|
Stock Trader and
Economy Student
J.P. Janssen's website.
The purpose of the site is to discuss ideas and principles about markets.
More
|

|
Jan/15/2007
Success Probability vs. Expectancy
Normally you should try to perform at your best in what you do. However, often such a circumstance occurs so that the rational thing is to decrease your expected performance. In almost all endeavors - be it personally, professionally, athletically, academically, socially, romantically, or any other thinkable area - this has to be considered.
Imagine the performance being normal distributed, and binomial - so that it's either failure or success but nothing in between. This may be the case if you are about to go bankrupt, and your future thus relies on a single event. The graph illustrates how taking more risk is a good idea when the odds are against you. The effort should be to maximize the green area minus the red.

Consider all the traders who use stop losses. I suspect this technique being something which saves the individual participants from devastating losses, but at the cost of decreased efficiency of the market. I.e. when a lot of traders sell to save their skins, the expectancy shall be increased.
I looked at every daily change in S&P 500 for 1990-2006. I compared them with the percentage move the next and the next five days. Not too surprisingly are the expectancy greater and the risk higher after big moves, and thus are the desperate sellers who are better off by selling at a discount giving gifts to the ones on the other side of the trade.
|