McCain Brings Hope to Pennsylvania

There had been some pushback from McCain operatives about whether the Republican in fact intended to concede Colorado, as CNN had reported yesterday. In politics, however, money speaks louder than words, and the New York Times is now reporting that McCain may not spend further advertising monies in Colorado, as well as four states won by John Kerry in 2004:

Democrats who monitor advertising spending now put at five the number of states where Senator John McCain is reducing his advertising – New Hampshire, Wisconsin, Colorado, Maine and Minnesota.

In essence, Mr. McCain’s campaign has decided to spread the advertising time he bought for the upcoming week in those states over the next two final weeks.

While station managers in the affected states said they were not ruling out the possibility that Mr. McCain would pump money back in before election day, on Nov. 4, the move represents a stark reordering of priorities.

Democrats were predicting Mr. McCain would use the savings to increase his advertising in Pennsylvania and, possibly, Ohio and Florida, all of which have become that much more vital should Mr. McCain have to concede states like Colorado and Wisconsin.

Note that it's not just Colorado on the chopping block, but also Minnesota, Wisconsin and New Hampshire. Michigan was conceded some weeks ago. Iowa and New Mexico are on life support. Essentially, McCain seems to be giving up on any path to victory that does not involve Pennsylvania -- a state that we presently project Barack Obama to win by 9.7 points.

Al Giordano has a good theory about what McCain is up to:

But here's what I think is going on at McCain strategy central: They're getting tired of the daily drumbeat on cable TV news and by newspaper pundits that says things like, "here are the six or seven swing states, all of them voted for Bush in 2004, Obama is winning or tied in most of them, and for McCain to win he has to run the table, taking every single one of them or it's over."

That message - that there is only one narrow Electoral College path to victory for McCain, while there are multiple ones for Obama - has cast a deathly spell over the GOP base's enthusiasm, which is now being reflected in paltry early voting numbers by Republican voters, especially in Nevada and North Carolina. And so they're trying to offer the faithful a belief in the suggestion that McCain, too, has multiple paths to win.

The senior staff seems to think it has convinced McCain to drop his reluctance to play the race card, with trial balloons afloatin' that Obama's ex-reverend will get an encore in the coming days in negative ads and such.

And if they're really going to go there - to try to make the campaign about race and, specifically, some white people's fears of pigmentation - then it would make total sense for McCain to temporarily ignore Colorado, where that message ain't gonna hunt, and shift focus to Appalachia and the South: Virginia, North Carolina, Ohio and, yes, Pennsylvania and even Florida being the swing states where racially charged politics have sometimes, in the past, worked for the Republicans, or, in Appalachia, where they worked for the Clintons during the primaries.

Actually, Al's theory is quite a bit more nuanced than that, and deserves a read in its entirety; among other things, Al thinks that McCain may be trying to use Pennsylvania, a state where his numbers have nowhere to go but up, in order to create a sense of "momentum" that may pay dividends in other states.

But if McCain's strategy is taken more at face value, it would seem to imply -- as Al suggests -- a potential re-appearance by the Jeremiah Wright All-Stars. That would conform with another detail reported in the Times piece, which that the McCain campaign wants to control its advertising message all for itself, rather than having to share with the RNC:

But the McCain campaign also needs the extra money to keep up with its current plans, due to a quiet decision it has made that most voters will hardly notice.

Until now, the campaign has been teaming up with the Republican National Committee to jointly produce a large percentage of its advertisements. By sharing the costs down the middle, Team McCain has been able to basically double the amount of advertisements it can run for its money.

[...]

The campaign has started to phase out those ads in these final days, deciding to stick to advertisements it can devote fully to Mr. McCain’s campaign message. That will greatly disadvantage Mr. McCain as he struggles to keep up with the far better funded Mr. Obama. But Mr. McCain’s aides have clearly decided a trade of volume for greater clarity is worth it.

There are real downsides to this: the RNC has a lot more cash on hand than the McCain campaign does. Without RNC assistance, McCain will be in no position to win a conventional air war, even if he limits his expenditures to small number of states.

Something like a Jeremiah Wright ad, however, would receive lots of free air time on afternoon talk shows and on the Internet, somewhat mitigating McCain's cash disadvantage.

The big problem with such a strategy, however, is this:

4,060,647
2,917,747
869,707

Those are the current numbers of registered and active Democrats, Republicans and independents in Pennsylvania. Democrats make up more than half the total -- 52 percent, in fact -- well outdistancing the Republican's 33 percent. Suppose that McCain were to split Pennsylvania's independents with Obama and win Republicans 92-8. He would need to carry 23-24 percent of Pennsylvania's Democrats to win the state; George Bush carried 15 percent.

As we reported yesterday, however, negative advertising does not seem to be a good strategy for winning over lapsed Democrats; on the contrary, Democratic solidarity has increased markedly in recent weeks, with Barack Obama now on target to win as much support among his party any Democratic nominee has in any recent election.

Frankly, I think Al may be giving the McCain campaign too much credit. My guess is that something like this happened: they ran their usual set of internal polls over the weekend, and saw themselves 5 points down in Colorado and Virginia. 8 points down in Minnesota and Wisconsin, and 10 points down in Iowa and New Mexico. But perhaps Pennsylvania came in at a -6 or something -- not much worse than the others -- and they decided: why worry about all those states when we can worry about just this one.

One of the things I emphasize at Baseball Prospectus is the importance of honest self-assessment. A team can get itself into tons of trouble by convincing itself it has 87-win talent -- making it a fringe playoff contender -- when it fact it has 80-win talent -- making it an also-ran. The same lesson probably applies to internal polling. In addition to the usual problems of optimism bias -- and the unresolved question of whether internal polling is in fact superior to public polling (especially for a campaign that is poor at voter contact) -- campaigns sometimes forget that internal polls contain of margin of error. If you're in the field in a dozen or more states each weekend, you are all but certain to wind up with one or two outliers. Perhaps Pennsylvania was such an outlier for John McCain.

If a campaign gets an internal poll that diverges from the consensus of public polling, it needs to ask itself why the divergence exists. If it cannot explain it, it should probably not treat the internal poll as actionable.

But as reported by the Los Angeles Times, the McCain campaign does not seem to have any particular reason why they think the public polls are wrong in Pennsylvania:

[McCain Political Director Mike Duhaime] said the campaign is operating three dozen offices in the state and is making hundreds of thousands of phone calls every week to identify and persuade potential GOP voters. The data mining efforts are aimed at identifying former Hillary Clinton supporters and independents who are prepared to consider McCain’s message. He said the internal data is “trending” in McCain’s direction and is showing “a lot of things” not apparent in the opinion polls.

Overall, Duhaime said McCain has drawn strong support from what he called a Democrats for McCain movement in and around Scranton, in the state’s western Rust Belt region. “That gives us optimism,” he said.

McCain anticipates good news as well, he said, in the south and central part of the state, near Harrisburg, York and Lancaster -– all cities that the candidate, his wife, Cindy, or running mate Sarah Palin have visited in the last few days.

What is the key phrase in that passage? "Anticipates good news". As in, the McCain campaign does not have any particular idea how they're going to win Pennsylvania, nor why the public polls have the state wrong -- they're just hoping their numbers are right, and hoping that something comes together for them.

As a famous Democrat once said, of course, hope is not a strategy.