The stock that got away: The ASX play Jun Bei Liu was happy to watch from the sidelines

Even the world's greatest investors have their fair share of misses, but what's important is what you learn as a result.

But it takes a brave investor to share those stories, and an even braver one to let it be published for a wider audience and help other investors improve their own decision-making.

In this series, we speak to some of Australia's leading fund managers on an investment opportunity that slipped through their fingers and the lessons they took from it.

Here, Ten Cap's Jun Bei Liu talks us through an unloved stock she exited too soon, but why she's sticking by her decision despite the subsequent price performance. 

What was the investment idea?

Every investor is asked about the one that got away, and the assumption baked into the question is that it's a story of regret — a stock you should have held, a profit you left on the table. 

Mine is a little different. Yes, Sims Limited (ASX: SGM) has run well past where I sold it. But the longer I look at where it sits today, the more comfortable I am that getting off the train when I did was the right call — even if the train kept going for a few more stops.

Let me set the scene, because the entry matters. On paper, Sims is exactly the kind of business a lot of fund managers won't touch. It's a metal recycler — scrap prices are notoriously volatile, earnings can swing from healthy profit to outright loss in the space of a couple of halves, and the disclosure has never made it easy to see what's really going on underneath. 

Historically it screens as a classic low-quality, deeply cyclical name, and plenty of people in our universe simply put it in the "too hard" basket.

What was your reasoning at the time?

Our investment process is focused on identifying money making opportunities regardless of the inherent (or perceived) quality of the company even if that involves businesses that others have written off, because that's often where a mispricing hides. 

Late last year something had changed. The lifecycle services side of Sims, the part that recovers and resells IT hardware, particularly memory chips, was quietly becoming a far more meaningful earnings driver, and we were reasonably confident on near-term scrap prices. 

Put those together and you had a genuine opportunity, even in a business we'd never call high quality. So we took a tactical position. We bought in November 2025 around $16, and we sold in February at $22. A tidy gain in a few months on a name half the market wouldn't go near.

Why did you exit? 

We sold because memory pricing looked toppy, and this is the part worth dwelling on, because it's the heart of why I'm wary today. 

Memory operates in a boom-bust cycles because its cyclical. Pricing surges when demand outruns supply — a new product cycle, a wave of data-centre build-out — and the chipmakers, seeing fat margins, all rush to add capacity at once. That capacity arrives in a lump, supply overshoots, and pricing can fall just as violently as it rose. 

Memory margins have whipsawed between negative and positive for decades; this is not a market that gently mean-reverts. The material gains you see at the top of a cycle are rarely sustained. 

So when DDR4 pricing looked stretched, we did what our discipline tells us to do: we banked the profit and stepped aside. Investing isn't about selling at the very top, it's about taking a good return for a defined risk and not overstaying a thesis you don't fully control.

What happened next? How did it perform?

Sims has rallied around 75% over the past year against a market up roughly 5%, and almost all of that re-rating rests on one thing: the surge in lifecycle earnings, which on consensus numbers has jumped to roughly half of group profit, up from under a fifth a year ago. 

It looks like the market is treating those earnings as a permanent higher base – the hype around AI an obvious driver. I'm not convinced it should, and, tellingly, nor are the sell-side analysts who cover it most closely.

What's your thinking now? 

Here's what makes me cautious. That earnings step-up is overwhelmingly a memory-pricing story, the resale of recovered memory is the largest and most profitable part of the mix, and the great majority of it is directly linked to where memory prices sit. 

The profit is also heavily skewed to the second half of the year, which tells you how sensitive the result is to spot pricing right now rather than to anything durable. The analyst community is uneasy about this. 

Even after big target-price upgrades, several remain reluctant to capitalise today's memory-driven margins into perpetuity, they point to a lack of clarity around the unit economics, to DDR4 pricing that has actually been subdued for much of this year, and to the simple fact that the kind of spike Sims is enjoying is rarely sustained through a full memory cycle – at least in the past.

That's the asymmetry I see at today's price – will the stock remain cyclical and mean reverting or has it entered a new regime where this earnings driver is structural. To make money from here, you have to back memory pricing staying elevated for longer than the cycle has historically allowed, and you have to believe the market is right to capitalise peak earnings as permanent. 

The upside is a continuation of a cycle; the downside is the cycle doing what it always eventually does. I took a good return on a well-defined thesis, and the same discipline that got me in is what's keeping me out now. 

Sometimes the one that got away is just a reminder that you don't have to be on board for every leg of the journey, especially the part where the risk is highest.

Managed Fund

Ten Cap Alpha Plus Fund

Australian Shares

View