People tend to predict that they will repeat their previous performance exactly. However, basic principle of statistics tells us that any given extreme performance is likely to regress to the mean over time. A business that is lucky in one year cannot expect to be lucky in just the same way the following year.
Exceptional performances tend to regress to the mean. The worst performance improve and the best — decline from one year to the next. Such instances of regression to the mean can occur whenever there is an element of chance in an outcome.
This is evident in real life situations. Gifted children have less successful siblings, shorter parents have taller children and firms achieve outstanding profits one year but not the next year.
Nevertheless, individuals assume that future outcomes will be directly predictable from past outcomes (e.g. sales). We tend to naively develop predictions based on our assumptions of perfect correlation with past data.
Under some unusual circumstances, we expect performance to regress. For instance: if a real-estate agent sells five houses in one month (abnormally high performance), his fellow agent does not expect equally high sales the next month. Thus, in some unusual situations individuals expect regression to the mean effect, but most of the times they miss it in less extreme cases.
Managers who fail to recognise the tendency of events to regress to the mean are likely to develop false assumptions about future results. Consequently, they make inappropriate plans and have unrealistic expectations for employee performance.
Bazerman, M.H. and Moore, D.A., 1994. Judgment in managerial decision making (p. 226). New York: Wiley.