I have been a marginally winning player (4 bb/100) at 5NL since I started playing a few months ago. I spent some time filtering my database to find spots where I'm torching money on fire. It turns out I'm losing 940 bb/100 when my action is "BC" on the flop, which is nearly identical to my "BF" line at -938 bb/100.

Filtering further, I can see that I'm more-negative in certain spots (OOP and as the pre-flop caller), but I'm not sure what type of EV is reasonable when faced with a flop raise. On one hand, my "BC" range is often going to be weak compared to villain's "R" and "XR" line so I expect this to be a pretty marginal spot; other other hand, I am able to selectively Call/Fold/Raise, so there should be an EV difference between those lines.

This is over a 40k hand sample, where a "BC" or "BF" line was taken 190 times. I expect variance to be affecting the results, and I know this isn't enough information to get help in specific spots. That said, how should I set my expectations after facing a flop raise? Am I just terrible post-flop (it's microstakes, so yes), or is this actually a tricky spot?

Filtering for specific circumstances and looking at winrates is not going to be useful at all for a 40k hand database. For this situation and filtering for this circumstance, getting reasonable bb/100 wr is likely going to require >>10mil hands just to put it into perspective.

For the situations you listed and as you filtered them, then having a negative wr is probably normal/fine. EV is typically taken as from the point of decision, but when filtering with a database like this it is showing your overall EV from preflop (which is likely negative).

To give you a quick and easy example of what I'm trying to say.... let's say you open preflop from the cutoff for 3 big blinds and face a 3-bet from the BTN... folding would yield -300bb/100, but any action that yields greater than that is preferred. For example, calling something like 88 might end up being -220bb/100, but that is still better than folding.