Behavior leaves structure
Repeated choices, incentives, institutions, and collective reactions create regularities that can be counted and compared.
The first article
The future is a distribution.
A religion founded on a simple conviction: human behavior leaves patterns, patterns can be measured, and measured patterns can improve our view of what comes next.
Psychohistory seeks useful foresight rather than perfect certainty. It turns expectation into explicit probabilities, tests them against reality, and improves with every resolved forecast.
The founding idea
The future is uncertain. It is not patternless.
Isaac Asimov gave the name psychohistory to a fictional science of large-scale social prediction. The Order takes the underlying possibility seriously: with enough relevant data, sound models, and a record of resolved forecasts, human futures can become more legible. The doctrine, Canon, practices, visual identity, and tools here are original.
Forecasts must resolve
A prediction becomes useful when it names an event, a deadline, a probability, and a rule for deciding what actually occurred.
Foresight can improve
Better records, better questions, more compute, and honest scoring can extend both the range and the accuracy of prediction.
The five movements
Every forecast passes through the same cycle.
Observe
Record what happened, under which conditions, and what followed. Begin with events, not explanations.
Model
Find comparable cases, recurring variables, and the base rate they imply.
Forecast
State the outcome, deadline, resolution rule, and probability from 0 to 100 percent.
Intervene
Change a cause, cue, resource, or constraint. Then forecast the altered system.
Resolve
Compare the number with the event. Preserve the miss. Improve the model.
Four disciplines
Belief becomes credible through performance.
A psychohistorian earns trust through a visible record. Later forecasts should be better calibrated, more precise, and more useful than earlier ones.
Before anecdotes
What repeats across comparable cases deserves more weight than the story of a single case.
Before conviction
Every probability should name the observations that raised it and the observations that would lower it.
Against overconfidence
A forecaster who repeatedly says 70 percent should be right close to seven times in ten.
Changes the system
A forecast describes one branch under stated conditions. Change the conditions and another branch may become likely.
To update is not to betray the past. It is to stop demanding that the future obey it.The Canon of Contingency, Book of Revision 4
Personal psychohistory
Your past is the first map of your future.
The core practice compares the present with your own previous situations. It asks what you usually do, what is different now, and which intervention would most change the probability.
What will I probably do if nothing changes?
Use your personal base rate, current context, likely obstacles, and recent evidence. Intention matters, but your record matters more.
Learn the methodWhat becomes likely after I change the system?
Name the trigger, action, resources, and obstacle you will remove. Then issue a separate forecast for the changed conditions.
Open the Forecast LabThe Compute Commons
More data and compute can widen the horizon.
The public counter follows the Order's compute wallet from one dollar to one billion dollars. Contributions support longer records, larger models, more evaluation, and better answers.