Saturday, March 21, 2009

Objective Decision-Making

I've been thinking with my buttons regarding the decision-making process of adjusting a condor. If you've been following this blog, you know I adjusted the OIH Low Prob Condor this week. After I did the adjusting, I kept thinking: now there is risk, but a lot LESS rewards on the trade.

Here is why: Once I cut 1x CALL spread, it took away a lot of my profits, and now I'm trading for an expected pay-off of $75. This is 6% ROI, for a risk of 18%... it is a 3:1 risk-reward ratio and I don't like it. So, what should have been the process here?

Being the Project Manager that I used to be, I went to an old tool I used to apply for decision-making: a decision-making tree. You can find more information at http://en.wikipedia.org/wiki/Decision_Trees, or simply go to a bookstore and grap a project management for dummies book, it should have a simple decision-tree there. If not, get another book :) Here is what a decision-tree would look like for this scenario:


The tree would have told me "do nothing" by a margin of 10 to 1.

Let me explain how the tree works:

1) The first 2 nodes are the decisions you are considering

2) For each decision you make, there are events and probabilities and pay-off associated with them. In the example above, if I close the CAL spread, I'm left with 3 possible events: OIH goes beyond 95, Stays between the strikes, or goes bellow 58. Then let's look at each event and its respective probability and pay-off:

a) There is 15% probability that OIH will go beyond 90, causing a loss of -284

b) There is 82% probability that OIH will stay between the short-stikes, causing a profit of $75

c) There is 3% probability that OIH will go bellow 60, causing a loss of -264

Repeat the process for the "Do Nothing" decision, and update the values, because the probabilities are the same regardless of what you do.

Finally, the tree simply weights the risks/rewards and probabilities and spits out the path that is most likely to give you the best rewards. In this case, I used all probabilities and values based on Friday before expiration.

TOS is great because it calculates the probabilities of pricing expiring beyond these limits and where your profit-loss would be. I'm consdering purchasing the decision-tree add-on for excell to help me on further exploration. What do you think? Have you done something like that for trading? I'm waitting for comments/feedback.