In this study, we use the concept of the linear recharge oscillator (ReOsc) model to diagnose the ENSO-dynamics in CMIP3 and CMIP5 model simulations. The ReOsc model parameters allow us to quantify SST and thermocline damping, SST coupling to thermo- cline and vice-versa, sensitivity to wind stress and heat flux forcings and separate atmospheric from oceanic processes. Our results show that the ENSO-dynamics and their diversity within the CMIP ensemble can be well represented with the linear recharge oscillator model diagnostics. We also illustrate that the ENSO dynamics show larger biases relative to observations and spread within the models than simple large-scale statistics such as SST standard deviation would suggest. The CMIP models underestimate the atmospheric positive and negative feedbacks, they have compensating atmospheric and oceanic errors, the thermocline damping is too strong and stochastic noise forcings in models is too weak. In future climate change scenarios we find significant changes in nearly all of these important elements of the ENSO dynamics, despite the fact that the ensemble shows very little changes in the overall ENSO variability.