Author: Xavier Fernández-i-Marín
November 19, 2012 - 2 minutesCatalonia Electoral forecast
After the last 5 polls yesterday we finally have the final forecasting for the catalan elections. Again, it is worth mentioning the historical importance of such elections, both at the Catalan, Spanish and International level.
Although the accuracy of the forecasting in previous elections was quite high, specially for the parties with more that 10% of the support, there are still two areas in which the model can fail terribly. First, in small parties. There is lot of noise in the estimation of the support for small parties. And second, the translation of the support received in final seats in the chamber. The algorithm has not been tested enough, and it relies in two assumptions that can make it fail miserably (namely, the turnout being the same, and the distribution of the support being also the same by districts). So I’m scared about it.
<img title="gr-prediccio_avui-caterpillar.png” alt="gr-prediccio_avui-caterpillar.png” src="http://blog.xavier-fim.net/wp-content/uploads/2012/11/gr-prediccio_avui-caterpillar.png"class="aligncenter” />
All in all, the most likely scenario from the forecasting is a maintenance of the support for the currently governing party, CiU (36.5% against 38.43% in 2010), with a similar number of seats. The real unknown, however, is who will lead the opposition. The silver medal is quiet disputed, with PSC, PP and ERC tightly fighting for it (although coming from very different initial positions). The green party (ICV) slightly improving results as well, and the two small parties with very different forecast (SI disappearing from the chamber and Cs doubling its representation).
That scenario would imply that the right keeps the majority. The topic for the next two years, however, will likely be the referendum for secession, in which case the parties that support such referendum have a very strong majority.