Imagine, if you will, it is an election year and pollsters being pollsters keep asking voters a "generic congressional ballot" question. Further imagine that through that entire year the Democrats retain a lead of anywhere from 1 - 19 points in the "generic" results. Let's say it is the last weekend in October before said election and the "big time" polls all show the Democrats ahead: Gallup by 1%, same over at Zogby, and Rasmussen has the Dems up by 6%. Why, it wouldn't be unimaginable to believe the Democrats were in for some good news come election night. However, that next Tuesday after the first Monday in November, the Democrats are stunned to learn they LOST four seats in the House. How can that be? They were favored for the entire year in the "generic congressional ballot" poll question and maintained that lead right up until election day?!?! Well, those results aren't imagined - that's exactly what happened in 2004. In June of that year the Dems were swooning over the 19 (!) point lead they enjoyed in the LA Times completely unbiased poll, but then come Election Day not so much with the swooning. Perhaps, it is because we do not have parliametary elections where we vote for the party nationwide; rather we have 435 seperate little elections where many factors - aside from the national "mood" - come into play. Perhaps, it could even be that most polls are conducted by those withering dinosaurs known as newspapers who might just hold the slightest little bias toward one party over the other.
Some conservatives - like the Weekly Standard's Fred Barnes - suggest you must add about 5 points to the GOP total in these "generic" polls to make them more accurate. That may explain why in 1994, the Democrats held a 6% lead in most generic polls immediately before getting their clocks cleaned. I think they are - 5 points added or not - pretty silly exercises that don't mean a whole helluva lot come Election Day. I am not saying that this year the GOP will gain even more seats in the House. However, I think the Democrats basing their House takeover on these numbers is foolhardy. After all, their lead in this year's LA Times June "generic" poll was only 14 points - Ouch!
BORU