The comic above was published by XKCD the day after the 2008 elections, and when users hovered their mouse over the comic, a text window read "Someday I'll be rich enough to hire Nate Silver to make all my life decisions. 'Should I sleep with her?' 'Well, I'm showing a 35% chance it will end badly.'"
Nate Silver is referenced in webcomics because he developed models to predict the 2008 Presidental and Senate elections for his blog FiveThirtyEight. This year, he has expanded his forecasts to include House and gubernatorial elections. I have a tremendous amount of respect for his site and find it slightly more addictive than crystal meth. However, I have to disagree with his projection for SC-02. Silver's model projects that Joe Wilson will win 59.2% of the vote and Rob Miller will get 38.0%, and that Wilson has a 99.4% chance of winning. Now, Representative Wilson is an incumbent Republican, in a good year for Republicans and a strongly Republican district, so I do think Wilson is the favorite, but I think that the margin of victory in the end will be much closer than 21 points.
Simply put, Nate Silver's model doesn't know that Joe Wilson is the "you lie!" guy. Incumbent politicians who are embroiled in scandals tend to struggle in their reelection bids. In September 2009, Wilson yelled "you lie!" at President Obama during his speech to Congress on health care, sparking national outrage. In addition, the House ethics committee is investigating Wilson's use of travel funds on Congressional trips to Afghanistan, and Miller has criticized Wilson sharply for this in advertisements.
Joe Wilson certainly has raised an impressive amount of money: $4.5 million, to be exact. However, Rob Miller has raised $2.8 million, and the combined fundraising total of $7.3 million makes this race the most expensive U.S. House race in the country. Despite Wilson's fundraising prowess, an expensive race isn't good for an incumbent. It's an indicator that the incumbent is vulnerable and both parties are prepared to fight it out over the seat. I think we can take the sky-high fundraising totals as a sure sign that this race will be a close one.
Finally, the prediction of a 21-point win just doesn't match the polling or the history of this district. A poll in early October put Wilson ahead of Miller by just 7 points, and no other poll has been published since early May. This race is a rematch of 2008, in which a non-scandal-ridden Wilson beat Miller 54-46. Now, one could argue that 2010 will likely be a much better year for Republicans than 2008 was, and I agree in general. However, I don't think that the shift in the nation's mood is strong enough to cause a 13-point swing in this race, especially given Joe Wilson's shenanigans since November 2008. I think the poll showing a 7-point Wilson victory is much more likely to be accurate than Nate Silver's projection of a 21-point Wilson victory, and it might even be closer than that poll shows.
This is a good point here. As advanced as Silver's formula may be, does it take into account variables such as scandals as is the case in this district? He does heavily rely on polls, which would indicate that voters are still supporting him despite all of the moral problems surrounding Wilson. However, the lack of a recent poll does bring into question Silver's accuracy in predicting who will win this district. It will be interesting to see how close he will be to correctly predicting the winner here.
ReplyDelete