I have recently started messing around with market breadth data to see if I can manipulate it in a way that I can determine if I should go long or short. Well enough of the gibber jab, let me give you what I came up with and maybe you can do me the favor of letting me know what you think. Add all advances and Add all declines together. Then multiply by 0.75. If advances or declines is higher than that number continue. Add all up volume and Add all down volume together. Then multiply by 0.75. If up or down volume is high than that number continue. Rule = Both of these two forms of market breadth data must be in the same direction. Either advance and up volume or decline and down volume. Now, basically what I'm trying to do here is find what the majority of the market is currently doing and act upon this information in a price valued manner. What I mean by a price valued manner is that when we go long we will look for signs of the market going short and when we go short we will look for signs of the market going long. In this way we will be entering our positions at a value, thus creating more profit. This is a theory of mine and I have done some live backtesting, but I would like to know your opinion of this matter. My professors at college don't really seem to be helping me that much as that I'm kind of on my own for this little project of mine. Maybe you can help me out or share some of your wisdom with me.
check here: http://newsletter.neoticker.com/category/trading-systems-that-work/ oddball and a few others breadth based.
Funny thing is I just sent Lawrence who wrote some of those articles an email right before posting this thread.
My interpretation. You are developing an index - counter trend - strong - exhaustion indicator. If number or issues + total volume is strongly (>75%) up you will take a short position and visa versa. The rationale being that everyone who wants to buy has already bought. It would be helpful if you presented your backtesting results. Why > 75 %? Why number of issues + total volume? If I am interpreting what you are trying to do correctly, this indicator, by itself, has the potential for disaster.
This system is flawed. Without knowing about cancels you can't possibly know if the bids are heavy due to offers being cancelled or due to buyers coming in.
thanks for the input everyone and yes this system is flawed as some of you have already stated. back to the drawing board for me.