The Large Hadron Collider’s Compact Muon Solenoid. Credit: CERN

CERN announced today that it had almost, but not quite, found the Higgs Boson, a subatomic particle that gives objects mass.

Two experimental teams reported that they had seen evidence of the particle between 115 and 130 gigaelectron volts (for comparison, a proton’s mass is just under 1 GeV); but they didn’t yet have enough evidence to declare definitively that they had found the Higgs.

So why all the excitement, for what sounds like no result at all?

It’s because particle physics is a game of statistics, and even a non-conclusive result can be pretty interesting. Protons are slammed into each other at close to light speed in the detectors that are dotted along the 27 kilometer circumference of the Large Hadron Collider, located outside Geneva.

These house-sized detectors capture the sub-atomic debris radiating out from hundreds of millions of collisions per second. Not surprisingly, the data is messy, and finding even a single collision that produces the signature of a rare event amidst this debris, such the one expected when a Higgs particle is created and decays into other particles, requires sifting through many terabytes of data. Because something that looks like the signature of the Higgs could also be produced randomly by the debris, multiple observations must be made, with each detected event contributing to the so-called confidence level of the reported results.

Getting enough data to have a confidence level of 95 per cent sounds is a pretty good start, but it still means that there’s a 1-in-20 chance that the result is due to pure chance—too risky to base the future direction of an entire field of science. Generally, physicists demand a “five-sigma” confidence level, which corresponds to a 99.99994 per cent level of certainly, before popping the champagne.

As reported by The Guardian, at today’s press conference one of the two experimental detectors searching for the Higgs at CERN, called CMS, produced evidence of the Higgs at close to the 95 per cent level, while the other, Atlas, produced results at the 2.3 sigma level, or about 98 percent confidence.

While still not enough for a definitive yes or no on the existence of the Higgs, those odds are good enough for many physicists to start investing the time in seeing how the result factors into many theoretical models, and represent a major milestone in our understanding of the laws that govern the universe

Reaching this point has been a challenge: the Large Hadron Collider is quite probably the most complex machine ever built, as Technology Review showed in this photo essay from May 2008, and as Noble laureate Jerome Friedmann discussed in his accompanying commentary.

Such machines are not without their problems: nine days after the LHC went online a helium leak knocked the machine offline for months. The LHC has still not yet run at the maximum energy levels for which it was designed—when it does so, hopefully the new results will confirm that today’s results are not simply caused by some very bad luck for CERN.