Fraud In the Polls?

Glenn Reynolds notes that Steven den Beste, one of the best of the bloggers, is back at USS Clueless after a hiatus, and is analyzing the last six months worth of Presidential poll data.
Den Beste’s findings are set forth in a beautiful chart; without looking at it, you can’t understand what he is saying. Ordinarily, we have no compunction about lifting graphics from other sites, but Steven’s chart is the product of so much work that I don’t feel comfortable reproducing it. So: follow the above link, study the chart, read den Beste’s commentary, then return here.
The two things that jump out at you when you look at the data Steven has collected are: 1) the long-term trend favors President Bush, and 2) there is something weird going on in September. Steven writes:

In my opinion, the polls were being deliberately gimmicked, in hopes of helping Kerry. In early August it looks as if there was an attempt to engineer a “post-convention bounce”, but it failed and was abandoned after about two weeks. But I’m not absolutely certain about that.
The data for September, however, is clearly an anomaly. The data is much too consistent. Compare the amount of jitter present before September to the data during that month. There’s no period before that of comparable length where the data was so stable.
The September data is also drastically outside of previous trends, with distinct stairsteps both at the beginning and at the end. And the data before the anomaly and after it for both Kerry and Bush matches the long term trendlines.
If I saw something like that in scientific or engineering data, I’d be asking a lot of very tough questions. My first suspicion would be that the test equipment was broken, but in the case of opinion polls there is no such thing. My second suspicion would be fraud.
In September, I think there was a deliberate attempt to depress Kerry’s numbers, so as to set up an “October comeback”. Of course, the goal was to engineer a bandwagon.

Others have voiced similar suspicions based on the ever-changing internals of some of the polls–oversampling first Republicans, then Democrats. But den Beste is the first person I know of to pull the data together and present it in a fashion that makes the thesis powerful.
Steven concludes: “The reason the Democrats and the MSM are getting frantic is that they’re losing.”
UPDATE: A reader points out that den Beste lifted the chart from Real Clear Politics. Actually, I see now that I just misread Steven’s post. I thought he took data from RCP and made the graph; but I see now that he clearly credited, and linked to, RCP. My mistake. It’s worth pointing out, though, that John McIntyre and Tom Bevan have established themselves as the number one news source for poll data and related fixes for political junkies. When both high-sophistication news outlets (USS Clueless) and low-sophistication outlets (the New York Times) are taking their data from the same source, it shows the RCP guys have cornered the market. Great work.


Books to read from Power Line