Five common stock screening errors and how to avoid them

The Stockopedia Stock Screens are an invaluable resource for the intelligent investor. They enable investors to search for stocks that meet their own unique investment criteria without relying on others for ideas. Screens are particularly beneficial to me as a value investor since I am always searching for the most unloved investment ideas. These types of stock rarely appear among the most talked about on discussion boards, at least when I start investing in them. However, by utilising the Stockopedia screening tools, I can identify these potentially lucrative investments early and get in ahead of the herd.

Since writing this value investing series, the stock screens I have created have led me to find several new investments that I wouldn't have considered otherwise. For example, my 52-week low screen highlighted several companies that made it onto my watchlist. After further research and patiently waiting for a good entry price, two stocks entered my portfolio: Angling Direct (LON:ANG) and National World (LON:NWOR) . Since purchase, these have risen 15% and 35% respectively and still look excellent value. Without this stock screen, I wouldn't have made these gains. Nor would I be up by around 35% on my average Luceco (LON:LUCE) buy price if this company hadn't stood out amongst my analysis of potential Buffett and Munger Stocks.

While a well-constructed Stock Screen can uncover many hidden gems like these, some common pitfalls can plague the screen creator:

Error #1 - Generating too many or too few results

Some value screens look for very specific conditions. For example, a screen intended to identify Ben Graham Net Nets may only have a few results unless we are in the depths of a severe bear market. (See the following talk I delivered back in 2020 for more information about this type of stock).

However, the vast majority of screens are not like this. Instead, investors use them either to create a list of stocks where they intend to buy all of them, a so-called quant strategy, or to narrow down the market to some of the most promising investing ideas. In both cases, there is an optimum number of stocks. For a quant strategy, this is a balance between investing in enough stocks to own a diversified portfolio and the costs of owning many stocks. When narrowing down the investing field, the investor wants to find enough interesting ideas without being overwhelmed by the quantity of further work. In both cases, I think between 30 and 50 stocks is ideal.

When I created a Buffett and Munger Screen, these were the criteria I chose:

While I wanted stocks at the upper end of the range of the Quality Rank and Gross Profit margin, the 70th percentile as the cut-off was partially chosen to give sufficient results. The screen at the time generated 49 results, right in my sweet spot. It may seem like a good idea to go for the 90th percentile and an earnings yield of 10% or more to find the absolute best companies. However, this only generates four results. A cursory look at these four companies suggests that none would be Buffett and Munger stocks once I utilised my Moat, Runway, Operational Leverage, and Low Multiple framework to assess them further.

Error #2 – Failing to base a screen on academic evidence or logic

While it is certainly possible to create a screen that screens for companies with the most broker Hold recommendations or the largest EPS per share, there is little reason to do so. The idea of screening is to narrow the field in finding market-beating investments. A screen usually starts with one big idea, which should ideally be based on academic evidence of outperformance, such as a research paper. This may be a value factor, such as high Earnings Yield, Momentum measures, or various quality metrics, such as Gross Profits/Assets. It may also seek to systemise the winning ways of a well-known investor, as many of the Stockopedia Guru Screens do.

Alternatively, it can be worthwhile to screen based on a logical idea. For example, screening for stocks that have failed to bounce during a recent bull market may find some great ideas. The theory being that cheap stocks passed over by the market may be due their day in the sun soon. However, unless the investor can explain the evidence or logic of why a screen is likely to find outperforming stocks, their screen is more likely to lead them astray rather than generate market-beating returns.

Error #3 – Not considering the best implementation of a screening idea

Examples of this error include not being clear if the screen is intended as a quant strategy, and therefore the intention is to purchase all of the screen results, or whether the screen is the first step in narrowing down the field. For me, a screen is always the start of the process, and the criteria I choose reflect this.

Another example of this error is failing to consider the time period with which to set criteria. For example, when I created my 100 Baggers screen, I used five-year averages for ROE and Gross Margin:

In using these criteria, I understood that the poor 2020 and exceptional 2021 many companies experienced due to COVID might distort their gross margin. If I only looked at recent history, I could be fooled into thinking a company was of higher quality than it really was. I also recognised that Return on Equity is one of the most mean-reverting series in finance and that if I wanted to identify companies with a moat, I would need to look at long-term measures.

Error #4 - Not taking account of industry-specific variation

When it comes to financial metrics, what is considered an acceptable level can vary by industry. So, for example, when I created my Price to sales Screen, I chose to look at the Current Ratio in Industry Group rather than over the whole market:

This is because I recognised that many low Price-to-Sales stocks would be retailers or similar companies that usually have negative working capital. If I chose a specific current ratio, it would exclude many such companies for which a lower-than-market-average current ratio is not a sign of distress. However, I still wanted to ensure some level of relative balance sheet strength. The answer was to use the rank of the “Current ratio in Industry Group” to exclude the worst-financed stocks in each industry.

Similarly, when I created my income screens, I found that I could create a more diversified and safer income portfolio by choosing the highest-yielding stocks within an industry group rather than the highest in the market. This was because the highest-yielding stocks at the time were dominated by property and financial stocks that all faced the same housing market risks. As a result, the "within industry" screen generated a much safer portfolio and only sacrificed a minor amount of dividend yield.

Error #5 - Investing without checking the data

Although every effort is made to correct data errors when reported, Stockoedia relies on third-party data providers, which are not infallible. This is particularly true regarding brokers' forecasts. For example, in 2020, many brokers reacted to the uncertainty caused by COVID by removing their forecasts from the market. However, the old estimates were not always removed from the data set. So while a stock screen will almost always find companies that meet an investor's chosen criteria, there are a few exceptions when the data is wrong.

Investors should always check the financial figures with the Annual or Interim Report from the company. Where possible, brokers' forecasts should also be checked. If brokers' notes are unavailable, then an investor may be wise to create their own model of the company's financials to confirm key metrics. Finally, before investing, an investor may want to create their own Stock Pitch article on Stockopedia. An investor will almost certainly learn something new about the company by writing one. I certainly did when I wrote about Angling Direct (LON:ANG) last week. The Stockopedia community will also act as a sense check for the idea. While the value investor should be wary of being overly led by the opinion of others, the Stockopedia community is knowledgeable. It will often point out when obvious mistakes are made.

By avoiding these common errors, investors can easily build high-quality screens to find great investment ideas. Happy screening!

[Disclosure: Mark owns shares in Angling Direct (LON:ANG) Luceco (LON:LUCE) National World (LON:NWOR) ]