The forecast ribbon is intentionally simple: a rolling linear regression that stays honest about demo data. Here’s how I keep it explainable.
The regression uses only the visible window of demo data and re-computes whenever the viewer changes the range. It keeps the math approachable and prevents claims about predictive accuracy.
I clamp the projection to business hours so the dashboard never shows overnight gaps that the sample feed can’t verify. If the regression cannot produce a stable slope, the ribbon hides and the copy explains why.
Most importantly, the surrounding paragraphs remind visitors that this is a UX demonstration. When you connect a real feed, you can swap in a more advanced model without changing the UI contract.
Copy these into your own playbooks or PRDs—the bullet points summarize the decisions the article supports.
Let me know if a signal resonates or if you want to sponsor a deeper dive.