In particle physics, data long outlives the detectors that generate it. A decade ago the 4,100-metric-ton Collider Detector at Fermilab (CDF) reached the end of its life and was shut down, stripped of its parts for use in other experiments. Now a fresh analysis of old CDF data has unearthed a stunning discrepancy in the mass of an elementary particle, the W boson, that could point the way to new, as yet undiscovered particles and interactions.

The W boson is massive, some 80 times heavier than a proton. Crucially, the W boson is responsible for certain forms of radioactive decay, allowing neutrons to convert into protons. Because its mass is constrained by (and itself constrains) many other particles and parameters within the Standard Model—particle physicists’ theory of fundamental particles and how they behave—the W boson has become a target for researchers seeking to understand where and how their best theories fail.

Although physicists have long known the W boson’s approximate mass, they still do not know it exactly. Plugging data into the Standard Model framework, however, predicts that the so-called W mass should be 80,357 mega-electron-volts (MeV), plus or minus 6 MeV. (One MeV is about twice the mass-energy contained within a single electron.) But in a new analysis published on Thursday in Science, physicists on the CDF collaboration have instead found the W boson mass to be 80,433.5 ± 9.4 MeV. The new measurement, which is more precise than all previous measurements combined, is nearly 77 MeV higher than the Standard Model’s prediction. Although these numbers differ by only about one part in 1,000, the uncertainties for each are so minuscule that even this small divergence is of enormous statistical significance—it is exceedingly unlikely to be an illusion produced through sheer chance. The well-studied W boson, it seems, still holds plenty of secrets about the workings of the subatomic world—or at least about how we investigate it. Taken by surprise, particle physicists are only beginning to grapple with the implications.

“Nobody was waiting for this discrepancy,” says Martijn Mulders, an experimental physicist at CERN near Geneva, who was not involved with the new research but co-wrote an accompanying commentary in Science. “It’s very unexpected. You almost feel betrayed because suddenly they’re sawing off one of the legs that really support the whole structure of particle physics.”

Questing for Quarks

It was a rough measurement of the W boson mass that allowed physicists in 1990 to predict the mass of the top quark with reasonable accuracy five years before that particle was first observed. Then, using the W boson mass and top quark mass, researchers made a similar prediction for the Higgs boson—which bore out spectacularly in 2012. More recently, physicists making such measurements have focused less on refining the Standard Model’s core competencies and more on probing its failures—it does not, for instance, incorporate gravity, dark matter, neutrino masses or a number of other troublesome phenomena. Poking at the places where the Standard Model breaks or otherwise deviates from observations, physicists say, is one of the best ways to search for “new physics,” their catch-all term for finding additional, possibly more fundamental building blocks of the universe. Until the CDF result, some of the Standard Model’s most promising discrepancies included an anomaly investigated at the Muon g-2 experiment at Fermilab and results from the LHCb (Large Hadron Collider beauty) experiment at CERN.

Small anomalies are a dime a dozen, and the vast majority are simply statistical fluctuations arising from the truly enormous numbers of subatomic events produced and recorded by typical particle physics experiments. In such cases, those anomalies fade away as even greater volumes of data are gathered. This latest anomaly appears more promising, though, because there is already so much preexisting high-quality information on the W boson’s mass, and the theoretical prediction of the particle’s mass has very low uncertainty. And, perhaps most importantly, the CDF collaboration has been extremely careful. The experiment was “blinded” to minimize the risk of human bias, meaning that physicists analyzing its data were kept in the dark about its results until their work was completed. When the CDF’s measured value for the W mass was revealed to team members in November 2020, “it was a moment of stunned silence,” says the study’s corresponding author, Ashutosh Kotwal. “The realization of what that unblinded number meant—that, of course, is pure gold.”

Since then, the results have gone through multiple further rounds of peer review—but that only guarantees the physicists have done their homework, not that they have found new physics.

Mining the Data

To measure the mass of a W boson, one must first build a particle collider. The Tevatron, which ran from 1983 to 2011, was a 3.9-mile (6.3-kilometer) loop where protons crashed into antiprotons at up to about two tera-electron-volts (TeV)—some 25 times the mass of a W boson. The CDF experiment, located along the loop, sought signs of W bosons in these collisions from 2002 until the Tevatron shut down.

But one cannot simply observe a W boson; it decays into other particles far too quickly to register in any detector. Instead physicists must infer its presence and properties by studying those decay products—chiefly electrons and muons. Counting carefully, the CDF team found about four million events in the experiment’s data attributable to a W boson decay. By measuring the energy deposited in the CDF detector by those events’ electrons and muons, the physicists worked backward to figure out how much energy—or mass—the W boson originally had.

This work took a decade because of the numerous uncertainties in the data, Kotwal says. To reach its unprecedented level of precision—twice as precise as the previous best single experiment measurement of the W boson mass, which was made by the ATLAS collaboration——the CDF team quadrupled their dataset and also used new techniques. These included modeling proton and antiproton collisions and conducting a new, more thorough examination of the decommissioned detector’s operational quirks—even using old cosmic-ray data to map its layout down to the micron.

That was enough to elevate the researchers’ anomalous result to remarkable heights of statistical significance: nearly seven sigma, in the parlance of statistics. Seven sigma means that if no new physics affected the W boson, discrepancies at least as large as the one observed would still arise from pure chance once every 800 billion times the experiment was run. Even in the world of particle physics, where astronomical numbers are the norm, this almost seems like overkill: the field’s “gold standard” threshold for statistical significance is only five sigma, which corresponds to a given effect appearing through chance once every 3.5 million runs. Crucially, the seven-sigma value of the CDF team’s new measurement does not mean that result has a 99.999999999 percent chance of being new physics. It does not even mean other measurements of the W mass are wrong. Rather a seven-sigma result means that whatever the CDF collaboration is seeing is not by chance. It is a call to further inquiry, not a conclusion.

Detective Work

To determine the anomaly’s source, corroboration from other experiments is needed. “It’s a very spectacular result,” says Guillaume Unal, ATLAS’s physics coordinator, who was not involved in the new study. “It’s a very complex and challenging measurement, and it’s also a very important one to really probe the Standard Model with good accuracy.” ATLAS is currently working to improve its measurement of the W mass, and Unal says using data from the LHC’s second run, which concluded in 2018, may allow them to get close to CDF’s precision.

In the meantime, theorists will pounce on this new result to produce myriad possible explanations. Although the LHC has ruled out many permutations of supersymmetry (SUSY)—a set of theories positing that elementary particles have “superparticle” partners—one culprit that could be shifting the W boson’s mass ever so slightly is a cohort of relatively light supersymmetric particles.

“Of course, [the LHC constraints] are becoming more and more stringent,” says Manimala Chakraborti, a theoretical physicist at the Nicolaus Copernicus Astronomical Center of the Polish Academy of Sciences, who is not part of the CDF collaboration. “But still, you can find regions of allowed parameter space for SUSY.”

At a time when new colliders are being proposed, and the LHC is preparing to launch another campaign of collisions after a massive overhaul, the announcement of a seven-sigma-magnitude anomaly from a long-gone experiment whose detectors have been cannibalized may seem strange.

But the collaboration continues to meet to assess and refine the fruits of the experiment’s run. “Detective work itself is what keeps us going,” Kotwal says. “The clues are all there.... It’s like Sherlock Holmes. The person may be gone, but the footprints are still there.”