Red herring, because this is a discussion about terrorism.
Oh, I am sorry; let me review my comments, in the specific context of terrorism.
The fact is that when people are the victims of terrorist violence, the most likely perpetrator is a fellow countryman - and religion tends to run in countries.
Your correlation is real; but it tells us exactly nothing useful, so it's utterly irrelevant.
Truth is not at issue here.
It doesn't matter whether or not you can identify a large class of people to whom a perpetrator is likely to belong; Unless you can identify a very small group of people to whom a perpetrator might belong, the information is of no value whatsoever.
I can predict with a very high degree of certainty that the next armed robbery committed in Las Vegas will be committed by one of the ~ 0.2% of the US population who are currently resident in Las Vegas. That's a MUCH smaller suspect pool than the ~1% of the US population who are Muslims; Just as soon as you can formulate a reasonable plan to prevent armed robbery in Vegas based on residency, we will be able to use the exact same methodology to leverage your incredibly useful statistic to prevent terrorism.
I look forward to seeing your proposal(s) that can use the known and very clear correlation between armed robbery in Las Vegas, and residency in Las Vegas, to protect the citizens from armed robbers.