Des' model gives Trump a current edge and Clinton the prevailing edge. Both sides have a strong chance of winning, but ultimately I think the baton will go to Trump.
Nationwide campaigns are tough nuts to crack analytically speaking. Right and wrong things need to happen on both sides in multiple areas. Things you can't adequately measure until the postmortem analysis unless you are working the campaign.
My view is that it's likely Trump here's why-
1. A tendency to out-perform the polls. That trend will continue to bear out after the primaries due to his draw with non-voters and even Democrats to a varying degree.
2. Many states that should be solid Clinton are in the toss-up category due to voter suppression (in the technical sense, not the legal sense). Clinton needs to mobilize the Obama constituency again if she wants a decisive victory. This isn't going to happen to that extent.
3. Early voting turnout is down overall. Even some states that rely mostly on EV have a Republican edge which is...bizarre.
4. Republicans tend to win on election day because more traditional voters come to the yard.
5. Clinton is coming from a hard late-campaign and has no more punches to land on her opponent.
This is, in many respects, a perfect storm. We'll see just how perfect.
I cannot see Clinton overcoming these factors. While the Electoral College and party stability favor Clinton, I think Trump crossed the proverbial Pyrenees at a moment opportune enough to have his Cannae tomorrow.
It'll then be up for Republican leaders to build a Romanesque system to keep the general in the field.