2012 election: When in doubt, make it up?

I'm having fun checking out these predictions from some conservative pundits over tomorrow's outcome.  Apparently when the data shows your guy is down, all you need to do is close your eyes, pinch your nose and pick your guy to win.

First, with Rove.  If you ignored his suspect analysis the last month, all you needed to do was look at his electoral map numbers, and see that Obama was ahead.  But with his final electoral map, he apparently decided that he would ignore the aggregate of polls, and simply focus on Rasmussen's latest polling number -- if it showed Romney with a lead despite an abundance of other polls showing otherwise, he marked it red.

Now, this is acceptable if you have poll numbers to suggest that polling had been going your way during the week; that's not what happened:

• In Colorado, three fresh polls (since Rove's last map) show Obama has jumped ahead.
• In New Hampshire four of five polls issued yesterday show Obama ahead with one showing a tie.
• In Ohio, just two polls in the last 32, have shown Romney with a lead; the average of Obama's lead during that stretch is 3.08 pp, and Romney's is 1.5 pp.
• In Iowa, five of the last seven polls since Rove's last map shows Obama with a lead; the average of Obama's lead during the last week is 3.8 pp, while Romney's average lead was 1.
• In West Virginia, three of the last four polls since Rove's last map, show Obama with a lead; if there was a state with a dead heat, this would be it, with both sides averaging 2.0 pp lead in the final week.

Second, oh for the the love of pundits!  No shame whatsoever, these men.  With the exception of Andrew Beyer, Dick Morris, George Will, Michael Barone and Dean Chambers all have Romney winning by large margins; Beyer has a map and margins closer to Rove's.  How'd they do it?

Well, they each selectively ignored the polling data, even if it showed Obama with 5+ point leads, simply because they were labeled as a battleground state.  For instance, Will and Beyer both think Romney will win in Nevada, despite even Rasmussen's last poll showing Obama up by 2, and the last time Romney held a lead in any poll in the state, was April.  The same holds true for Will, Morris and Barone picking Romney to win in Pennsylvania; you have to actually believe that, despite Rasmussen's last poll two weeks ago showing Obama with a 5 point lead, and in a state where Romney hadn't led in any poll going back to February, that Romney would win because of ad money pouring in, late.

It's all good though.  Because here are all these conservative men who have laid down their predictions, just days before the election.  And you know what my prediction was, three months ago -- more on that, after the election.