According to the Googles, there have been approximately 14,000 trading days since the Nasdaq Composite was founded, and it has fallen four-percent or more 147 days, meaning that we get a decline like this roughly every 100 trading days, which means about twice a calendar year, very roughly.
That's newsworthy, but also not extremely unusual.
The more interesting detail than nasdaq overall is just what sectors are having bad days and how concentrated that is. Seems to be predominantly AI companies, which is very relevant to the HN crowd.
> meaning that we get a decline like this roughly every 100 trading days, which means about twice a calendar year, very roughly.
Stock volatility is typically clustered. So, we can't make this type of claim. Just like this statement is flawed: Chicago experiences below freezing weather 10% of the time, so every 10 days you see below freezing weather.
It does change the point! If it happened roughly twice a year, it's not as noteworthy when it happens than if tends to happen every other year, but four times.
It's more noteworthy when it happens the more clustered it tends to be. If it were typical that we'd go ~7 years with none then have ~16 in a single year, it would be huge news when it happens.
It's funny: for all the "pattern recognition" people try to apply to markets, they often miss the very basic ones. "Record highs" are incredibly common, and "corrections" also happen regularly.
June had two similar NASDAQ corrections, July had one as well. Sept was flat, but October was a straight downward trend. On the flip side, April to June was a constantly increasing.
The last time the NASDAQ was this low was all the way back in April 2026...
Happens every time Democrats win office, then it slowly trickles back up. The anti-datacenter vote screws up a lot of plans but the lack of power generation had already screwed up much of them anyway. He who shouldn't be named said that all these software people are in for a rude awakening when they try to scale real world infrastructure at the rate they're use to scaling virtual software infrastructure.
There's always market volatility in the time surrounding elections, no matter which party dominates. Once the uncertainty settles down, the markets usually do as well.
The fate of the Bipartisan American Affordability and Jobs Act will be highly instructive. It's a bill that's fabulous for renewable energy, good for affordability, good for fossil fuel interests, and could likely pass with a filibuster-proof and veto-proof majority if the Dems don't sabotage it.
Almost all of the interconnect backlog is renewables, so this helps renewables far more than it helps fossil interests, but some progressives want to sabotage the bill because it could theoretically help fossil fuel interests. Yet, there's no way a "better" bill would get past filibuster or Trump's veto.
We'll find out if Democrats let the perfect be the enemy of the good enough. Whether they're willing to compromise to get things done.
Alot going on today. a 4% drop in a day is pretty darn big.
3 different pods are being liquidated that i've heard of. And the data center trade has been unwound in just 2 days is a pretty violent shock.
Lots of rumors about situational awareness also in alot of trouble. I feel for this guy a bit. He did really well to get a good slug of Anthropic shares that made his investors alot of money.
But since about July when it was clear that Anthropic was done going up like a rocket ship his investors have probably been hounding him about how he's going to make money outside of Anthropic.
his first foray into the public markets was a disaster and might set a record for fastest wipe out from start to finish that forced him to have to sell his public book escape a wind up.
But due to Anthropic taking a possible IPO valuation cut he's had to go back to the public markets at the worst time and if rumours going around today are correct, he's looking at how to liquidate his public portfolio for the second time in the quarter.
Thinking a 5% drop in one day is a big deal is recency bias. Other people have pointed out that statistically it is not that rare. We just have been living through an amazing run for equities.
I am not saying I know it will be fine or I know it will be terrible. I am saying I have lived through many 5%+ drops and they happen on the way up and down.
> And the data center trade has been unwound in just 2 days is a pretty violent shock
Is this true? Coreweave is down ~66% over the last 6 months, with that decline occurring somewhat linearly over that period (i.e. not last 2 days). Very similar chart for Applied Digital.
The Nasdaq rebounds so fast though. it's not that big of a deal. bought the dip. I think it's just noise. markets sometimes look for any reason to sell
Given the election result the AI bailout ain’t coming, combined with growing concern AI spending has gotten too forward on its skis. That’s freaking out the markets.
No new frontier (Astra / Fable) releases in a while. Actually none since the “pacing the frontier” post. Safety maxxers might cause an AI winter if this is what’s holding them back.
Leaks, for what they're worth, indicate that the next GPT model is soon (tomorrow rumored), and that Fable 5.5 is already ready-to-go. That said Fable is likely being held to drown out said GPT release, since Anthropic is already ahead w/ Opus 5.5.
even if frontier AI development stops where it is today, i would have a hard time calling this an AI winter. I cant imagine AI wont be a major part of my dayjob for the forseeable future
> investors accelerated their move away from technology stocks on fears of a data-center buildout slowdown.
So, approximately how many of those investors are going to lose their jobs and their money for betting on data center buildout while also going about the data center buildout in the most obnoxious, exploitative, electorate-alienating way possible?
Intel lost 50% of its value in a week. But it's up 50% from last year.
Anyone paying attention knows the market is in a weird place, and large swings today are the small swings of the past. It's hardly a "plummet" when the amortized line for the past month, including today, is mostly a linear decline.
