Updated March 30, 2026
- Fermilab's final muon g-2 measurement achieves 127 ppb precision.
- A revised Standard Model prediction using lattice QCD closes the anomaly gap.
- The twenty-year discrepancy reflected theoretical input errors, not new physics.
For twenty years, the muon's wobble looked like a crack in physics. Fermilab's final muon g-2 measurement confirms the crack was in the mathematics.
The story begins in 2001, at Brookhaven National Laboratory. Physicists measuring the muon's magnetic moment found a number that disagreed with the Standard Model's prediction. The discrepancy was small, a fraction of a part per million, but it was stubbornly persistent. It survived every reanalysis. It survived every skeptical review.
By 2021, Fermilab's Muon g-2 experiment had confirmed Brookhaven's result. By August 2023, the combined data reached 5 sigma, the threshold physicists use for declaring a discovery. Something appeared to be tugging on the muon that no known particle could explain.
What is muon g-2?
The muon is a heavier, unstable cousin of the electron. Quantum theory predicts its magnetic moment should deviate slightly from a value of 2. Measuring this deviation, called g-2, tests whether all known particles and forces are accounted for. Any mismatch between measurement and prediction could signal undiscovered physics.
The Muon g-2 Result: 127 Parts Per Billion
On June 3, 2025, co-spokespersons Peter Winter and Marco Incagli stood before the Fermilab collaboration and announced what six data-collection runs had yielded. The experiment had exceeded its own design specifications.
The result was, by any standard, a remarkable technical achievement.
The collaboration's systematic uncertainty dropped from 280 parts per billion at Brookhaven to 78 ppb. That is more than a threefold improvement. Lawrence Gibbons at Cornell and Simon Corrodi at Argonne, the analysis co-coordinators, oversaw a dataset four times more sensitive than the original Brookhaven measurement.
Key figure
127 ppb
Final measurement precision, beating the 140 ppb design goal
"We not only achieved our goals but exceeded them, which is not very easy for these precision measurements," the collaboration noted. The quiet understatement is characteristic of a field where decades of patient work can hinge on the fourteenth decimal place.
The Theory That Caught Up
The measurement was never really the problem.
The problem was the prediction it was being compared against. Calculating the Standard Model's prediction for muon g-2 requires accounting for hadronic vacuum polarization. That is the contribution from quarks and gluons flickering in and out of existence around the muon.
For decades, theorists estimated this using data from electron-positron collision experiments. Seven such experiments contributed measurements. They disagreed with each other, sometimes sharply.
In 2023, a Fermilab-affiliated analysis identified pion pair data as a source of systematic error in the theoretical inputs. Meanwhile, a different approach was gaining credibility. Lattice QCD computes the strong force directly, simulating quarks on a discrete space-time grid.
Gilberto Colangelo of the University of Bern, a member of the Theory Initiative's steering committee, spent five-hour review sessions scrutinizing the lattice results. Nothing wrong was found. "The fact that they agree on the result is a very good indication that they are doing the right thing," he told Scientific American.
The fact that they agree on the result is a very good indication that they are doing the right thing.
Gilberto Colangelo, University of Bern
In May 2025, the Theory Initiative published its updated prediction using lattice QCD exclusively for the hadronic vacuum polarization term. The new prediction: 116,592,033 (with an uncertainty of 62) in units of 10-11.
Fermilab's measurement: 116,592,071 (with an uncertainty of 15).
The gap is 38 in those units, well within the combined uncertainties. The anomaly, physics' most celebrated hint of new particles, had quietly dissolved.
A Judgment Call, Not a Calculation
The Theory Initiative's decision to abandon data-driven hadronic vacuum polarization was not purely mathematical. It required over one hundred theorists to make a collective judgment about which evidence to trust.
Of the seven electron-positron experiments feeding the old prediction, only one agreed with lattice QCD. Hartmut Wittig of the University of Mainz, another steering committee member, put the stakes plainly. "The answer to whether there is new physics," he observed, "may depend on which theory prediction you compare against."
The theorists chose the method where independent groups converged on the same answer. They set aside the method where experiments contradicted each other. That choice carries genuine weight.
Some physicists argue the data-driven approach is not fundamentally flawed. It may simply be limited by the quality of its experimental inputs. Better collision data could, in principle, restore it. The debate remains unresolved, though the lattice QCD consensus continues to strengthen.
The Standard Model Holds, for Now
Fermilab's storage ring has collected its last muon. The experiment is complete.
And the Standard Model, the framework physicists have spent decades trying to break, remains intact in this sector. Kim Siang Khaw of Shanghai Jiao Tong University, a Fermilab collaborator, offered a characteristically pragmatic view. "Every physics study is a work in progress," he noted.
More On Particle Physics
Inside the Atom: How Molecules Became Microscopic Particle Colliders
MIT physicists turned radium molecules into tabletop nuclear probes, using electrons as messengers from inside the atom's core.
→The question has shifted. It is no longer "what new particle explains the anomaly?" It is now "why did two decades of electron-positron data point in the wrong direction?"
Independent verification may come from J-PARC in Japan. A muon g-2 experiment there uses an entirely different technique and aims to begin collecting data around 2030.
Until then, the most precise measurement of the muon's magnetic wobble tells a quieter story than anyone expected.
The Standard Model was right all along. The mathematics just needed time to catch up.
Sources
- Primary Research: Muon g-2: Most precise measurement of muon magnetic anomaly (Fermilab, 2025)
- Additional Context:
- Sabine Hossenfelder video on muon anomaly (YouTube)
- A Blockbuster Muon Anomaly May Have Just Disappeared (Scientific American)
- Fermilab's final word on muon g-2 (CERN Courier)
- Theory Initiative 2025 update (arXiv)
- Phys. Rev. Lett. 131, 251803 (2023)
Fact Check: Claim-by-Claim Verification Verified
All 13 claims verified against primary sources. Two corrections applied during check: measurement value updated from intermediate Run-2/3 result to final all-runs value; "fourfold" changed to "more than threefold" for accuracy.
Sources used for verification
- Fermilab press release - news.fnal.gov
- Scientific American - scientificamerican.com
- CERN Courier - cerncourier.com
- Theory Initiative update - arxiv.org
- Phys. Rev. Lett. 131, 251803 - journals.aps.org
