Color as evidence
In the maps this desk makes, color is never decoration — it's a claim. That rule is the difference between a research artifact and a pretty picture.
Most data visualization uses color for mood. The palette is chosen to feel calm, or premium, or energetic. It photographs well. It matches the deck. And it says nothing — because if you asked what the blue means versus the orange, there’d be no answer. The color is there to look designed.
For a desk that sells rigor, that preference runs the whole business backwards. If the color of a dot on a trust map is chosen to feel good, the map is a decoration wearing the costume of analysis, and a skeptical technical reader — the exact person these maps are for — smells that instantly.
So the rule here is strict, and it’s simple: if a color doesn’t carry a claim, it doesn’t go in.
The rule
Across every artifact this desk produces, color maps to one thing only — the evidence state of what it’s marking:
- Bright — verified. There’s direct public evidence for this. You could go check it.
- Muted — unverified. Plausible, inferred, or asserted, but not yet backed by the public record.
- Violet / pink — contested. The evidence points both ways, or different audiences read it in opposite directions.
- Grey — noise. Present, but not load-bearing for the argument.
And there’s a legend, every time. The legend is a contract: the promise that the color means what it says, and that you can hold the map to it.
If the color doesn’t carry a claim, it’s decoration — and decoration is where trust goes to die.
Why it matters more here than anywhere
On most sites, color-for-mood is harmless. On a trust desk, it’s self-defeating.
The entire pitch is that these maps are evidence-led — that a bright dot earned its brightness. If the color is actually chosen by feel, then the rigor is a costume, and the first person to notice is the one whose opinion you most need. You can’t sell “we cite the public record” with a chart that colors itself by vibe. The discipline is the credibility.
The same map, twice
The fastest way to see it is to color one map two ways.
Before/after, side by side: the same confusion map colored by mood (a pleasant, meaningless gradient) and colored by evidence state (bright/verified, muted/unverified, violet/contested). Same nodes, same layout. Only the second one can be acted on.
The mood-colored version is prettier at a glance and useless the moment you try to use it — nothing tells you which misreading is confirmed and which is a hunch, so you can’t decide what to fix first. The evidence-colored version is a work order. Fix the bright, contested, high-cost node before the muted, low-cost one. The color did the arguing.
The broader rule: every choice is an argument
Color is just the sharpest case of a wider truth: every choice in a visualization is an argument, and most of them are invisible.
The axis range persuades. Whether the baseline is zeroed persuades. Log scale versus linear can turn the same numbers into two opposite stories — one a smooth climb, one a series of violent booms. What’s omitted persuades most of all, precisely because nobody sees the omission. A chart doesn’t just show data; it makes a case, and it borrows authority from looking neutral while it does.
Our own Bitcoin four-year-cycle piece is the honest example: the same price history on a log axis and a linear axis tells two different stories, and choosing the axis is itself an evidence decision, one we’d have to defend rather than hide.
The standard
Which gives a clean test for any artifact that leaves this desk, and it’s the line to end on:
If you can’t write the legend, you haven’t earned the color.
The moment you can state exactly what each color claims — and stand behind it — you have a research artifact. Until then, however good it looks, you have a decoration. On a trust desk, only one of those is worth publishing.