Karl Rove's Election Prediction Map

RCP is listing it as a toss-up as well. I’d say it leans Romney, but if things swing toward Obama this fall, I’m betting it’s one of the first chips to fall.

NYT reported that Arizona and Virginia are toss-up states for 2012.

The DoJ’s civil-rights lawsuit against Sheriff Arpaio could go some way towards winning Obama the Latino vote there, I daresay.

It occurs to me that he’s giving out pessimistic data to stir up his base.

The sidebar on methodology makes me wonder how much worth these predictions have.

Data that thin might as well come off a ouija board.

No, this is how you stir up the base:

Think about what it means if Rove is truly fudging the numbers in favor of Romney. Even with a favorable push, he’s saying, Romney can’t win. How is that firing up the base? That’s telling them to stay home, save their time and money, and concentrate on 2016.

It’s part of a narrative he’s pitching. He’ll change his tune in the summer and fall, saying we can win states X, Y, and Z, but only if everybody gets out and votes!

You guys are talking like Rove is making an interpretation. Isn’t the map just a compilation of averages of most recent polling data for these states? I think unless he’s just totally fudging numbers there is only so much “narrative building” he could be doing. Rove the Political Consultant makes his money because people think he understands politics, I don’t believe it would help his business if he was seen to be making predictions based on some overarching “narrative.”

At least according to his methodology statement it is an average of all polls within 30 days of the most recent poll, so unless he is lying (and in this scenario it would discredit his reputation in an area of personal importance to his pocketbook) I don’t see that he’s doing anything other than reporting information:

There are a couple things he could do. One, label as “unknown” anything without recent polling data or at least label them more clearly. Two, not publish anything at all if it isn’t favorable to his candidate. The very fact he’s making this public, even if it’s completely true, means that he is probably doing this to increase the chances of his party winning. Which isn’t a bad thing but you need to take that into consideration.

He’s setting a low baseline (from the GOP POV) from which he’ll plot an upward trajectory with subsequent maps. Like anyone else, he can make a pretty good prediction on all but say 4 states. You can bet because it’s Rove, that there’s a strategic component to his map.

Yeah Rove doesn’t put toilet paper to his ass without thinking 14 steps ahead.

And how did he arrive at this particular methodology? Why did he use the average of all polls from within the last thirty days of the most recent poll instead of the last sixty days or the last fifteen days? Or the average of all polls or just the results of the most recent poll? You can bet all thse options were tested to see what the results would look like and then decisions were made about which polls and what range of dates were included. By choosing what data he used, he was able to greatly influence what outcome he achieved.

What, you think a turd blossom just happens?

How can mere numbers capture this sort of raw enthusiasm?

Polls this early are essentially meaningless predictors. Check out the 2008 polls (middle of the page) over time.

Obama has tons more money and I also think he has tons more campaigning skill and likability that Romney. I fully expect these things to work in Obama’s favor over the course of the campaign. Now if the economy takes a turn for the worse then all bets are off. But if the slow, but steady economic progress continues I just do not see Romney winning.

OTOH, Romney has on his side a lot of people who, in spite of everything, really, really, really want him . . . well, want the Republican to win; and some of those have a lot of money and power.

Remember, McCain is from Arizona. That had something to do with the fact that McCain won Arizona so handily (53% to 45%). But most of Arizona’s neighbors went blue (NM, CO, NV, CA), and continuing increase in the voting Latino population would suggest that Arizona is a plausible state for an Obama pickup if Obama’s doing well in general.

What’s interesting about the money situation is how much of Obama’s money comes from small donations versus how much Romney has gotten form those who have already maxed out their $2500. As this shows, Obama has 10 times as much money on hand and has gotten only 19% from big time donors as opposed to Romney’s 57%. It seems to me that Obama will have a much easier time continuing to raise money as thousands of folks kick in $10 and $50 donations a shot. The Obama team has this down to a science. Of course there is the SuperPAC money, which could be the equalizer, but my guess is Obama will have a significant money advantage throughout the campaign.

It’s not hard to see Romney winning all those toss-up states. It would only take something that boosts him 3-4 points nationally. Contagion from the euro crisis could do the trick alone. The important thing to remember are that those toss-up states are not independent events. Winning all of them is not like calling six coin tosses correctly.

Furthermore state polls generally tend to lag national polls and it will be interesting to see how the recent national tightening affects the state polls. At this point the latter are probably not frequent enough to be reliable not to mention the fact that many swing voters will not make up their minds about Romney till the conventions and debates.

I am also not sure how the gay marriage issue will play out in Ohio. The issue was critical to Rove’s mobilization strategy in 2004 and it’s not crazy to think it could boost Romney by a couple of points on election day.

Having said all this I do think the battleground favors Obama. It was always clear just by looking at the 2008 results and checking an electoral map. Romney will probably have to win either Colorado which Obama won by 8 points or New Hampshire which he won by 10 points.

Just to give an example of how unreliable these state polls are here is Coloradofrom RCP. The latest poll is almost a month old and shows a tie. The one before that shows a 13 point Obama lead. The other polls seem to be from last year.

The bottom line is no one really knows what is happening in Colorado right now and it’s probably the second most important state after Ohio. I think Obama will win it in the end but that’s not based on hard polling evidence.