I recently had a friend ask me a couple of questions about investing. Here are my answers below.
Stock Screening And Idea Generation
He asked:
How do you come up with your investment ideas? I get many of my investment ideas from reading other people’s research or just general media exposure and extrapolating ideas. The problem I see with this is that I’ll miss the smaller companies that almost have no exposure and likely the better upside.
TLDR: everything you look at is whatever survived your screening process; so this is an insightful question, but our answers will change as we grow as investors.
Generally speaking there is nothing wrong with the ‘random walk’ when coming up with ideas. However, one way to make random exposure a more useful idea generation strategy is to expose yourself to ideas from as many quality sources as possible. I see screening as just another way to gain potentially sensible exposure. In a sense, spending time on a stock that an intelligent, successful, and knowledgeable investor likes is looking at the output of a screen. Of course, you must choose your screens wisely. Some sources will promote stocks based on the most upvotes (reddit) the most clicks (Motley Fool) or which companies are paying them (smallcaps.com.au and similar sites).
For example, the algorithm that promotes stocks in Strawman is a screener, and as it incorporates track record I think it would be a decent one. In comparison, looking at the 10 most talked about stocks on hotcrapper is usually damaging to the mind (if not the wallet).
Some simple screens you can do on Commsec can be powerful. For example, I think I first came across Pro Medicus methodically going through every revenue generating stock below $200m market cap, from A-Z. I still do similar screens today. Another simple screen is to screen for EBITDA growth, as it can point to a company that is around an inflection point.
At some point I’ll get back into the quantitative screening I used to do a lot of. Bots can screen; and I have helped build quite a few screening algorithms. For example, in a prior role I wrote the analysis code for a bot that outputs automated articles based on a company when they meet certain positive flags. You can access the output of that with a simple google search. Part of the reason I’ve stepped back from weekly emails is that I’m putting more effort into the more sophisticated scoring algorithms that can be used to unleash some quite powerful data analytics for screening. I look forward to sharing some of those learnings when I have them.
Another simple screen that has yielded good results for me is a dividend screen. If you look for revenue growth over -5% and then rank by dividend yield you can probably find a fair few stocks that are relatively cheap. Dicker Data is a fine example of a company that initially stood out on this screen (given its consistent growth and large dividend yield).
Ultimately, the investor must take information in, and use it to direct your attention at an opportunity set you are capable of recognising and navigating. This is key, because that should allow you to repeat successful processes, with a view to subtly improving them each time. As a result, there is no limit to your sources of ideas; it could be the output of your twitter screen, or from reading anything from bot generated Simply Wall St articles to the thoughtful thesis of a top notch fund manager. In many ways, I recommend casting the net wide, then using your built up knowledge to quickly sort companies into groups that might interest you.