keys, but I wouldn't go quite so far as to say that the national numbers, especially the swings in the national numbers are completely meaningless.
My guess is that there is at least some correlation between national numbers going up or down and how the numbers are going up or down in the swing states. I have no data to back that up, but it does seem intuitive.
A couple of the links I posted do look at the delegate counts and the polls in the swing states.
In both of those Clinton's advantage has gone up:
FiveThirtyEight - Clinton's lead increased
http://projects.fivethirtyeight.com/2016-election-forecast/
63.7% Clinton
36.2% Trump
Electoral-vote.com - Clinton's delegate lead increased
http://www.electoral-vote.com/
317 delegates Clinton
221 delegates Trump
Electoral-vote.com is more raw numbers from the most recent polls it seems to me, while FiveThirtyEight massages the numbers a bit based upon their estimates of the reliability of the polls, how recent they are etc.
Or at least that's how it looks to me.
The numbers from Predict It are free market estimates of those wanting to put money on predicting the outcome. Presumably there is a mix of those who are partisan, some who are playing gut feelings, perhaps some insiders & also some who try to evaluate things in some detail... from polling data, news articles, gut feelings & perhaps even some insider info.