Mike posted the analysis of EPD hit locations when individual BBC tiles see nothing both for the AuAu case and the isobar case. In order to get a quantitive idea of how large/small the compression is, I drew the phi distribution of the EPD hits ring by ring when a certain BBC tile doesn't fire. There are 24 bins covering the whole phi range from -pi to pi. I used the phi of a random point on the tile when filling the histograms so I don't have any empty bins for ring 1. In this study, a "hit" on an EPD tile corresponds to an nMIP value greater than 0.8. 

Some of the plots from the Au+Au case is shown below and a full collection of the plots can be found in the attachment. 


Some of the plots from the isobar case is shown below and again, a full collection of the plots can be found in the attachment. 




Conclusion
In both cases, the 'suppression' is about a factor of 10 and the Au+Au is not showing more suppression than the isobar data.