The results strongly supported our hypothesis. We tested for overreaction in various ways, but as long as the period we looked back at to create the portfolios was long enough, say three years, then the Loser portfolio did better than the Winner portfolio. Much better.