We've had a couple of single-day 4% drops over the past few months too. Oh no, the market is the lowest it's been in... 3 months!? This is the largest decline we've seen in... 4 months!?
Could it be a market correction? Could it be a full-on collapse? Sure.
But is a single 4% drop worth a headline "plummeting into correction territory"? Not really.
Reporting on the stock market generally is overwrought (stocks always "soar!", "plummet!", "rocket!") because the kind of typical events are inherently repetitive. Sports reporting is the same.
That said, the Nasdaq hasn't seen a lot of -4.8% days recently. This particular article isn't egregious.
That's newsworthy, but also not extremely unusual.
The more interesting detail than nasdaq overall is just what sectors are having bad days and how concentrated that is. Seems to be predominantly AI companies, which is very relevant to the HN crowd.
Stock volatility is typically clustered. So, we can't make this type of claim. Just like this statement is flawed: Chicago experiences below freezing weather 10% of the time, so every 10 days you see below freezing weather.
Although I didn't claim "twice every year like clockwork" in my top-level comment either. It's just a very basic statement of the base rate.
Make it eight-times every four years or sixteen-times every eight years. Or maybe "Four-times every other year, in a clustered fashion".
I don't think any of that really changes the point.
It's more noteworthy when it happens the more clustered it tends to be. If it were typical that we'd go ~7 years with none then have ~16 in a single year, it would be huge news when it happens.
June had two similar NASDAQ corrections, July had one as well. Sept was flat, but October was a straight downward trend. On the flip side, April to June was a constantly increasing.
The last time the NASDAQ was this low was all the way back in April 2026...
See https://www.cnbc.com/2026/11/04/data-center-ballot-measures-...
Almost all of the interconnect backlog is renewables, so this helps renewables far more than it helps fossil interests, but some progressives want to sabotage the bill because it could theoretically help fossil fuel interests. Yet, there's no way a "better" bill would get past filibuster or Trump's veto.
We'll find out if Democrats let the perfect be the enemy of the good enough. Whether they're willing to compromise to get things done.
3 different pods are being liquidated that i've heard of. And the data center trade has been unwound in just 2 days is a pretty violent shock.
Lots of rumors about situational awareness also in alot of trouble. I feel for this guy a bit. He did really well to get a good slug of Anthropic shares that made his investors alot of money.
But since about July when it was clear that Anthropic was done going up like a rocket ship his investors have probably been hounding him about how he's going to make money outside of Anthropic.
his first foray into the public markets was a disaster and might set a record for fastest wipe out from start to finish that forced him to have to sell his public book escape a wind up.
But due to Anthropic taking a possible IPO valuation cut he's had to go back to the public markets at the worst time and if rumours going around today are correct, he's looking at how to liquidate his public portfolio for the second time in the quarter.
His wife is chief of staff to Dario isn't she? Not saying this guy isn't smart but let's call a spade a spade here.
I am not saying I know it will be fine or I know it will be terrible. I am saying I have lived through many 5%+ drops and they happen on the way up and down.
https://www.cnbc.com/2026/09/11/leopold-aschenbrenners-situa...
Is this true? Coreweave is down ~66% over the last 6 months, with that decline occurring somewhat linearly over that period (i.e. not last 2 days). Very similar chart for Applied Digital.
Wednesday - "Market soars because of election"
Thursday - "Market plummets because of election"
Can we at least agree that no one has any idea what is going on?
Am I crazy? Wasn't Astra released just over a month ago? And wasn't Fable 5.1 released around the same time?
How often do we expect frontier models to be released?
Fable 5 was June 9 > Fable 5.1 was September 1 (85 days)
Opus 5 was July 24 > Opus 5.5 was September 22 (61 days)
Yeah, we're not due for a model refresh until the end of month if current release roadmaps of 2-3 months hold.
AI is not going anywhere and will only grow more and more.
So, approximately how many of those investors are going to lose their jobs and their money for betting on data center buildout while also going about the data center buildout in the most obnoxious, exploitative, electorate-alienating way possible?
Still +4% YoY.
24/7 sensationalist news is just exhausting.
Anyone paying attention knows the market is in a weird place, and large swings today are the small swings of the past. It's hardly a "plummet" when the amortized line for the past month, including today, is mostly a linear decline.
We've had a couple of single-day 4% drops over the past few months too. Oh no, the market is the lowest it's been in... 3 months!? This is the largest decline we've seen in... 4 months!?
Could it be a market correction? Could it be a full-on collapse? Sure.
But is a single 4% drop worth a headline "plummeting into correction territory"? Not really.
That term has a very specific meaning in that context.
https://en.wikipedia.org/wiki/Market_correction
[ For the record, the link I originally included: https://www.wsj.com/livecoverage/stock-market-today-dow-sp-5... ]
That said, the Nasdaq hasn't seen a lot of -4.8% days recently. This particular article isn't egregious.
This is a bigger jump, but it's also immediately after elections and other activity so I'd expect higher volatility than normal.