“Adam Bonica”:http://homepages.nyu.edu/~ajb454/ has come up with a “new method for forecasting House elections”:http://ideologicalcartography.com/2010/10/27/forecasting-house-elections-with-fec-records/. Bonica’s method – inspired by “Sandy Gordon’s”:http://politics.as.nyu.edu/docs/IO/2602/short_forecast.htm approach of using the past success of political experts to forecast election results – relies on campaign donors as “experts”. Adam “writes”:http://ideologicalcartography.com/2010/10/27/forecasting-house-elections-with-fec-records/:
bq. [Sandy’s] paper gave me the idea of treating the hundreds of thousands of donors who had given in previous election cycles as de facto expert raters by looking at the percentage of funds given to winning candidates in previous election cycles. The idea behind this is simple. Some contributors give a greater proportion of their money to candidates that go on to win, while others spend the majority of their money on losing candidates. All else equal, the more money a candidate raises from the type of donors who give to winners, the more likely he is to win.
He also notes:
bq. Despite the handicap of excluding all information from polls, InTrade, expert raters, and other data sources used to forecast elections, the model’s predictions are remarkably accurate. In fact, the out-of-sample predictions for House races outperform the polls…. [Additionally, a] major advantage of using campaign finance data to forecast elections is that it costs next to nothing. There is no need to commission polls or pay expert raters. One needs only to collect freely available data and fit a model.
bq. The bottom line: the model favors Democrats retaining their majority with a loss of between 19 to 40 seats.
So we have another suggestion that poll-based forecasting might be overstating Republican gains. In writing this post, I keep coming back to “Sandy’s comment”:https://themonkeycage.org/2010/10/more_on_forecasting_and_reacting_to_2010.html that forecasts of the 2010 elections were beginning to bear all the hallmarks of a bubble: more and more people say it will be a landslide because more and more people say it will be a landslide. I’ve been thinking more about where the “bubble effect” could play out. Certainly one option is in expert prognosis. Another could be in dealing with likely voters. Remember that the trick to getting polls to accurately predict elections – and especially close elections – is to get the relationship between registered voters and likely voters right. I wonder whether we can rule out the possibility that pollsters could be tweaking their likely voter models to get the results where the “bubble” says it should be? Remember as well the concerns about “cell phone only users”:https://themonkeycage.org/2010/10/closer_races_and_cell_phones_i.html and whether these are overstating the likely vote for Republican candidates. So in conjunction with Adam’s results, perhaps a few more reasons to think the poll-based predictions might be running a bit pro-Republican?
Bonica’s predictions are “here”:http://ideologicalcartography.com/2010/11/01/campaign-finance-based-forecasts-for-the-2010-midterm-elections/; his methodology is described in more detail “here”:http://ideologicalcartography.com/2010/10/27/forecasting-house-elections-with-fec-records/.