Can the Democrats create a firewall?

[°Leprechaun]

It's way too early to predict accurately, but my guess is option 2. Yes

I'd would have called it a "blue" wall, but that might confuse, because of the atlas color scheme.

[°Leprechaun]

There are three scenarios, I am not predicting these exact maps, but use them to illustrate a Trump win, a bad loss, or a landslide.
A win for the Democrat regardless of how big that win is, may be the most likely scenario. However, if the election is close (and it still could be) Trump could win.

Trump wins

Trump loses badly

Landslide for the Democrats

Which map would you chose, then tweak it, or are these maps way off?

[°Leprechaun]

Depends on your definition of firewall. I don't think either side has a resounding victory in the case of them winning. My guess, though it's extremely far out and none of us really know what we're talking about, is that it comes down to a few close states and nobody has a huge mandate (e.g. winning WI by 5 points, PA by 6 or whatever).

This concept was talked a lot about in 2016. To have a real firewall, it would have to be solid.
If polls show Trump losing WI, PA, & MI by more than 10 points (for example).
That would be a solid firewall. If this stays in place until a day or two before the election, that would make the election results seem inevitable, but certainly given the almost certainty of 2016, more skepticism will be advisable.
It is possible that another state could replace WI since it is the weakest of the three.
MI seems likely to flip, unless Trump gains between now and a year from now, and PA is probably more likely to flip than WI which brings the D to 268, not quite enough for a firewall.
ME2 and NE2 could bring the D to 270 but those would be narrow wins, perhaps not as likely as WI or some other state.

edit: AZ as a lean D state is over-rated. It may be fool's gold and a Trump win there would not surprise me at all. It may be more likely a tossup if not a lean R.