When Markets Stop Closing, Do Prices Still Clear?
Why continuous trading may delay price discovery—even as it accelerates trading
For a long time, markets had a built-in advantage for figuring out price:
They forced everyone to look at the same thing at the same time.
The open. The close. Earnings prints.
Those weren’t just operational artifacts. They were moments of coordinated attention density—when liquidity, capital, and opinions all showed up together.
Nasdaq moving toward near-continuous trading chips away at that.
The obvious takeaway is:
More trading → faster price discovery.
I’m not sure that’s right.
A “market clearing price” isn’t just where supply meets demand.
It’s where attention converges.
Historically, that convergence was:
synchronized
liquid
visible
Now it’s increasingly:
staggered
uneven
dependent on who happens to be awake and active
Instead of one clearing event, you get a sequence of smaller ones.
Prices don’t converge all at once.
They accumulate consensus.
A completely hypothetical example.
Say a company releases material news at 2:00 AM ET.
Overnight: thin market, mostly models and global funds → stock trades up 3%
Europe opens: agrees with the move → now +4%
US pre-market: some pushback → +2.5%
9:30 AM: real liquidity shows up, larger holders reposition → closes the day +1%
So what was the “right” price?
At each step, the market cleared—but only for the participants who were there.
The final price didn’t emerge until attention and liquidity actually concentrated.
It’s tempting to say: software will solve this.
And to a point, it does.
Models can read:
order flow
momentum
short-term imbalances
They can make very good predictions over short horizons.
But they’re predicting based on who is currently in the market.
They can’t see:
who is waiting for liquidity
who only trades during US hours
who will show up with real size
You can model where the market is going.
You can’t model who hasn’t arrived yet.
So you end up with a two-stage system:
Exploration
happens in thinner markets
fast, continuous, model-driven
prices move quickly, but not always with conviction
Validation
happens when liquidity concentrates
slower, more deliberate
prices actually “stick”
The clock starts earlier.
The finish line doesn’t move much.
This has a subtle implication.
We tend to think more access, more data, more trading = more efficiency.
But what actually shifts is the bottleneck.
Not information.
Participation.
Who trades—and when they choose to—starts to matter more than the data itself.
If that’s right, then not all prices carry the same weight.
A move in thin, continuous markets may look like information—but it may just be participation.
And trading on that distinction is the difference between reacting to noise and committing capital with conviction.
None of this is entirely new.
Markets have always depended on attention.
What’s changing is how that attention is distributed.
And that’s enough to make something we could mostly ignore before start to matter.
Continuous markets promise constant price discovery.
What they may deliver instead is constant price formation—without resolution.