🧠 How to avoid bad ETFs?

73% of investors rely on past performance. Is that enough?

Author

Nessrine Slimani
March 12, 2026

Do you spend more than 6 minutes analysing an investment before buying? Congratulations, you're already doing better than most investors.

A study covered by the Wall Street Journal found that most investors spend only 6 minutes analysing stocks before buying, and 73% of that time is spent looking at price charts and past performance.

6 minutes. Of which 4 are spent looking at numbers that, as we've read a thousand times, don't guarantee future results. So if past performance doesn't predict the future, why is it the first thing you see when searching for an ETF or a stock? And more importantly, is it really useless, or do we just not know how to use it properly?

In this edition, you'll understand why past performance is more useful for avoiding bad ETFs than for finding good ones, and what criteria to look at instead to make a choice without nasty surprises. We're focusing on ETFs here, not stocks—if you want to know why, I cover it in this newsletter.

Does past performance predict the future?

Morningstar is one of the leading platforms for analysing and comparing investment funds. Its 5-star rating is based on risk-adjusted past performance, in other words, how much a fund has returned relative to the risks it took.

The higher the rating, the better the fund performed in the past. In 2017, the Wall Street Journal conducted a survey of 14 years of Morningstar data, covering more than 10,800 funds between 2003 and 2016. Among the funds that had a 5-star rating, only 12% kept that rating five years later.

That's because good performance is often the result of a favourable context that doesn't repeat itself (a mix of luck, sector timing, and more). So when you invest long-term, you can't rely on the last 5 years' performance to predict the next 10 years. So is past performance at least a good indicator of tomorrow's losers? Partly.

Samuel Lee, former strategist at Morningstar, writes it himself: when past performance provides useful information, it's generally to disqualify, because active managers who persistently underperform tend to keep doing so.

This is what Mark Carhart first demonstrated in a landmark 1997 study covering nearly 1,900 US equity funds over more than 30 years.

His conclusion: the only truly robust persistence he observed was that of bad funds, as funds in the bottom decile continued to underperform reliably, while good funds saw their advantage disappear quickly.

And fees play a central role in this story: bad funds often combine poor strategy and high fees, a combination that worsens underperformance and makes it last.

What criteria should you look at to choose an index ETF?

There are three criteria that don't change with market cycles.

1. Fees (TER)

The TER - Total Expense Ratio - is the annual cost of holding an ETF, expressed as a percentage. Unlike performance, it's predictable. ETF providers sometimes adjust their fees up or down, but this is rare.

I dedicated an entire newsletter to the impact of fees on performance, you can find it here to better understand why it's so important.

2. Liquidity

Liquidity is your ability to get your money back when you need it, at the price you expect. This month, BlackRock had to block half the withdrawal requests from its private credit fund, too many investors wanted to exit at the same time, and the fund didn't have the liquidity to cope.

For an ETF, the risk is different but the logic is the same: low daily trading volume can make selling difficult or costly. Make sure the fund has at least €100 million in assets to avoid this risk.

3. Tracking difference

An index ETF's mission is to replicate its benchmark index as closely as possible. The tracking difference measures how well it does this: it's the gap between the ETF's performance and that of its index over a given period.

If your benchmark index gains 10% over a year and your ETF gains 9.3%, the tracking difference is -0.7%.

Two ETFs can have the same fees and yet one tracks its index much better than the other. Between losing -0.7% in performance or -0.15%, you'd obviously choose the second ETF.

Key takeaway

The problem is relying exclusively on past performance to choose an ETF or a stock. No one can predict the future: not analysts, not algorithms, and past performance combines too many factors (luck, timing, sector cycles) to be a reliable GPS to the future.

But it's still useful for disqualifying certain funds.

Combined with stable factors over time like fees, liquidity, or tracking difference, you're less likely to make a mistake. I hope this newsletter helps you see things more clearly.

Take care,
Nessrine

P.S. My first book (🎉) is available for pre-order! Everything we've explored together in the newsletters, structured and deepened. You can reserve it here on Amazon.

Where are you with today's topic?

Login or Subscribe to participate in polls.

Articles, research, and studies consulted for this edition:

Important reminder: This content is for educational purposes only, not investment advice. Make sure to do your own research before getting started. And remember that all investments, ETFs included, carry risks of capital loss.

Reply

or to participate.