User:Mz7/48-hour hypothesis

The 48-hour hypothesis is the proposition that only the first 48 hours of a request for adminship (RfA) are significant enough to affect the final outcome. This idea was first proposed by Bradv on IRC around 2019–2020.

Background
On the English Wikipedia, request for adminship discussions typically last for seven days before a bureaucrat closes them. During this time, participants can vote "support", "oppose", or "neutral". If an RfA has greater than 75% support at the end of seven days, then the discussion is typically closed as "successful", and if an RfA has less than 65% support, then the discussion is typically closed as "unsuccessful". RfAs with a support percentage between 65% and 75% are said to be in a "discretionary zone", where bureaucrats may determine the final outcome at their discretion, usually following a discussion called a "crat chat". A candidate may withdraw their request for adminship at any time before its conclusion.

Mathematical argument
The more participants that have already voted in an RfA, the less impact that new votes have on the support percentage. Mathematically, the second person who votes at an RfA has the ability to change the support percentage by a maximum of 50% (if there was previously 0 supports and 1 oppose, then a support vote would change the percentage from 0% to 50%). On the other hand, the hundredth voter will have much less impact: they can only move the support percentage by a maximum of 1% (and this would only happen if they opposed when there were 99 supports and 0 opposes, or supported when there were 0 supports and 99 opposes).