It's down 17% in a week, but could this awful investing strategy actually work?

"Don't catch a falling knife", "be fearful when others are greedy", "time in the market beats timing the market", "diversification is the only free lunch in investing", "the market can remain irrational longer than you can remain solvent."

So ingrained is the agreed-upon investing wisdom that anyone with even the slightest exposure to markets can likely rattle off a few of these mantras like they can the days of the week.

Another classic that speaks for itself is "don't chase trends". 

Every investor worth their salt knows that the key to a successful portfolio is having a long-term strategy and blocking out the short-term noise. 

Following the hype is a recipe for disaster. By the time you jump on the bandwagon, most of the returns have already been realised and chances are you'll just be someone else's exit liquidity. 

And yet, given how important momentum and sentiment are in shaping markets these days, and how frequently themes develop, maybe there is more juice to be squeezed once a thematic has already left the station. 

There's plenty of recent examples. 

Anyone jumping on the gold trade when it first passed US$4,000 in late 2025 would have looked foolish... until it took off again, peaking at almost US$5,500 a few months later. 

Anyone backing the ASX 20 at the start of year, when every man and his dog was saying the ASX was already too top-heavy and over-concentrated, has fared better than anyone backing the opposite. 

The ASX 20 is up 4.9% year-to-date. The ASX 200 ex-ASX 20 is down just over 5%.

So I decided to put the conventional wisdom - don't chase trends - to the test.

The rules are arbitrary but simple. 

For the rest of the year, my hypothetical $10,000 portfolio will go all-in on a single thematic ASX-listed ETF at the start of each month. 

I'll pick the theme that I think is generating the most hype at that moment in time, based predominantly on recent performance, i.e. which thematic ETF has performed the best over the last month. But I'll also take into account certain macro factors, media coverage and market sentiment. 

At the end of each month, that ETF is then sold and the capital fully reinvested in a different ETF that captures the next trending theme. 

It is the ultimate hype-chasing strategy. By all conventional wisdom, it should be doomed to fail. But will it? 

As way of benchmark, I'm going to be using the S&P/ASX 200 Net Total Return (ASX: NXT). If this portfolio beats out the best the ASX has to offer, I'll consider that a success. 

Month 1 (July) - All-in on AI

What better way to kick off this experiment than with the all-conquering AI trade? 

After progress on peace talks on the war in Iran and the reopening of the Strait of Hormuz, markets have been able to turn their attention back to arguably the most dominant theme of the last decade - AI. 

And the obvious way to jump back on the AI hype train is with FY26's best-performing ASX ETF.

That's right, our first thematic investment will be the Global X Semiconductor ETF (ASX: SEMI), which returned 166% in the 2026 financial year, and returned almost 14% in June alone. 

SEMI tracks the Solactive Global Semiconductor 30 Index, which includes semiconductor companies like Micron, AMD and SK Hynix - some of the best-performing stocks in the world over the last 12 months. 

As the picks and shovels play of the AI cycle, semiconductor companies continue to post record profits and smash earnings forecasts, making this the obvious candidate for those looking to ride the trend right now. 

As Global X's Marc Jocum told my colleague Vishal Teckchandani:

"Semiconductors were the defining investment theme of FY26, as the AI trade broadened beyond the hyperscalers and Magnificent Seven into the hardware supply chains powering the next phase of AI infrastructure."

On 30 June, our hypothetical $10,000 portfolio went all-in on SEMI, purchasing 214.54 units at a price of $46.61.

Unfortunately, we're not off to a great start. 

SEMI has crashed 17% since the start of July, leaving the experiment with a mountain to climb one week in. 

Innovative or insanity?

It's off to a terrible start, but can this strategy turn things around? You'll have to check back in at the start of next month.