Shares for the future: top-ranked company causes tricky situation

As shares analyst Richard Beddard confesses, his methodology for stock selection is by no means perfect, and neither is he as its operator.   

I have been writing about the Decision Engine for many years, but a few weeks ago I presented it for the first time. It was just about the most terrifying experience of my somewhat privileged life.

Part of my fear was the unfamiliarity of talking to an audience, and I should say thank you to the Twitter followers who advised me on preparing the talk – particularly the wise investor who told me not to try and tell everything.

Part of it was imposter syndrome.

The Decision Engine is a collection of spreadsheets that calculate profitability and financial ratios and collate my judgments on how each business will make more money, what could stop them, and whether we will all benefit. Standing in front of the brightest of the bright, I feared the house of cards might fall down.

Happily, it did not, and here we all are at the first UK Conference of Student Managed Investment Funds, hosted by The University of York:

Photo: Kirkpatrick Photography

As I wrote last month, the theme of the talk was how I use an investing system to reduce bias and improve my decisions.

I explained how scoring shares helps me evaluate the qualities of a good investment independently so I am not, for example, in thrall to the telegenic hero running the company and ignoring the fact that he is always asking us for more money.

And I explained the design of the Decision Engine, which focuses my attention on the company’s score rather than its price, and nudges me to buy or sell shares.

The million-dollar question

Bringing the talk to a close, I had a stab at answering the million-dollar question: How well does it work?

The answer might surprise you, considering I use the Decision Engine to inform my decisions about the Share Sleuth portfolio here on interactive investor and that portfolio has grown sixfold since I started it in 2009.

I think it helps, a bit. There are problems with the financial data.

This is not a criticism of SharePad, my primary source of financial data. My conclusion recognises the fact that accounting numbers are not reality but an imperfect measure of a company’s finances.

There are also problems with permeability.

The Decision Engine is supposed to shield me from the influence of the near-random movement of share prices and associated market chatter, like the apocryphal tin foil hats worn by conspiracy theorists that supposedly protect them from mind control.

It does not stop readers reminding me when one of my favourite shares is down, though. Somehow market noise trickles in through fissures in the foil hat.

• Richard Beddard: look at the right numbers and this stock will impress
• Read more articles by Richard Beddard here

Incidentally, such emails, which are meant kindly, do not deter me from including my email address at the bottom of these articles. They are a small price to pay for all the very helpful emails about businesses and what makes them tick, and the opportunity to clarify and explain when I have unwittingly glossed something over.

The most obvious source of bias is me. The Decision Engine might tell me to buy or sell a share, but I do not always take its advice.

This may or may not be a good idea. As I wrote last month, Daniel Kahneman, a psychologist whose work inspired the Decision Engine, found that intuitive decisions made after the discreet evaluation of individual criteria are often as good as the scoring algorithms that use the same criteria (although he also recommended we use the algorithms regardless).

My somewhat grandiose choice of name for the system gives the impression that it is an efficient decision-making machine, when in fact it is far from efficient. It throws grit into my investing process, slowing me down, forcing me to reflect on my conclusions, and preventing me from rushing to judgement, which helps reduce bias. But it is full of holes through which bias can creep.

Thinking the unthinkable

During the Q&A one of the students asked how I go about deciding which Decision Engine candidates to research, and whether there is an opportunity for bias to creep in there.

It takes quite a while to score a company, so naturally I am selective. With a bit of study, I must believe I will be able to understand how a business makes money.

It must already be profitable, and preferably cash-generative, which shows it can make money.

I favour companies with experienced executives who own lots of shares. And I prefer them to have grown largely under their own steam.

Conversely, I avoid companies that repeatedly raise large amounts of money from investors and lenders to acquire other businesses or develop products that may never make it to market.

I do not even consider gambling and tobacco companies, believing they do mostly harm.

I also avoid businesses like housebuilders and miners, where profits depend substantially on factors out of their control: movements in the prices of the property they own, rather than the industry of their employees.

• Share Sleuth: my first trades of 2022 lead to a new holding
• Richard Beddard: could this firm be a stock market darling again?

The more stable a company – as indicated by past profitability, the experience of management, the source of the growth and the strength of its finances – the more likely it is to have a coherent strategy, and the more likely I am to be able to work it out.

Rules of thumb like these – psychologists call them heuristics – simplify decision-making, but they can also introduce bias (systematic errors) if they are wrong more than they are right.

Potentially, the pre-selection of candidates for the Decision Engine is another hole in my tin foil hat. But given the performance of the portfolios I have made from the shares I have selected, I have some confidence these heuristics work.

One question leads to another, and I confessed to my questioner that I have never thought to address whether the subsequent analysis adds value. Perhaps just by selecting companies I think will score well, I am doing 80% of the work with only 20% of the effort...

Photo: Kirkpatrick Photography

40 shares ranked

Since last month’s update, I have re-evaluated vehicle tracking company Quartix (LSE:QTX), kitchen supplier Howden (LSE:HWDN), Bunzl (LSE:BNZL), a distributor of everything mundane and ordinary, and Tracsis (LSE:TRCS), a gaggle of technology and consultancy firms supplying the transport industry.

The scores of all 40 companies have been influenced by moves in their share prices too.

A brief word of warning is required about top ranked XP Power (LSE:XPP) The company, which I have admired greatly for many years, announced earlier this week that at a trial in the US the jury had ruled against it in a case brought by Swiss rival Comet.

Comet says the verdict proves XP Power stole trade secrets and used them to develop its next generation Radio Frequency (RF) power products, a relatively new part of XP Power’s product portfolio.

XP Power has been fined $40 million, not much less than it earned in the year to December 2021.

It disputes the verdict, and is considering its next steps, which could include an appeal.

The Times reports the current product range is not affected, but the roll-out of new products could be hampered.

This is a tricky situation for The Decision Engine (i.e. me). The cost is perhaps the smallest problem, if it is a one-off.

But the first of XP Power’s core values is integrity. I have always scored the company 2 out of 2 for fairness, which does not seem tenable if it has stolen trade secrets.

I have no expertise in law, so I will not be able to form an independent view on the court case or the likelihood of the verdict being overturned.

When I come to review XP Power in the coming weeks (it has recently published its annual report) I will probably feel compelled to reduce its score at least until there is a final resolution.

• Scores and stats: Richard Beddard. Data: SharePad and annual reports
• Shares marked with an asterisk* score less than 5 out of 6 for Profitability, Risks and Strategy. They are more speculative
• Click on a share's name to see a breakdown of the score (scores may have changed due to movements in share price)

Richard Beddard is a freelance contributor and not a direct employee of interactive investor.

Richard owns shares in many of the companies he scores and ranks. He weights his portfolio so the biggest holdings are in shares with the highest scores.

For more information about Richard’s scoring and ranking system (the Decision Engine) and the Share Sleuth portfolio powered by this research, please read the FAQ.

Contact Richard Beddard by email: richard@beddard.net or on Twitter: @RichardBeddard