Before I start this evening, I wanted to point out something that the Googlebots dragged in.? It’s a list of all the RSS feeds currently created for Aleph Blog.? Note: I only created two of those, the main feed, and the comments feed.? The others were created by readers.? To that end, if you want a specialized Aleph Blog feed, there is probably a way to create it.? Suppose you want all my articles on insurance or personal finance, but not my icky macroeconomics posts.? There is a way to do it.? So if you only want a *part* of the Aleph Blog, set up a customized RSS feed.

One more unrelated trivia question: as of the end of 2011, how many states does Berkshire Hathaway have an insurer domiciled in?? The answer did not surprise me, but it was interesting learning the answer.

-=-==-=-=-=–==-=-=-=-=-=-=-=-=-=–=-==-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

Here’s a comment from a reader on yesterday’s post:

Love the blog, and really appreciate each new submission you create.

I?d love to hear more about ?valuation metrics, sentiment variables, and other factors.? Are these quantitative formulas that I could adopt for myself? I like the idea of ranking your current and prospective holdings. What method do you use to assign a value to each company for grouping?

First you need to read this article of mine.? As for valuation metrics, I use past earnings, forward earnings, book value, Free cash flow, and sales — I give higher weights to free cash flow, book value, and sales.? Earnings are more commonly manipulated.

Other variables: one year price momentum, four-year price momentum, insider buying/selling, yield, neglect ($ volume traded versus market cap), realized stock price volatility, net operating accruals, asset growth and sales growth.

I adjust the weightings period by period.? The changes usually aren’t big, but I adjust the weights to reflect what is out of favor.

For values that are infinite, I input 999.? One more note: I use the ranks of each variable against the stocks owned and contenders, not the actual value to do my rankings.

Here’s a comment from another reader on yesterday’s post:

A bit off topic, David, but something I?ve been thinking about:

Given your comments about incremental improvements, what do you think about the idea that investment manager performance should be judged not by the returns of the portfolio, but by the returns of the new ideas and sold ideas versus the static portfolio? i.e. measure the value of the decisions made recently versus those made in the past?

Given the wide availability of 13-Ds and our ability to copycat others with low-turnover strategies, might this not be a better measure of the value investors are getting for the fees they?re paying?

I do one unusual thing with my stocks.? I list them by order of purchase.? Why? It gives me a feel for whether old or new ideas are working.? But my experience has been that stocks I sell tend to do badly, with some exceptions.? Please don’t talk to me about the sale of NTE, which I sold because the management became more aggressive, and I did not like that.? I really failed there.

But here is the problem: investment managers choose the overall bias that they invest toward.? They should bear some responsibility for their choices, unless they have a fleet of funds that cover almost all relevant areas.

In my own investing, I have not seen any difference over the long haul regarding my new investments versus older investments.? There are times when new or old do better, and times where they are pretty similar.? Since I score my investments quarterly, older investments have to show promise regularly.? Those that show less promise get tossed.? Those with more promise get kept.

The important thing in investing (leaving aside tax implications) is focusing forward.? If something is cheap relative to prospects, don’t sell it.? It is far better to simply understand which of your existing investments are the most marginal in expected returns, so that you can use those as a source of cash when you find promising opportunities.

Full disclosure: long BRK/B