Bitcoin probability forecasts, explained
What a Bitcoin probability forecast says, what it does not say, and why ranges are more honest than a single price target.
Published by Marcos Mendes · BTC Horizon
A probability is a model statement
A forecast such as “a 30% probability that Bitcoin finishes above a target tomorrow” describes the output of a particular model using information available at publication time. It does not mean the event is guaranteed to occur three times in every ten attempts, and it is not a statement of certainty about the market.
The probability becomes useful only when the same calculation is repeated under fixed rules and compared with what later happened. A model that repeatedly assigns sensible probabilities should be rewarded; one that sounds confident at the wrong moments should be penalised.
Why a range is better than one target
A single price target hides uncertainty. A forecast distribution can instead provide a central estimate together with intervals containing a declared share of the model’s expected outcomes. Wider intervals express greater uncertainty; narrower intervals make a stronger and more easily falsified claim.
Intervals must also be checked after maturity. If an advertised 80% interval contains the realised price only half the time, it is too narrow or otherwise poorly calibrated, regardless of how convincing individual charts looked.
Finishing above is different from touching
The probability that Bitcoin finishes above a target at a particular time is not the same as the probability that it touches that target at any point before then. Touch probabilities require assumptions about the path taken between publication and maturity, not only the final distribution.
How BTC Horizon approaches the problem
BTC Horizon publishes experimental 24-hour, 7-day and 30-day forecasts using CMC BTC/USD inputs and outcomes. Each edition is frozen and tracked through maturity. The current fixed formula has no demonstrated predictive edge; adaptive challengers are tested privately before any replacement.
Methods and evidence
These notes explain BTC Horizon’s implementation and evaluation rules. Worked examples are illustrative unless explicitly described as observed records. Articles are prepared with AI assistance using the project’s code and documented records; they are not independent peer review.