Outlier

An outlier is a data point that sits far from the rest — for example one house priced at 50 million dollars in a town where everything else is under a million. It might be a genuine rare case or a mistake, such as a typo that added a zero.

Outliers deserve investigation, not automatic deletion. A real extreme value can be the most important point in your data (a fraud case, a record-breaking event), while a data-entry error should be fixed. Decide based on what the value means; many statistics like the mean are very sensitive to outliers, which is why the median is often preferred for skewed data.