Reasonable acceptance rate coexist with posterior spikes

[jpfeifer]

As we are talking about prior-posterior plots, I would need to see the mod-file or the prior.
See also here [Evaluation (robustness) of the Bayesian estimation results)