Whether or not you honestly can't see the difference between a rational conclusion of atheism, verses a post hoc rationalisation for atheism is not my problem.
In my professional life, I'm a transportation engineer. One of the things I do is transportation demand modelling.
When estimating transportation demand of a zone (i.e. a small portion of a city), we generally like to base it on the land use in the zone. However, there are times when we have incomplete information (or when conventional assumptions break down) and we have to figure it out another way.
One technique is just to assume an arbitrary demand: for no particular reason at all, we assume that there are 100 trips between every zone pair. We then assign this to the network, estimate the traffic volumes on all the roads in the study area, and compare our estimates to actual count volumes. Where there's error (and initially, there will be tons of error), we adjust the travel demand for the zones that contribute to the traffic volume on that road. We then repeat this process until the error of the estimated traffic demand gets down to acceptable levels.
This way, by starting with a completely arbitrary assumption and testing it, we'll get to a fairly good understanding of reality... one that's often much better thanan understanding based on plausible inferences but that's never tested.
Basically, I don't care too much about the basis for some initial position. The vital thing - whatever the basis for that initial position - is that your position be tested with evidence. Whatever one's reasons for believing in God or not, we can ask ourselves whether what we see is a better fit with the idea that God exists or with the idea that he doesn't.
... and if every observation we make is consistent with both ideas, then God is irrelevant.
I think this approach is rational and reasonable. Do you?