LHCb's One-in-Ten-Billion Decay: First Evidence of B⁰→μμ After a 42-Year Search
On 3 November 2026, at a seminar at CERN, the LHCb collaboration announced something particle physicists have been chasing since the early 1980s: the first evidence, from a single experiment, that a neutral B meson can decay into two muons. The process, written B⁰→μμ, is predicted by the Standard Model to happen only about once in every ten billion such particles. After a 42-year search, the signal has finally risen above the noise — at 3.2 standard deviations, enough to call it evidence, not yet enough to call it a discovery.
That distinction matters, and it is the whole story. In particle physics, 3 sigma is "evidence"; 5 sigma is "discovery". A 3.2-sigma result means there is roughly a one-in-1,500 chance that the bump in the data is a statistical fluke. It is a strong hint, but hints have dissolved before, and everyone in the room at CERN knows it. What makes this one different is the machine behind it: the upgraded LHCb detector, running since 2024, collects data at up to five times its previous rate, and this analysis combined the new 2024–2025 data with everything recorded between 2011 and 2018, sifted with machine-learning techniques and particle-identification methods sharp enough to pick a handful of real decays out of billions of collisions.
The result also sharpens a second measurement. Alongside the B⁰ signal, LHCb produced the most precise measurement yet of its sibling decay, B_s→μμ, first observed in 2015, as well as its lifetime. Both measurements agree with the Standard Model — which, paradoxically, is both reassuring and slightly disappointing. These rare decays are prized precisely because they are places where new physics should show up first: any undiscovered particle or force operating at extremely short distances would nudge the decay rate away from the Standard Model's prediction. So far, the rulebook holds, but the test has now moved into territory no experiment has reached before.
Physicists at Birmingham, who lead parts of the analysis and built key detector technology, describe this as a window onto the "Zeptoverse" — distances around a billionth of a trillionth of a metre, where undiscovered particles and forces may be hiding. That framing is honest about what this result is: not a crack in the Standard Model, but a new instrument pointed at the place where a crack would appear. A B meson lives for about one and a half trillionths of a second, and in that instant it samples physics at energies and distances no other probe can reach. Watching ten billion of them decay, and counting the handful that go to two muons, is one of the most precise null tests in science — and a null test that keeps passing is itself information, because it rules out whole families of theories that predicted a different rate.
The question now is how long it takes to cross from evidence to discovery. Significance grows roughly with the square root of the data collected, so moving from 3.2 sigma to 5 sigma needs roughly two and a half times the effective signal — a plausible haul for the upgraded detector over the next few years, but not a guaranteed one, since analysis improvements, detector performance, and the sheer luck of how the next data fluctuates all play a part. That is the prediction I am making with this article: that B⁰→μμ will reach the 5-sigma discovery threshold by the end of 2028. If it does, the rarest decay ever seen at the LHC becomes a confirmed fact of nature. If it fades, that too will be informative — a reminder that in this field, the line between a breakthrough and a fluctuation is measured, not wished, into existence.