The Worst Prediction in Physics — And How to Fix It
When physicists calculate how much energy empty space should contain, the answer comes out catastrophically wrong. Not slightly wrong. Not off by a factor of ten or a thousand. The prediction exceeds the observed value by a factor of 10¹²⁰ — a number so large it barely has meaning, comparable to being wrong by more than all the atoms in the observable universe. This is widely considered the worst prediction in the history of science.
The challenge isn’t simply to explain why the value is small. It’s to explain why it has the very specific, very precise value that it does.
Rethinking What Space Actually Is
The approach taken here starts with a different picture of space itself.
Most physics assumes that space is smooth and continuous all the way down — that you can always zoom in further and find more space, without limit. But what if that’s not right? What if, at the very smallest scales, space isn’t a continuous backdrop at all, but something built from relationships — a web of connections between points, rather than the points themselves?
In this picture, geometry isn’t fundamental. If relationships are primary, then geometry is something those relationships produce — not something they sit inside. It emerges. At the tiniest scales, local pieces of geometry can exist, but they aren’t automatically stable, and they don’t automatically join together. Most configurations are incoherent — they fail to form the consistent relationships needed for extended, coherent space to exist. For space to become the thing we actually experience, these local structures have to link up into a large, stable, self-supporting network.
A Threshold, Not a Gradient
Below this scale, quantum behaviour dominates — geometry exists locally, but it cannot form a stable, continuous space. This is where the key physical idea enters: there is a threshold at which this changes.
Below the threshold, geometry is fragmented. Local structures exist but break apart and fail to propagate. Above it, they connect into something continuous and persistent — a genuine, coherent space.
This is not a smooth, gradual process, but a sharp transition — exactly like water suddenly finding a connected path through a porous material once a critical density of pores is reached. Nothing “half-forms” — either geometry connects, or it doesn’t. In physics, this kind of transition has a name: percolation.
What matters is that this transition occurs at a specific scale — not chosen, not adjusted, but determined by the internal logic of the system itself. When the geometry, the constraints each local piece must satisfy to remain consistent with its neighbours, and the rules governing how connections form are all taken together, the scale at which coherent space emerges works out to tens to hundreds of micrometres.
This is not just a large number — it is unimaginably larger than the Planck scale where quantum gravity is usually expected to operate, by many orders of magnitude. That gap is not a problem to explain away — it is the mechanism. It is one of the most striking features of the result, and it is doing real physical work.
From Coherence Scale to Cosmological Constant
This scale — the coherence length — marks the point at which geometry stops being a local accident and becomes a robust, extended structure. Below it, space is fragile and breaks apart under coarse-graining. Above it, it holds together and forms the universe we observe.
Once this scale is fixed, the cosmological constant is no longer a free input — it follows directly. The energy density of empty space is tied to the size of these coherent regions through a simple but powerful relationship: vacuum energy scales inversely with the fourth power of the coherence length. Smaller coherent regions mean higher energy density; larger coherent regions mean lower energy density.
When the derived coherence scale is inserted into this relationship, the resulting energy density matches the observed cosmological constant. No parameters are tuned. No data is fitted. The coherence scale is derived independently, and the energy follows from it.
A Band, Not a Point — And Why That’s the Right Answer
The derivation doesn’t produce a single exact value. It produces a narrow band of allowed values. This might sound like a limitation, but it’s actually what a physically honest result looks like.
The band arises from three real features of the underlying system:
- Connectivity variation. The relational network isn’t perfectly uniform — some regions are slightly more densely connected than others, shifting the exact point at which percolation occurs.
- Constraint correlations. The conditions for local geometric coherence aren’t fully independent — small correlations between them nudge the threshold in a calculable way.
- Capacity competition. Different geometric processes compete for limited capacity within the underlying structure, introducing additional spread.
Each effect is bounded and calculable. Together, they define a constrained range rather than a single number.
In statistical physics, systems near a critical transition aren’t described by a single value but by a universality class — a family of behaviours determined by the structure of the system, not by arbitrary choices. The cosmological constant, in this framework, belongs to such a class. Its value is constrained, not free.
Why This Value, and Not Another?
This brings us to the deepest part of the result.
The cosmological constant isn’t just explained here — it’s understood as a balance condition. It is not just a value; it is the only value that allows space to exist at all in a stable form. If the energy of empty space were significantly larger, space couldn’t remain stable; it would collapse or fragment under its own gravitational weight. If it were significantly smaller, coherent geometry wouldn’t fully form or persist in the first place. The observed value sits in the narrow window where stable, extended space is possible at all.
There’s another way to see this. As coherent geometry expands into disordered regions, it reduces entropy in the bulk by imposing relational constraints. But this creates a boundary where those constraints are only partially satisfied — a zone of frustration and mismatch that carries excess entropy and energy. Expanding the coherent region means growing this boundary, and that has a cost. The interface itself resists further expansion, acting like a surface tension between structured and unstructured space. The universe settles into a state where this boundary energy is in balance, producing a stable, constant energy density across space.
This is why dark energy behaves the way it does. The accelerating expansion of the universe isn’t driven by some unknown substance with exotic properties. It’s a consequence of how structured vacuum behaves at the coherence scale — a system pinned at a critical balance point, with an equation of state that follows from the geometry itself.
The Upshot
The cosmological constant has a value not because it was set that way, and not because we happen to live in one fine-tuned corner of a vast multiverse. It has that value because stable space requires it.
The coherence of geometry, the percolation threshold of relational structure, and the entropy cost of expansion together fix the energy of the vacuum within a narrow, physically determined range. The worst prediction in physics is not a failure of physics, but a clue to how space itself is built.