Odds of Making Money in the Stock Market

I was surfing through some blogs and came across Value Investor India, a real gem of a blog that is not updated frequently but has some really good posts. It is based on value investing concepts and has some really good posts on arbitrage and industry overviews. Rohit, the author, has obviously gone deep into the internet to dig out some really good stuff, such as this arbitrage returns evaluator, which seems to be based on what I read in Buffetology (or maybe that's just how all arbitrage stock opportunities are evaluated).

Anyways, one post caught my attention, as it seems to be a pretty interesting way to look at timing the market. Basically, the concept is that you should take the current PE and be able to judge the odds of making money in the market based on how often the PE has been higher than this in the past. If today's PE is towards the higher end, then chances are you will lose money in the medium term.

The NSE, and perhaps even the BSE, site allows visitors to download historical index data (open, high, low, close, PE etc) into excel sheets. Now, once you have downloaded the data, it is fairly easy to count the number of times the PE has been higher than the current PE and divide by the total number of days in the historical period considered in order to get an idea of the probability of it being higher than today in future.

Example:
If today's PE is 18 and you see that over the past 1000 days the PE has been >18 50 times and <=18 950 times, then chances of you making money in the future are 5% and chances of losing money are 95%. So you might want to keep your money in the bank for now.

The obvious flaw in this approach is that as markets in India mature and growth slows down, future PEs might be generally lower than past PEs, thereby throwing your calculations out of the window. Further, since you are looking at an index PE, it would be wise to invest in an index fund on this basis. It might even be better than a passive SIP and yield better returns...

Why would this not work for individual stocks? Because individual stock PEs should not be anallyzed statistically as they are completely dependent on management and factors affecting the individual company. For a basket of stocks such as the index, the approach does provide a good rule of thumb.

If you have bought stock funds in a period when the odds were good of PEs rising in future, you'd probably have really done well, benefiting from the increased PE as well as rising earnings, a double benefit!

The example given in the post was that of the stock market crash during the UPA elections when apparently the odds were 10:1 of PEs rising afterwards - and of course the market soared over the next couple of years!

Anyone want to try this out and let us know today's odds?