We calculate partial Bayes factors to quantify how the feasibility of the constrained minimal supersymmetric standard model (CMSSM) has changed in the light of a series of observations. We take as “training” data the approximate knowledge that was available before LEP, and take our comparison model to be the Standard Model with a simple dark matter candidate. Partial Bayes factors are then computed, using as “inference” data the LEP2 Higgs constraints, 2011 XENON100 dark matter constraints, 2011 LHC supersymmetry search results, and the early 2012 LHC Higgs search results. We find that LEP and the LHC strongly shatter our trust in the CMSSM, reducing its posterior odds by a factor of approximately two orders of magnitude. This conclusion is robust under variation of priors, but may be avoided if the CMSSM is not required to explain the (g-2)µ anomaly.