Yeah, polling is garbage in general but using it to see relative change is actually like the one thing that it's good for.
So, your assertion is that Biden is slipping lower and lower and lower in the polls the more he does. You picked the one state where he's slipped the most, to make that point. If I did the opposite, I could pick North Carolina, and say that gaining 1.7 points since before he did his press conference means he's killing it, and that press conference restored the confidence of the voters.
Probably a fairly accurate metric -- since you're going to ignore, for reasons which will be obvious to anyone who knows what the national polls show, the national polls -- could be to add up all the swing states and see how things have changed.
In the last week, Biden's gained an average of 0.56 points in all the swing states. If you saying him losing 0.4 points in PA since the press conference means he's losing ground, then I have demonstrated that zooming out to a non-cherry-picked-to-the-single-worst-state view shows the exact opposite happening.
Similarly, in the last month, Biden's lost an average of 0.8 percentage points in all the swing states averaged together. You could write an article about how even in the face of an objectively catastrophic debate performance, less than 1% of the voters abandoned him, pointing to the resilience of his support because most of the voters (unlike the media) are smart enough to realize that one bad debate doesn't all of a sudden mean that etc etc you get the idea. Oh, also, that means he's been gaining ground back since the debate, after dipping lower than 0.8 points initially, which kind of makes sense since the debate was such a horrifying fuck-up.
See? Primary sources are fun. That's all based on the Nate Silver chart of all swing states that you sent me.