Investment learning starts with knowing the past

I have always had problem with some of the finding behavior finance research. The research is good at pointing out flaws but less effective at offering ways to improve performance and behavior other than stop doing bad things. Nevertheless, I have found some the behavioral research work by Markus Glaser and Martin Weber very helpful with how to offset some biases. They find that there is no correlation between return estimates and realized returns, but they see difference in behavior based on experience. See their work, "Why inexperienced Investors Do Not Learn: They Do Not Know Their Past Portfolio Performance".

Through their survey work, the researchers find that most investors overrate themselves. They are just too confident in their abilities and there is no link between this confidence and actual performance. Some of their other research concludes that this overconfidence causes excessive trading, and more trading is not better for investors. Nevertheless, there is some hope for investors because they find that experiences leads to better estimates of past returns. Investors become more realistic after they gain education and experience. Now, looking closely at the results it may be hard to draw too many conclusions from a limited sample, but the intuition seems to be correct. 

In the school of hard knocks, experience helps us be grounded in reality. Perhaps this is what you should be asking for from an experienced manager, a measure of reality that they may not always be successful, drawdowns will happen, and their estimates of what they will be able to achieve are close to what can be produced. Reality and learning leads to less overconfidence and perhaps better long-term returns.