After an eventful two years, has my quality screen beaten the market?

Two years ago I created a screen to look for potential quality compounders – those rare high-quality companies that can generate consistently high returns for many years, reinvesting their profits successfully and potentially delivering long runs of market-beating growth.

The intervening period has seen periods of elevated volatility and some unexpected events. In this piece I want to look at the performance of my original selection and consider how this might be improved.

A quality compounding approach to investing requires fairly long timeframes. Two years isn’t really enough to see the benefits of true compounding. But I think it is enough to check which stocks from my original list are still performing and to look for some new contenders to consider.

Compounding Quality screen revisited

Before we go any further, here are the links to my original article and to the compounding quality screen itself.

As a quick refresher, here’s a snapshot of the screening rules I used to try and identify companies with the potential to be quality compounders:

When I originally looked at this screen in May 2023, it generated 35 stocks.

At the time of writing (16 July 2025), there are only 20 stocks in the results.

Of these, 12 remain from the original May ‘23 selection and eight are new.

How did the 2023 vintage perform?

I’ll return to the latest screen results shortly, but first I want to take a look at how my original selection has performed over the last two years or so.

To model the aggregate performance of the whole selection, I created a model portfolio with equally-sized starting positions.

The results are pretty underwhelming. My selection has largely just tracked the FTSE All Share index since May 23, with both achieving a capital gain of around 16%:

The reason for this humdrum performance becomes clear when we look at the range of share price performance from the 35 stocks in the portfolio. While there were some big winners, there were some notable losers as well, including a near-total loss with Argentex:

One selection of stocks over a two-year period isn’t a scientifically rigorous test of an investment method. But this clearly isn’t a very satisfactory result.

What steps might I have taken to improve the performance of this selection?

A key element of investing is avoiding big losses. If we can manage this, then it becomes much easier to achieve a market-beating portfolio performance. It seems clear that what was needed here was a more active approach. This could potentially have taken two basic forms.

Select a subset of stocks

Using the original screen as a starting point, I could have carried out extra research to narrow down my selection to a smaller number of companies with stronger prospects.

This seems to be consistent with a compounding quality approach, which is traditionally a long-term and discretionary strategy. The problem is that this kind of qualitative research is very difficult and often relies on guesswork about the future.

The StockRanks might not have been much help either. Back in May ‘23, there was no clear difference between the future winners and losers:

They all started out with promise, but problems emerged at different times. When that happened, a reactive decision was needed – continue holding for hoped-for long-term returns, or sell and cap losses?

This chimes with Ed’s findings in his multibagger article last week, which highlighted the need to pay close attention to what’s happening with long-term holdings and to be prepared to move on.

There’s some overlap between my selections and the stocks in Ed's article. While my screen didn’t require stocks to be multibaggers, I was certainly hoping to find stocks with similar characteristics that could perhaps become glorious future multibaggers.

Use a rules-based approach

Philosophically, I think a rules-based approach to selling is at odds with the idea of compounding quality. But in this case at least, it could have worked.

One simple solution would have been to use a stop-loss rule to prompt me to sell stocks when they fell more than 25% below their purchase price.

If I had followed this approach with the original selection, the portfolio would have generated a capital gain of 22% over the last two years – a useful 6% above the wider market.

To redeploy the resulting cash, I might also have recycled the cash from sold positions into newer picks from the screen, while continuing to hold the best-performing stocks.

Of course, this approach still isn’t ideal. There’s also the question of what I should do with winners. For example, a stop-loss rule of this kind would allow a stock to double and then fall back to its purchase price without prompting me to take action.

A trailing stop-loss is one possibility, but I find these complicated, prone to misfiring and not reliably available on all platforms.

My preferred approach would probably be to echo the approach I use in the SIF portfolio and continue holding shares if they still qualify for my screen after a minimum holding period – I’d suggest 12 months in this case.

Compounding Quality results: July 25

To wrap up this piece, I want to take a look at the screen results today. Which stocks remain from the original selection – and what are the new compounding quality contenders?

At the time of writing, 20 stocks qualify for this screen (down from 35 in May 23). Here’s the full list from 16 July 2025:

I’ve broken this down into two subsets – original May ‘23 stocks and new entrants.

First of all, here’s how the May ’23 stocks have performed until now:

With only one loser and an average gain of 58% in just over two years; the performance of these stocks has been comfortably above average. This seems to support the idea that it makes sense to cull underperformers and to periodically review whether shares still qualify for the screen.

Looking ahead, are any of the stocks that are new to the screen likely to achieve similar results? It’s impossible to be sure. But I think there are some stocks that might be worth a closer look in the list below, which I’ve ranked by StockRank:

In a follow-up article I’m planning to take a look at a selection of these stocks and consider whether they have the attributes required to become potential quality compounders.

Alongside this, I’m going to update the model portfolio I've created for this piece with a view to revisiting it in another 12 months.

Disclosure: at the time of publication, Roland owned shares in IG Group Holdings and Sage.