A reader left a question that I think is pretty damn good. He asked “what are the variables that I use that allow you to do a forward analysis of an asset with any meaningful probability of being right.”

As someone who believes in top down management I focus on big picture issues and then try to figure the best way to invest toward the big picture while at the same time building in some sort of counter strategy in case my assessment is wrong.

So instead of looking at a stock from the bottom up and saying ok, here is a stock with such and such valuations that will make more widgets next year for a slightly smaller cost and buying it, top down looks at the world for things like supply and demand issues, turning points in cycles or big long-lasting themes as examples. The world is not going to have enough drinking water, the US had an unsustainable real estate price boom built on very low rates or China is going to be much more globally relevant over the next decade.

Another type of analysis might be at the sector level. Take a couple of years ago when the curve inverted. The history of curve inversions is bad. Is there any reason why this might be different? Are the arguments made by the people saying it’s different this time compelling? Based on a bunch of excesses we all knew about, even if we could not quantify, no, those arguments were not compelling.

The best way I could describe the above is that I am trying to put the odds in my favor as often as I can by trying to find the forest. This is far more important than the stocks, or funds, that are ultimately chosen to capture the big themes.

Before the financial sector blow up I used to say, as an example, the best thing would have been to sell all tech in the summer of 2000 as opposed to trying to find the one or two names that might have gone up as the Nasdaq dropped 80%. Now the financial sector is another example. The best thing would have been to sell all financials instead of trying to find one or two that might to do well.

Once you know that things look good for the oil market (using an example from a couple of years ago that shows how simple this can be) what then is the best way to capture this effect taking into account a good balance of potential price appreciation and volatility? Over the last five years it would be very difficult to pick a stock that went down, save for maybe Yukos. Clearly some energy stocks have been relative laggards or leaders but if you got oil right one way or another you probably have a winner on your hands.

For me the energy theme meant looking at foreign stocks because every country has a big oil company, this is an easy sector to add foreign exposure. Having done a little reading I knew that Norway was (and obviously still is) an important destination for oil.

There aren’t too many big oil companies in Norway. If oil up is good for Norway it must be good for Statoil (STO) was the thinking. I could have been wrong of course but the wind was at my back when I bought it. Obviously I still had to learn the company. Norway’s production is declining, the company is very unhedged, they have a big global footprint and I had to learn the numbers but it would have been hard for that stock to have been a horrible pick.

I think this is all very simple, you either see it that way or not but sorting out the big themes and then looking for the best way in is the best description I can give.

Another reader left a comment that included the following;

Diversify. Eat right. Have friends. Work out. Buy Adopt a dog.

Pretty hard to argue with that. If you can’t really eat right how about just cutting out the soda?

Last 5 posts by Roger Nusbaum

• The Big Picture for the Week of November 15, 2009 - November 14th, 2009
• Process Drilldown - October 23rd, 2009
• Sunday Morning Coffee 10-18-09 - October 18th, 2009
• A Little Followup From This Morning - October 8th, 2009
• Wednesday Roundup - October 7th, 2009