
Confidence is not evidence about odds
How sure a person feels at the start is an accurate reading of that person and nothing else. Everybody standing in the same line was let in by the same feeling, so the group's rate does not move when one
A formally documented quantitative projection, established before the start of an activity or the collection of data, forms a pre-registered estimate. This kind of estimate defines an explicit baseline against which actual future results are compared, ensuring that initial expectations are transparently recorded and cannot be retroactively adjusted by the holder of the seat. Its scope extends to any quantifiable aspect of a project, from expected resource consumption to anticipated performance metrics, and it ceases to apply once the actual, observed values are definitively known and recorded in the permanent record.
The mechanism for a pre-registered estimate involves committing to specific figures or ranges at a predefined point, often through a formal submission or public statement that is difficult to retract, making the initial projection binding for future assessment.
A founder’s decision to publish a pre-registered estimate transforms an internal expectation into an external datum that becomes part of the public record. This declaration creates a fixed reference point, making the founder accountable to themselves and any partners for the stated projection and its subsequent variance. The estimate provides a measure of anticipated performance or resource need before the work commences, directly influencing the allocation of attention and capital.
It stands as an unchangeable record, against which the efficiency of the operator’s execution can later be evaluated, independent of the outcome’s favorability, thereby shaping the perception of the operator’s foresight and planning ability and influencing future exchanges.
This recorded projection provides a crucial witness to the initial state of knowledge and expectation. By creating an artifact that cannot be altered once released, a pre-registered estimate safeguards against the cognitive bias of adjusting one’s recollection of prior belief to align with a known outcome. This record allows for an objective assessment of prediction accuracy, offering valuable feedback on the predictive models or assumptions used in the work.
Its value lies in establishing a transparent history of expectation, enabling an audience to validate the founder’s forecasting ability over a sequence of engagements, fostering confidence where claims are consistently met or systematically improved through demonstrated learning. The witness function holds even when the estimate itself is imperfect, as it records the process of prediction, not just the result.
The practice of establishing a pre-registered estimate carries a specific operational cost for the founder. The energy required to conduct thorough preliminary research, model potential outcomes, and defend the chosen figure consumes significant hours prior to actual execution, directly impacting the capacity of the operator for other work. This upfront investment in analysis and documentation reduces immediate productive capacity, representing a direct trade-off in the present moment, as resources are diverted from active building to forecasting.
Should the actual results deviate substantially from the pre-registered estimate, the founder incurs additional costs, not only in explaining the variance to partners or clients but also in dedicating future attention to recalibrating predictive models or adjusting subsequent plans. This consequence can diminish the perceived reliability of their predictive process, potentially affecting the terms of future exchanges or the ability to secure new provisions. The load is constant across different scales of operation, from a simple task to a complex project, compelling a founder to balance the rigour of prediction against the immediate demands of execution.
The ongoing cost can also appear in the form of deferred opportunities, where a founder might hesitate to adapt a plan mid-course if it means diverging from a publicly committed estimate.

How sure a person feels at the start is an accurate reading of that person and nothing else. Everybody standing in the same line was let in by the same feeling, so the group's rate does not move when one
The terms are written where they belong. Each field holds three desks, and every entry a desk writes raises the terms it uses, each term given a meaning, a mechanism and the places it appears. The record on these pages stays first person and hand made; the fields grow the nomenclature.