For the past two weeks or so, many on the right have been wringing their hands about a series of polls showing Barack Obama leading Mitt Romney, sometimes outside the margin of error. And although Gallup showed the race narrowing into a tie last week, this week we’ve seen the president build and expand his lead on the Republican nominee.
In two posts today, Jim Geraghty addresses this, suggesting in the first post that the Republican is behind, but not by much. In the second, he cites one of his “regulars, he accounting-minded poll watcher nicknamed ‘Number-Cruncher’” who offering a wealth of data* comparing polls in recent years to the final result, notes that pollster errors seem generally overwhelmingly to favor the Democrats.
Now, if this was just random chance of mistakes, you would see pollsters being wrong in both directions and by about the same margin in each direction at the same rate – sometimes overestimating how well the Democrats do some years, sometimes overestimating how well the Republicans do. But the problem seems pretty systemic – sometimes underestimating the GOP by a little, sometimes by a lot.
Read the whole thing.
*In 1992, for example, Gallup overestimated Clinton’s final margin by six points. Sixteen years later, they would overestimate Obama’s margin by four.