lvert\vec v\rvert\,\vec v$ together). The constant $k$ carries the bullet’s ballistic coefficient — a higher BC makes $k$ effectively smaller, so drag bites less. The second term, $\vec g$, is gravity, pulling straight down. That’s the entire core of exterior ballistics: drag backward, gravity down, integrated over the flight.

So a higher BC does not mean “less drag” in some absolute sense — it means the bullet behaves like a more slippery, heavier version of the reference shape, holding velocity longer and drifting less. That is genuinely valuable. But it is a scaling of a borrowed curve, not a measurement of your bullet.

Where a single BC quietly lies

Because a BC pins your bullet to a reference shape, it is only as honest as the shape match. When they don’t match — a modern boat-tail described against blunt G1 — the true drag curve and the scaled reference curve diverge, and the divergence is worst exactly where it hurts: near transonic speeds, at the far end of a long shot, where you most need the trajectory to be right. Manufacturers sometimes paper over this by publishing a banded BC (different values for different speed ranges), which is really an admission that one number was never enough.

What a custom drag curve buys you

The honest answer to a borrowed, mismatched curve is to stop borrowing. A custom drag function is a measured $C_d$-versus-Mach curve for your specific bullet — the bullet’s own drag signature across the whole speed range, rather than a scaled stand-in.

Two things change when you feed the model a real curve instead of a BC:

• You use sectional density, not a BC. The BC bundled aerodynamic shape and sectional density into one number because it had to borrow a shape. Once you supply the bullet’s true drag curve, the shape is no longer borrowed — so the model uses the bullet’s sectional density (weight over diameter squared, a plain physical fact) together with the measured curve. Nothing is being scaled onto a reference anymore.
• The transonic region gets honest. The single biggest failure of a BC — drift where the reference shape stops matching — simply goes away, because you are no longer pretending your bullet is a scaled G1 or G7. The curve is your bullet.

This “true drag” capability is the thing competing tools typically sell as a separate premium module. It is not magic — it is just refusing to borrow a curve when a measured one exists. And it slots straight into the same honesty hierarchy that runs through everything: a measured drag curve beats a fitted BC beats an estimated one. When the model is scaling a reference, it says so; when it has your bullet’s own curve, it uses it.