Gallup’s daily updated tracking poll of the presidential race shows Romney with a solid 6 point lead.

But that’s far out of step with the general weight of the polling. Real Clear Politics’ average of the polls, which includes Romney’s big Gallup edge, shows Obama with a .1 percent edge. Nate Silver’s much more complex model at FiveThirtyEight shows Obama with a 1.1 percent edge right now.

As it turns out, there are plenty of reasons not to trust Gallup’s numbers in its likely voter model.

From Ezra Klein’s About that Gallup poll: Is Romney really up by 7? And will Obama win the election anyway?:

Dig into the poll, and you’ll find that in the most recent internals they’ve put on their Web site — which track from 10/9-10/15 — Obama is winning the West (+6), the East (+4), and the Midwest (+4). The only region he’s losing is the South. But he’s losing the South, among likely voters, by 22 points. That’s enough, in Gallup’s poll, for him to be behind in the national vote. But it’s hard to see how that puts him behind in the electoral college.

So if you accept the Gallup tracking poll as definitive, as Jennifer Rubin at The Washington Post rather lamely did today, then you’re embracing a nightmarish scenario of a popular vote/electoral vote divergence in which Romney gets the most votes but gets beaten handily in the Electoral College. Obama can lose the entire South (including Virginia and Florida), and still win the election, as the map at left shows.

Of course, Gallup’s own numbers are nonsensical. Obama is entirely likely to get swept in the South, but he’s unlikely to lose by more than 10 points in Georgia and is likely to be within a point or two (one way or the other) in Florida. It’s just not plausible that the other states could drag that margin so much higher for Romney.

Also, the divergence between Gallup’s likely voter result (+6 for Romney) and registered voter result (+1 for Romney) reflects an extreme edge — one that seems very unlikely — for Obama among registered voters who won’t vote.

And Klein points to another clear problem with the likely voter model:

“The likely voters model takes into account changes in the response to questions about how closely they’re following and how enthusiastic they are,” [Frank Newport, editor-in-chief of Gallup] said. “It’s not just capturing underlying movement — it’s representing changes in enthusiasm.”

There was certainly a surge in enthusiasm among Romney supporters after the first debate and many dispirited Obama supporters too. But everybody still just gets one vote, and polls have showed movement within a very narrow range for months now.

Also, FiveThirtyEight has pointed out some absurd swings in Gallup’s election polling going back many years. See Gallup vs. the World:

In 2010, Gallup put Republicans ahead by 15 points on the national Congressional ballot, higher than other polling firms, which put Republicans an average of eight or nine points ahead instead.

In fact, Republicans won the popular vote for the United States House by about seven percentage points — fairly close to the average of polls, but representing another big miss for Gallup.

Apart from Gallup’s final poll not having been especially accurate in recent years, it has often been a wild ride to get there. Their polls, for whatever reason, have often found implausibly large swings in the race.

In 2000, for example, Gallup had George W. Bush 16 points ahead among likely voters in polling it conducted in early August. By Sept. 20, about six weeks later, they had Al Gore up by 10 points instead: a 26-point swing toward Mr. Gore over the course of a month and a half. No other polling firm showed a swing remotely that large.

Then in October 2000, Gallup showed a 14-point swing toward Mr. Bush over the course of a few days, and had him ahead by 13 points on Oct. 27 — just 10 days before an election that ended in a virtual tie.

As Silver notes in that post, these wild swings aren’t reason to give up on Gallup. It’s likely tracking enthusiasm in ways worth noting, even if not a good predictor of actual outcomes. (Other Gallup products seem to fare much better, however.)

It’s going to be a really close election — likely no more than 2 points separating Romney and Obama. Obama seems to have the edge in both the popular vote and the electoral college, but a lot can happen in the next three weeks.

Tagged with →  

One Response to What’s up with Gallup tracking poll giving Romney a big lead?

  1. [...] I wrote a number of posts critical of Gallup both before and after the general election in November. In mid-October, reading the work of Nate Silver and others, I even asked: What’s up with Gallup tracking poll giving Romney a big lead? [...]