Polymarket price history: reading probability charts

Record market, outcome and price basis. Yes for one monthly threshold and Yes for a neighboring threshold form different series. Even within one event, history must remain tied to its outcome identifier so the comparison concerns the same question throughout.

An illustrative move from 0.28 to 0.42 changes the selected outcome’s price scale. It does not say that BTC itself increased by 0.14 dollars. Comparing the contract with the asset requires a separate series labeled in its own measurement units.

History can use observation intervals with different granularity. Documentation advises reading the observation-window width instead of assuming equal spacing between every point. The final observation also need not fall on a familiar hourly boundary chosen for the display.

Convert observations to one time zone before comparison. Two labels of 12:00 from different zones can denote different instants. Separate the last observation time from page-loading time and the historical window requested for the chart being displayed to the reader.

For illustrative observations of 28% and 42%, the absolute change is 14 percentage points. The relative change is (42 − 28) / 28 = 50%. Both calculations are valid for those numbers, but they answer different questions about the same series.

Show starting and ending observations alongside the change unit. Otherwise, increased by 50% might be misread as a new level of 78%. If the starting observation is missing, do not infer relative movement from a guessed baseline or another contract’s price.

Empty history means no observations were obtained, not a zero probability. A line connecting sparse points does not prove smooth movement between them. Identify the available granularity and avoid inventing intermediate trades or precise intrainterval prices from the connecting segment.

A news story coinciding with a chart change does not establish causation. Explanation requires timing, relevance to the condition and supporting evidence. Without those, describe measured movement only rather than assigning a confident cause that the available series cannot demonstrate.