Survivorship Bias

Survivorship bias is the error of drawing conclusions only from matters or entities that survived a filter, such as funders still operating or cases that reached trial, while ignoring those that dropped out. It makes outcomes look better than they were.

A track record built only from positions that resolved excludes positions still open, which tend to be the slower and weaker ones. A fund that reports returns on closed positions alone overstates its performance.

The correction is to keep the open and abandoned matters in the denominator, to measure by vintage and to show unresolved capital explicitly.

Survivorship bias is not the same as censoring, though the two are related. Censoring is incompleteness in duration data. Survivorship bias is a missing set of matters in the analysis.