Electoral Statistics
All throughout the primary season, I was religiously following Jay Cost on RealClearPolitics.com. His phenomenal skills at extracting every last bit of useful information out of the voting results and exit polls are second to none. Check out his data rich four part demographics review of how Obama prevailed over Clinton.
These days I get my statistical simulation kicks at 270toWin.com, which currently gives a 74% chance of a McCain victory in November. This is in contrast to a more naive approach presented by Electoral-Vote.com, where they simply assign a state to whoever is leading at the moment, regardless of the polls' margins of error. The latter site, which also happens to exhibit some liberal undertones, predicts an Obama landslide by over 100 electoral votes, although just two months ago they projected a narrow defeat. Simply put, as far as I can tell -- and contrary to the popular belief -- Obama's ascent to presidency is still far from assured.
These days I get my statistical simulation kicks at 270toWin.com, which currently gives a 74% chance of a McCain victory in November. This is in contrast to a more naive approach presented by Electoral-Vote.com, where they simply assign a state to whoever is leading at the moment, regardless of the polls' margins of error. The latter site, which also happens to exhibit some liberal undertones, predicts an Obama landslide by over 100 electoral votes, although just two months ago they projected a narrow defeat. Simply put, as far as I can tell -- and contrary to the popular belief -- Obama's ascent to presidency is still far from assured.