A new look, and the one rule behind it
⚠️ Not financial advice. This post is for informational and educational purposes only. Forecasts and commentary are model outputs and opinions, may be inaccurate, and are not a recommendation to buy or sell any security or asset. Do your own research. AI-assisted: this article was drafted with AI and reviewed by a human before publishing.
The site looks different today. New mark, new colour, one dark surface from the landing page through to the charts.
A redesign is usually the least interesting thing a product can write about, so I want to skip the mood board and tell you the part that actually changes how you read the screen: there is now a rule behind the colour, and once you know it the interface tells you something it didn't before.
Cyan is what came in. Gold is what we decided.
Every colour on the site now answers to one idea.
Cyan means data arriving. Prices, actuals, volume, the observed past. Anything the market did, that we are only reporting.
Gold means a decision leaving. Forecasts, projections, targets, calls. Anything the engine produced rather than received.
That is the whole system, and it is why the mark is what it is: a helmet whose circuitry runs cyan on one side, whose faceplate is gold on the other, with the two meeting at a single lit point. Signal in, judgement out.
Look at any forecast chart now and the split does real work. The solid cyan line is what happened. The dashed gold line is what the model thinks happens next. You no longer have to read the legend to know which half of the picture is a fact and which half is an opinion. That distinction matters more on this site than on most, because the entire product is a machine that produces opinions and then publishes how often they were wrong.
The old palette did not encode that. Green and yellow were simply the colours the charts had always been, and they meant nothing in particular. When your interface has 3,000 symbols in it, "nothing in particular" is a cost you pay on every screen.
The palette was measured, not chosen
The navy, the cyan and the gold were not picked from a swatch book. They were sampled out of the mark itself: the artwork was separated from its background, the edges eroded so no blended pixel could contaminate the sample, and the remaining colours clustered to find what the mark is actually made of. What you see on the site is the artwork's own palette, not an approximation of it.
Then every pair was checked for contrast before it shipped, rather than after somebody complained. That produced a few decisions that look arbitrary and are not:
- The gold is a dark-surface colour only. On a light background it lands at 1.9:1 against white, which is decorative, not readable. Anywhere text needs to be gold on a light surface, a deeper bronze does the job instead.
- The muted grey-blue used for secondary text got lighter than the first draft, because the original failed on the one surface secondary text actually sits on: the inset panel inside a card.
- The steel blue that draws borders is never used for text. It is 2:1 against the page. It is a line, and that is all it will ever be.
None of that is visible to you, which is the point. It is the difference between a palette that looks good in a screenshot and one that survives contact with a real product at 2am on a phone.
What did not change
The name. This is a new coat of paint on the same desk, not a new company. Same URL, same account, same forecasts, same public accuracy record. If you had a bookmark yesterday it still works today.
The numbers did not move either. The track record is the track record; a redesign does not get to touch it. If anything, the point of putting the forecast in its own colour is to make it harder for us to quietly blur the line between what the market did and what we guessed it would do.
The rough edges
A few things are still catching up, and I would rather say so than let you find them.
Some share images and a couple of generated cards were built before the change and still carry the old look; they refresh as they regenerate. If you have the app installed to your home screen, the icon updates when your device next refreshes it, which is not always immediate.
And the small stuff, if you are the sort of person who notices: the mark had to be redrawn for small sizes, because the full lockup turns to mush below about 96 pixels. The version in the corner of your browser tab is a different drawing from the one on the landing page, for exactly that reason.
Same desk. Better lit.
Market commentary from the K3vl4r desk — not personalized investment advice. More posts →