All Bubbles Explode
- MAC10
- Sep 23, 2025
- 3 min read
All asset bubbles inevitably end badly, so people always ask me - when is inevitable?
Normally September is the worst month of the year, however so far September has been the best month of the year. Therefore logic dictates that bubble implosion has been pushed into the fourth quarter when investors will be least likely to expect it:
Barron's September 23rd, 2025:
"If this stock market were a house, it would look great from the outside but its dimensions would raise architectural concerns."
"Fewer than one in 10 stocks are at 12-month highs."
As the Minneapolis Fed’s Neal Kashkari said last week, it looks like some parts of the market don’t need interest-rate cuts, while others need them badly."
Indeed. Economic cyclicals that need rate cuts are floundering, while big cap Tech which doesn't need rate cuts is sky-rocketing. Similar to the economy, Fed policy is being pulled in two different directions on markets as well - asset inflation in large cap momentum and asset deflation in economic cyclicals.
Here is where it gets interesting:
The formation of this rally is quite similar to 2021. There was a melt-up early in the year 2021 and early this year in January through mid-February - both times, Nasdaq 52 week highs peaked (lower pane). Then came a crash. Followed by a renewed melt-up to new index highs which was driven primarily by large cap momentum stocks and full of breadth divergences. What we notice however is that in 2021, the second peak in new highs (lower) pane took place in Q4. Whereas it appears that the second peak in Nasdaq new highs this year already took place earlier in September. Which means this year's schedule for large cap momentum rollover is earlier than it was in 2021 when it was around the week of November 8th.
Continuing with the Barron's article above:
"It’s a familiar story and repeat of 2023 and 2024, when tech heavyweights pushed the market higher against a backdrop of rocky Federal Reserve policy.
That’s not necessarily a problem, unless a shock to the tech sector causes it to unwind. Valuations also look stretched. But Nvidia and Co.’s stardom mostly masks wider weakness."
It's not a repeat of 2023 and 2024, it's an acceleration of the trend that started two years ago and has gained in size and magnitude throughout this time.
But does it actually take an exogenous "shock" to cause the Tech sector to unwind? Or does it take a stampede into a massively over-owned sector while ignoring broader weakness? After all, there was no "shock" that caused the Dotcom bubble to explode. It merely went vertical and then collapsed in on itself.
Likewise during the 2021 Gamestop melt-up - The market went straight up and then imploded for no particular reason other than the fact that the last fool was found. This of course is the largest melt-up since Gamestop.
We've had two mini crashes already as shown via the S&P semiconductor index below. In the first two crashes each time Nasdaq down volume reached a new record. More recently, on a minor two day pullback at the end of July this year, we see the second largest down volume in history. So one doesn't have to be a genius to predict that this melt-up will see the largest down volume in history by a wide margin.
In summary, bullish pundits are informing us that valuations no longer matter.
Because nothing matters, until it explodes.
