"It was a difficult decision," admitted Roger Cashmore, CERN's director of research to Physics Web. "We wanted to give the experiments the opportunity to better test the effect without compromising the LHC schedule." If the experiments continue to measure the same effects, the extra data will reduce the likelihood that the observation is a statistical fluctuation from 6 in 1000 to 1 in 10 000.

Over the next five weeks the experiments will focus solely on collecting as much data as possible at a centre-of-mass-energy of 206.6 GeV. The gamble will cost CERN about an extra SwF 7 million (about £2.7 m), says Cashmore. However, the decision sets the scene for a nail-biting end to the accelerator's illustrious career.