25 September, 2010

RCP Strangeness


Real Clear Politics has suddenly changed their Poll Averages on both the Kirk vs. Giannoulias and Brady vs. Quinn races -- without there having been any new polls released.

RCP currently has Mark Kirk at +2.0, when they used to have it at +1.3. Bill Brady is currently listed as +9.3, when it used to be +9.2. And like I said, neither race has had a poll released to explain the changes.

Best guess is, RCP discovered some discrepancy in their math, and have now corrected it. I noticed earlier this week that Kirk's RCP average actually went down (from 1.9 to 1.3) with the last Rasmussen poll, despite the fact that the poll showed Kirk up by 3 percentage points. However, I took that as a sign that more numeric weight was being attached to the previous poll (which showed Kirk with a 4-point lead on a sampling almost double Rasmussen's), and hence an overall narrowing of the average. But I guess something still wasn't funky, so it appears RCP took a look to see what was going on.

Now. Watch the Giannoulias campaign accuse RCP of trying to tip the election in Kirk's favor by "cooking the polling books", as it were.

No comments: