If you mean this season then the sample size is too small, it's probably too small even over the course of a full season come to think of it. And no matter how big the sample size, the only meaningful comparison is one we're not able to make anyway - what would our record be if the goalies were flipped? We can't know that right?
As far as I can tell, there is simply no way to know for sure what would be best so it's all speculation. I'm quite comfortable with my opinion though - starting Andersen against our divisional rival MTL Saturday night is the way to go easily. And same goes for the Boston game earlier this week. And I don't follow other teams closely but I'm gonna go out on a limb and guess that that's the way most other NHL coaches do things, perhaps someone more knowledgeable will chime in here.