How to create rules for picking winning shares like Rolls-Royce before they take off: ED CROFT

Daily Mail​

18-07-2025

If you open an account using links which have an asterisk, This is Money will earn an affiliate commission. We do not allow this to affect our editorial independence.
In the second part of our series on how to pick shares, Ed Croft, the founder and chief executive of Stockopedia, explains how to establish rules that will help you pick winning companies and avoid those that turn out to be duds.
He runs through why he considers certain elements important and how the site built its StockRanks system that spotted opportunities such as Rolls-Royce, Jet2 and Games Workshop shares before they took off.
Flying high: Rolls-Royce shares have soared over recent years after their pandemic slump and Ed Croft says a rules-based stock-picking approach could have spotted the opportunity
In the first article of our share picking series, we tackled the problems many investors face – story-driven speculation, tip chasing, and the trap of seeking more and more information, which often brings overconfidence rather than better results.
One answer is using a rules-based approach, based around the characteristics of shares proven to identify winners.
When you know that there's a persistent pay-off to buying the highest quality, value and momentum shares, there can be a real mindset shift.
But for many, that's where the journey stalls.
Because once you realise that rules matter, the next step is to create your first set.
Price to Earnings Ratio of less than 12? Tick. Return on capital above 15 per cent? Tick. Debt under control? Tick. You build a rational set of logic, and it feels good.
Until you hit a wall.
Rules are essential - but they can be restrictive
I still remember the buzz of creating my first stock screen. Stock screens are essentially checklists of rules that can narrow a universe of thousands of stocks down to a manageable list.
My first was based on Jim Slater's criteria from his excellent Zulu Principle book. I had a whole list of 'must have' criteria which would find me reasonably priced, quality growth shares. But how many candidates did the screen produce?
Just three. Two were rather small and illiquid and the third was some obscure foreign-listed firm. It was quite disappointing. And it certainly wasn't an investable portfolio.
Checklists can be powerful – don't underestimate them. They add discipline to your investing and help filter out the noise. But they can be extremely restrictive.
If a stock has a P/E of 12.4 but you've screened for less than 12, should it really be cut out? Of course not. But a strict set of rules won't catch it.
So what do you do?
You start raising all your cutoffs – you find a broader set – but something feels amiss. You know your cutoffs are keeping out some of the best candidates in the market.
Scoring every stock in the market
The breakthrough for me came when I stopped thinking so binary – in pass/fail terms – and came across the idea of scoring. It was Joel Greenblatt, in his excellent 'The Little Book that Beats the Market', that sowed the seed.
What if, instead of demanding that a company have a P/E of less than 12, you scored every stock in the market for how low the price to earnings ratio was? And another for how profitable it was – using its 'return on capital'.
Rather than just having a hard cutoff for 'cheap' or 'good' shares, you could create a gradient – with 'cheap, highly profitable' stocks at one end, and 'expensive, unprofitable' shares at the other.
It's a fundamentally different approach. You're no longer left with just a few stocks, you have a score for every stock in the market. And what's even better, you can then compare any stock against any other.
Stockopedia
Stockopedia is an analysis site for share-picking investors that This is Money's team has used for many years.
Its share data and StockRanks provide exceptional insight.
There are also practical tips, Model Portfolios, and simple, consistent rules-based strategies.
As a special offer, This is Money readers can get 25 per cent off a Stockopedia membership *
How will you score stocks?
You can do this even if you are a stock picker, looking at stocks on a case by case basis. You can build a set of solid rules, which can even include qualitative assessments like 'how experienced and trustworthy is the management', and grade stocks between zero and ten for each rule.
I know some of the best investors in the UK that do this. It does require judgement, but it removes a lot of bias from your investing process, and helps you avoid getting too sucked into a story.
But a more data-first approach allows you to compare all the stocks in the market.
If you like doing your own work, you can do this yourself – you will need access to a financial database to export some data. Some are free on the web, though you do have to be careful with data quality. But it's really worth it.
Choose some key metrics across a few of the quality, value and momentum dimensions.
For example, as I've described, you might choose the P/E ratio and the return on capital (a key profitability measure). Score each one as a percentage from zero to 100, where 100 is 'best'.
Total the scores and rank again. You can then use this score to check your own stock ideas against. It's not hard. And it works.
> Check out the illustration in this spreadsheet
It uses institutional quality data from Stockopedia's database (correct at the time of publishing).
Take a look and see if you can find any of your shares within it – their scores might surprise you
While this is just a simple example, the benefit of this kind of gradient-based thinking is huge. It can add so much rigour to your process, but more than this – it really gets results.
Putting it all together to catch share price moves early
Scoring the market for a couple of financial ratios isn't really that robust.
Just scoring for 'value' based on the P/E ratio can leave opportunities on the table – what about companies that are cheap relative to their company sales, but are just turning profitable? When their sales grow, they can see huge profit expansion.
So at Stockopedia, we built a system that takes this further. Every stock gets a daily score – out of 100 – based on its quality, value, and momentum profile – but each of these scores is built from a range of financial ratios to give a more robust and rounded guide to their relative merit.
We call these measures the StockRanks and they have a terrific track record of finding stocks that perform.
Top ranked stocks through time have included many of the very best multibaggers in the UK market – stocks like Rolls Royce, Games Workshop, Jet2 and more – before they took off.
The top 10 per cent of ranked shares – those in the 90+ range – have on average returned 11.9 per cent annualised. And the lowest ranked 10 per cent? Well, more than 75 per cent of them end up losers – with an average annualised return of -17 per cent.
Now, this doesn't mean every 90+ ranked stock is going to be a winner. Of course not. A good score is not a buy recommendation. Individual stocks do their own thing, profit warnings happen and markets go down as well as up. But across groups of stocks, over the longer term, the odds weigh in your favour. That's what matters.
Building a portfolio that captures the 90+ effect
In the third article of this series, we'll explore how to move from understanding the pitfalls of emotional investing to applying a rules-based approach that works.
Avoiding behavioural mistakes such as poor stock selection, under-diversification, and reactive trading, we'll show how to construct and manage a rules-based portfolio that has a high chance of beating the market.