The probability of Obama winning the Presidential Elections

The probability of Obama winning the Presidential elections in November appears to be pretty robust. Here is why --

(1) Based on public polls, it is becoming evident that Obama is most likely to hold all the states (Michigan appears close but it is trending Obama's way) that Kerry won in 2004. That would give Obama 252 electoral votes. Add to this Iowa which is almost certain to go to Obama -- 7 electoral votes. That gives Obama 259 electoral votes.

On the other hand, McCain -- at this stage -- is not likely to win all the stages that Bush won in 2004 (284 electoral votes.) McCain is almost sure to lose Iowa (7 electoral votes.) Add to this mix, Ohio (20 electoral votes), Virginia (13 electoral votes), Colorado (9 electoral votes), and New Mexico and Nevada (each with 5 electoral votes) -- there is substantial doubt if McCain can hold these states. That puts McCain at about 225 electoral votes.

(2) The more compelling arithmetic is this. Repeatedly, in large numbers of public polls about 37-38 percent of the likely voters identify themselves as Democrats and about 30 percent as republicans. That leaves about 30 percent as independents.

Assuming both Obama and McCain get 80 percent of their party votes, the differential gain for Obama would be about 6.5 percent. In the cross-over vote of 20 percent, Obama would lose about 1.5 percent leaving a net lead of about 5 percent for Obama. Assuming that Obama and McCain split the independents, Obama's overall lead would be about 5 percent.

The only caveat is the potential Bradley-Wilder effect which accounts for about 5-7 percent.

(3) What adds to the odds of Obama's victory are the extra-ordinarily low approval ratings (in low 30s) of President Bush and the perception that the country is heading in the wrong direction (over 64-70 percent of Americans say this.)