Avoiding analysis paralysis by fiddling with my screens again

I'm pondering whether I should continue investigating Latchways. Having decided it's probably a very good company, but reaffirmed my commitment to cheap shares it seems wasteful using up a valuable day on a company that probably isn't cheap enough to add now. Why not wait until such time it is? And focus on cheaper companies.

But there's a problem. At what price would I add it to the portfolio?

One way to establish that is to conduct a full investigation. Put Latchways' financial statistics into a 10 year spreadsheet, generate the charts, check for bogies like craven management, large pension funds or operating leases, virulent new competition and dependence on a small number of customers or suppliers, and then assume a 10% earnings yield and work out the price that would achieve it.

The second way defers the work until its necessary by waiting until the earnings yield drops below, or at least towards 10%. The difficulty with that is my screens aren't set up to do it. They only look at companies that have reported their full year results in the last four months. For eight months of the year companies disappear from them.

That's because I base my decisions on the information in annual reports and bearing in mind the latest annual report refers to events in the previous financial year, it can quickly be superseded by events.

To monitor companies like Latchways all year round I'd have to remove the date restriction from Nifty screen, but I could do that so long as I remembered to check events since the companies last reported were not sufficient to undermine their records of consistent profitability and current financial strength, which is what the screen looks for. Since Nifty companies like Latchways should be models of consistency, the frequency of cataclysmic events undermining the screen should be relatively low.

I'm also dropping the F_Score from the Nifty screen. The F_Score is an indicator of current trading (at least at the company's year end), it differentiates between companies getting into trouble, and those getting out of it. It's incredibly predictive, when applied to companies in distress, but less so applied to consistent performers.

Here's the revamped screen:

It's on probation. I'll update the Thrifty 30 shortlists page as usual at the beginning of next month.

The revised screen retains and ranks the other three variables. The companies at the top have consistently high profitability (ROE) over the last 10 years, low leverage (Assets/Equity), and low market values (relative to book value).

Although the ratios are slightly different, the focus on profitability and value is similar to Greenblatt's Magic Formula with two added pieces of security. It explicitly favours unindebted companies, and those companies that have been profitable over a long period of time. I think that means the results of the screen will include fewer companies that flash brightly for a year or two but are doomed not to shine in future years, either because of financial weakness or unreliable profits.

Only the top 10% of companies meeting the criteria are shown in the table, and Latchways just about squeezes in even though it's the only company with an earnings yield of less than 10%.

It's a difficult company to shake!