I feel like you didn't read the OP.
I know that HPF on everything is a bad idea. I know that whatever I capture is what I've got to work with. And I know that trial and error is ultimately how I'll figure this out for myself.
The question again, is: How does HPF at the source differ from HPF in the box, if it does at all, specifically when it comes to phase issues in the low end? I'm looking for opinions/experiences with these two techniques from people who have used this type of filter in both situations.
I don't know where I've gone wrong here... I thought my original post was pretty clear.
I think everyone has tried to answer the question, but maybe jumped on the premise you stated about doing it to everything. Ok, you get that was a mistake.
First, the big difference at the track level, answered by many, is that you may be cutting out something that is fundamental to the source signal, if you apply a HPF at the source. So, it's really important to understand the frequency content of your source, and exactly what the HPF is is cutting, and how much. If you cut even a little of the source fundamental, you have a track that is not what was original. You might try to boost something in the missing area, but it's not the same.
Now, that, of course, doesn't mean to not use them, and that's part of your clarified question, what's the difference between ITB and something earlier. I guess it depends, but generally you don't want stuff that isn't part of the source to be something that can actually corrupt/distort the source signal, and unwanted low frequencies almost certainly have higher frequency overtones/harmonics that are going to interact with your source, as well as contribute to the amplitude of what is hitting all the components from start to finish. You may have trouble understanding why the vocal sounds low when it seems like you recorded at proper levels, when the problem is very low or subsonic content in the source signal. You can take it out ITB, but it may have already introduced other noise into your track that you cannot get out without taking part of the content you want away. In an ideal studio, you'd filter everything that wasn't part of the source. We do the best we can with mic placement and HPFs at home, but only when we can do it knowing that we're not taking away from a fundamental.
I did find this snippet (as it applies to mastering, but certainly true at the track level) about HPFs, and how a shelf might be a better choice in some places. I think a corollary (my assumption) is that when you use them, you want to be pretty precise on the cutoff frequency and slope, so they don't all end up doing bad things at exactly the same place. Doing something like engaging a hardware HPF switch (with no other controls) on a preamp and using that on every single track might be setting you up for some problems.
Your Audio Mastering Questions, Answered | Are You Listening? Season 2 Episode 7