White Projected To Lose Governor's Race

One of the most respected election prognosticators in the business is Nate Silver of the political blog FiveThirtyEight.com, who has recently been hired by the New York Times as a political columnist. Silver is not a pollster, but probably knows as much about polls and polling as any pollster or politician.

As a mathematician, he examines the polls and their strengths and weaknesses (such as who was polled, what was asked and how, size of the poll, length of poll, etc.), and using mathematical models combines the averages of all polls taken, poll biases, demographic data, political climate and historical results to make a projection of what will happen on election day in individual races. He was uncannily accurate in the 2008 election.

If he can be believed, and I think he can be, then Bill White (the Blue Dog Democrat) has very little chance of unseating Republican Rick Perry. Silver has been projecting this race since February. His initial projection in February was that White had a 21.5% chance of winning. His projection of White's chance of winning since then has never been over 17% and currently resides at 16.5%. That's a very slim chance of winning, especially considering it has remained very constant for the last eight months.

Silver also projects what he believes the vote percentage will be for each candidate on election day. These projections have also been remarkably constant -- meaning there has been very little (if any) movement in this race since February. Here are the projections since February:

FEBRUARY

Perry...............51.8%

White...............45.4%

MARCH

Perry...............52.2%

White...............44.8%

APRIL

Perry...............53.3%

White...............43.7%

MAY

Perry...............52.7

White...............44.4%

JUNE...............53.7%

White...............42.9%

JULY

Perry...............52.2%

White...............44.4%

AUGUST

Perry...............52.8%

White...............44.0%

SEPTEMBER

Perry...............51.9%

White...............45.1%

The polls (weighted for bias and other factors) that were used in making these projections were:

Rasmussen (1/17, 2/1, 2/22, 3/3, 4/14, 5/13, 6/16, 7/13, 8/22)

Public Policy Polling (2/4-7, 6/19-21)

YouGov (2/1-7, 5/14-20)

Texas Newspapers (2/2-10)

It is easy to see from these numbers why Bill White desperately wants to debate -- and just as easy to see why Rick Perry has no interest in granting him that debate. Unless White pulls off some remarkable feat or Perry commits some incredible blunder, Perry's chances of winning in November are exceptionally good (about 83.5%). With these numbers Rick Perry would be a complete fool to submit to a debate.

There is still about two months until the election and anything can happen in that long a time. However, I have to believe that unless Bill White can significantly change these numbers by the first part of October, this race is over. And that won't be easy considering the remarkable constancy the race has shown over the last eight months.