Presidential Election Prediction

UPDATE (11/09) -  Not too great on this one. Pretty abysmal in fact. Lesson- 1) Don't listen to GOP operatives who speak against major sustained polling analysis.

Against all the odds (and perhaps some clamors of bias) I am predicting that Romney will win the Presidential Election by an electoral college margin of 295 - 243. I believe that his route to victory will be found via wins in the swing states of Colorado, Ohio, Pennsylvania and Virginia.

Here's how he wins. 

First- the 'swing' states that I believe are now 'safe' for each candidate. It is my opinion that the traditional conservative support which exists in Florida and North Carolina will place those states firmly in Romney's bracket. Conversely, I also now believe that Wisconsin, Iowa and Michigan will go for Obama. The union presence and traditional Democratic affiliation of many voters in those three states will be a great boost for the Obama vote. 

Having banked the electoral college gold mine of Florida (29 EC VOTES), victory in the following swing states will enable Romney's victory.

COLORADO (9 EC VOTES) - Romney has closed Obama's lead in Colorado. Obama will be unable to replicate his 2008 voting success among younger voters, while Romney will attract greater support from independents and from a considerable Republican base which is far more energized than in 2008. Colorado will be close, but I believe it will fall into Romney's bracket come election day.

OHIO (18 EC VOTES) - By most estimates, Romney has closed the poll margins to within the margin of error. Winning Ohio will require a number of actions to fall in Romney's favor. First, Romney must be able to mobilize significant proportions of his base to get out to the polling booths. Based on the energy with which Republican leaning voters are now engaged in the election,  I believe this will occur. Today's New York Times has a good piece on both campaigns voter outreach efforts in Ohio. Second, Obama will need to see a depressed turn out of polled likely voters who currently say they will support him. I also expect that this will occur. Come election day, reduced enthusiasm for Obama will mean that a small but significant minority of his leaning supporters do not turn out to vote. Third, as yet undecided voters (who are a small minority of the likely voting total) must break heavily in Romney's favor. Again, I believe (and historic 'late undecided' trends suggest) that Romney will be able to persuade these voters to enter his name.

         To become reality, Romney's victory in Ohio will therefore center on three key actions- his campaign's ability to 'get out the vote', the Obama campaign's inability to motivate a comparative level of enthusiasm and the influence of undecided voters breaking in Romney's corner and turning up at the polling stations to reflect their decided opinion.

PENNSYLVANIA (20 EC VOTES) - For me, has become the most exciting state in the run up to election day. As recent polls have broken towards an increasingly pro-Romney trend, the Romney campaign has made Pennsylvania a major target. Interestingly, one of the most recent Pennsylvania polls has put the race at 47%-47%. The Romney campaign evidently made an error in not engaging a major focus on Pennsylvania before October. Regardless, now that the polls have tightened and the Romney campaign is expending resources in the state, I believe that Romney can pull out a shock win. Romney's victory in Pennsylvania will require that rural, conservative leaning voters turn out in significant proportions. In addition, Romney must undercut Obama's 2008 lead in urban/suburban areas. Ultimately however, I believe that Pennsylvania will go for Romney for the same reasons behind my suggestion of a Romney Ohio victory. Put simply, between likely voters and the few undecideds still in the race, I believe that Romney will be able to attract more voters than Obama. From research and from having spoken to people on the ground in Pennsylvania, I firmly believe that the Romney campaign has a statistically significant advantage over the Obama campaign in terms of voter enthusiasm. For me, enthusiasm is the crucial, always understated element of electoral success. Being a likely voter is NOT the same as turning up at a polling station.

VIRGINIA (13 EC VOTES) - As with the other swing states, Romney has closed the gap on Obama since the debates. The polls in Virginia remain very close - some show Romney leading, some Obama and some even. Ultimately, as with the other states, 'getting out the vote' will be critical for Romney if he is to win. Romney will need to generate high turnout in the traditionally more conservative southern areas of Virginia. In addition, Romney will have to undercut Obama's lead in Washington DC suburbs like Fairfax, Alexandria and Arlington. Again, I believe that undecided voters will break in Romney's favor and that the enthusiasm of the Republican base will propel Romney to victory.

CONCLUSION - The above predictions are definitely bold. However, I believe that Romney can and will achieve an upset victory on Tuesday night. The enthusiasm of the Republican base exceeds that of the Democratic base. And enthusiasm gets voters to the polling station. Romney now has comfortable metrics in terms of likability, leadership and ideas. These strengths will place him in a good position to take advantage of undecided voters who get to the voting stations. Obama will be unable to rely upon the base of support he received from young Americans in 2008. He will also attract far less independent voters. Taken together, these dynamics matter. Voting wins elections, participating in a phone poll does not. And... as Harry showed us in 1948, it's not over until its over.

Thoughts on CBS/NYTimes Polls

CBS/NYT have just released their latest polling data. From an objective standpoint, I don't believe that the data is accurately reflective of the race as it stands at the moment. These new poll results show Obama leading in Florida by 48-47%, in Ohio by 50-45% and in Virginia by 49-47%.

My issues with the results are as follows.

1) The polls also suggest that Romney is leading with Independent voters. The Virginia poll indicates Romney's lead with this voting block is significant. If this is the case, Obama cannot be ahead. Voter affiliation is broadly split about 30-35%/total affiliation- for both the Democratic Party and the GOP. Independent affiliation is estimated at around 35%/total affiliation. So... it doesn't take a genius to work out that the support of a credible majority of independent voters will lead to an overall poll majority. For me, this issue alone renders these latest polls as weak data sets.

2) Obama leading in Florida? The trends in Florida poll data over the last three weeks have shown systematic and robust movement in Romney's favor. For Obama to have retaken a lead seems to me a highly unlikely prospect. I am highly confident that Florida will go Republican on November 6th. The significant amount of time that Obama has spent in Florida recently provides a clear indication of his campaign's concern over the state.

3) The Ohio data is broadly reflective of poll data over the past few weeks. This data has indicated Obama holding a confident base margin of 3-5% points advantage over Romney. My personal belief is that Obama currently has a 1-2% point lead with likely voters who have not yet voted, but a 2-3% lead when early voters are included in a grand tally. I believe that many Ohio voters have given up answering the phone seriously when the pollsters call. It gets tedious when you get 3-4 calls a day.

CONCLUSION- Speaking to friends (both Democrat and Republican) who are engaged with ground level politics, the feeling remains that Romney has the momentum but that Obama retains a lead in Ohio. My personal belief is that Romney has a reasonable opportunity to win Ohio- but that this outcome will require the few remaining as yet undecided likely voters to join his camp. However, I also believe that Romney has a good chance at winning Wisconsin (where the GOP ground game is formidable) and Iowa (which is rural conservative). In addition, I think that Romney will win Colorado, Virginia and Florida. Having considered the polling trends over the last few weeks and having spoken to a lot of skilled politicos, I now genuinely believe that Romney has a route to the White House that does not involve Ohio. Is this route tough? Yes. Does it afford Romney room for error? No. But... is it possible? Yes.

UPDATE 11/04 - Wisconsin and Iowa look to be consolidated behind the President now. I was too optimistic about trend data vs union capacity to GOTV.