...because if you do, you might wind up with what Chris Cillizza wrote over the weekend, in which he concludes that Mitt Romney's ceiling is basically what George W. Bush got in 2004 -- that he "he has a ceiling of somewhere right around 290 electoral votes."
That's just plain wrong. It's an illusion -- it looks like that now, because Barack Obama is up a bit in the national polls, and because no Republican has done better than Bush '04 for quite some time. But as Harry Enten tweeted, if that was the case Bill Clinton couldn't have won a large electoral college victory in 1992. Indeed, the late 1980s was full of talk about a Republican "electoral lock," which magically went away as soon as a Democrat led in the national polls by a few points.
Real simple: the parties are roughly even nationally, all things equal. All things are generally not equal, which means that in any particular presidential election it's possible to see fairly large swings, in both directions. If the swing produces more than about a three point win, the system will produce an electoral college landslide, just as it did for Barack Obama in 2008.
Seriously. Look, the campaigns have to pay at least some attention to the electoral college, because they need to target their resources. But if you're trying to figure out who is going to win the election, or the margin of victory, you're just going to be a whole lot better off if you pretend the electoral college doesn't exist until at least after the conventions.