So let us look at this...
These are last season's EPL statistics when either, both or none of Romero and Micky played:
View attachment 21305
You would expect that our best defensive performance would be when both played, followed by when one and worst with none. It does not follow that pattern. The only statistically significant difference is between the games when both played vs when only one of them played. Most interestingly, there is no statistical difference between games when both played (which only included two games against top 6 opposition) as opposed to when neither played.
Not saying that having Romero and/or Micky in the team is not important; but the perception that this season's much better xGA is because at least one of the two was always available is not borne by science... it is much more logically plausible that the improvement is down to the coach; even though some will never give him that....