There have been some different numbers thrown about referring to President Obama's chances for being reelected based upon various economic indicators. If one looks at consumer spending an unemployment, the economy hasn't improved enough to assure an Obama win:
For my first variable, I took the percentage increase in real consumer spending in the last six months of the year in which the election was held, compared with the previous year's last six months. That's a more tangible sign of confidence, in my view, than surveys of confidence.
For my second variable, I examined the rate of unemployment for married males over the last six months of each election year. This provides a more continuous measure of labor's pain over the past half-century than the overall rate of unemployment, whose demographic mix has changed over time.
To compute my measure of voters' economic well-being, or VEWB, I subtracted that bad thing, the married-male unemployment rate, from the good thing, the rate of increase in real consumer spending, and got a single number. [...]
Last November, Obama's VEWB would have been the second-worst since 1956 (behind only Jimmy Carter's in 1980). Sure, the president's number could turn at least flat or maybe even positive by November, but that's quite unlikely. However, as my data show, it isn't always the economy (stupid) that determines presidential elections. But it sure helps the incumbent if the economy is healthy when he seeks a second term.
Do you think such metrics are valuable in determining a candidate's chances or are they better at predicting the past than anything else?