Page 1 of 2
Model, accurate since 1980, predicts Romney in a landslide
Posted: Thu Aug 23, 2012 6:07 am
by _bcspace
Two University of Colorado professors, one from Boulder and one from Denver, have put together an Electoral College forecast model to predict who will win the 2012 presidential election and the result is bad news for Barack Obama. The model points to a Mitt Romney victory in 2012.
Ken Bickers from CU-Boulder and Michael Berry from CU-Denver, the two political science professors who devised the prediction model, say that it has correctly forecast every winner of the electoral race since 1980.
"Based on our forecasting model, it becomes clear that the president is in electoral trouble," Bickers said in a press statement.
To predict the race's outcome, the model uses economic indicators from all 50 states and it shows 320 electoral votes for Romney and 218 for Obama, according to The Associated Press. The model also suggests that Romney will win every state currently considered a swing state which includes Florida, Virginia, North Carolina, Ohio, Pennsylvania, Wisconsin, Minnesota, New Hampshire and Colorado.
The professors' model shows a very different picture than what current data suggests. Currently, The Huffington Post's Election Dashboard shows Obama with 257 electoral votes to Romney's 191 with only six "tossup" states including: Colorado, Florida, Iowa, North Carolina, Ohio and Virginia.
Berry cautions that just because the model has worked in the past, doesn't mean it will work this time. "As scholars and pundits well know, each election has unique elements that could lead one or more states to behave in ways in a particular election that the model is unable to correctly predict," Berry said in a statement. Some of those factors include the timeframe of the current economic data used in the study (the data used was taken five months before the November election, but Berry and Bickers plan to update it with more current data come September) as well as tight races. States that are very close to a 50-50 split, the authors warn, can fall in an unexpected direction.
http://www.huffingtonpost.com/2012/08/22/university-of-colorado-pr_n_1822933.html?icid=maing-grid10%7Chtmlws-main-bb%7Cdl1%7Csec1_lnk3%26pLid%3D195989
Re: Model, accurate since 1980, predicts Romney in a landsli
Posted: Thu Aug 23, 2012 8:39 am
by _Drifting
Will it be Satan's fault if Romney doesn't get in?
Re: Model, accurate since 1980, predicts Romney in a landsli
Posted: Sat Aug 25, 2012 7:14 pm
by _Brackite
I am Not Predicting here that either Mitt Romney or Barack Obama will win in a landslide in this coming up Presidential election.
I have now Predicted here that this is going to be so very close of a Presidential election, that this will end up going to the U.S. Courts. I think that the State that will be so very, very close in will be at Ohio.
This Presidential election is very likely going to be like Bush vs Gore was back in 2000.
Re: Model, accurate since 1980, predicts Romney in a landsli
Posted: Sat Aug 25, 2012 8:47 pm
by _EAllusion
This is making the rounds in conservative circles. The "accurate since 1980" line is quite misleading, as this model was invented recently and retrodicted back as far as 1980. But when you look at how they did that, there's some hokey reasoning that makes it appear that they retroactively fit the data with fudge factors rather than came up with a coherent prediction model. This is a shame, as there are solid prediction models out there doing real work. They tend to predict a narrow Obama victory, incidentally.
Re: Model, accurate since 1980, predicts Romney in a landsli
Posted: Sat Aug 25, 2012 11:31 pm
by _moksha
If this model was not around in 1980, then claims of its predictive powers could also be post dated to 1790. You need to think longitudinally when on a Rocky Mountain high in Colorado.
Re: Model, accurate since 1980, predicts Romney in a landsli
Posted: Mon Aug 27, 2012 5:31 am
by _bcspace
Will it be Satan's fault if Romney doesn't get in?
Ultimately, yes. But I would place the blame on those who serve Satan such as Democrats and those who don't vote even though eligible to do so.
Re: Model, accurate since 1980, predicts Romney in a landsli
Posted: Mon Aug 27, 2012 6:24 am
by _Bond James Bond
My model successfully predicted the election of every President and predicts a narrow Obama victory. You can have access to my math when we have access to Romney's tax returns from 2004-2009.
Also a Romney landslide is hilarious. Unless Obama kills a white woman during the Convention he's got 200+ votes locked up tight.
Re: Model, accurate since 1980, predicts Romney in a landsli
Posted: Mon Aug 27, 2012 11:23 pm
by _EAllusion
Bond James Bond wrote:Also a Romney landslide is hilarious. Unless Obama kills a white woman during the Convention he's got 200+ votes locked up tight.
Europe continues to teeter on the edge of a broader collapse. That could give Romney a landslide. Otherwise, it's hard to fathom how that is going to happen. Nate Silver's model currently gives it a 1% shot, and that feels like a generous nod to modeling uncertainty.
The model that thread is in reference to is an econometric model. Collectively, those give a prediction of a minor Romney victory. This model tends to be on the extreme end of pro-Romney, and even calling its results a "landslide" is dubious.
Re: Model, accurate since 1980, predicts Romney in a landsli
Posted: Tue Aug 28, 2012 12:28 am
by _Brackite
Gallup now has Obama at 46%, and it now has Romney at 47%.
http://www.gallup.com/poll/150743/Obama-Romney.aspx
Re: Model, accurate since 1980, predicts Romney in a landsli
Posted: Tue Aug 28, 2012 1:54 am
by _EAllusion
How strange that you never post any polls, the vast majority up to this point in fact, that show Obama up. Why might that be?