In their 1985 study, "Does the Stock Market Overreact", Bondt and Thaler investigated
whether or not investors tended to overreact to unexpected news. Stocks in the study were
grouped into two portfolios, a winner’s portfolio and a loser’s portfolio, based upon past excess
returns in the 5 years prior to portfolio formation. This was opposed to grouping the stocks
according to some firm specific piece of information, such as earnings announcements. What
Bondt and Thaler found was that the loser portfolio tended to "outperform the market by, on
average, 19.6%, thirty-six months after portfolio formation" (Bondt and Thaler 1985). While the
winners portfolio earned "about 5.0% less than the market" (Bondt and Thaler 1985). The
conclusion that they drew from these findings was that there was indeed what appeared to be
overreaction effect in the market and that this overreaction was asymmetric, that is to say it is
much larger for losers than winners. Bondt and Thaler's study was significant in that while it did
not fully disprove weak-form market efficiency, it did bring to light a significant anomaly.