By Joe Nocera - “Unbelievable job numbers,” tweeted Jack Welch, the iconic former boss of General Electric on Friday morning, moments after the Bureau of Labor Statistics released its September jobs figures. “These Chicago guys will do anything,” he continued. “Can’t debate so change numbers.”

...Point No. 1: the idea that a handful of career bureaucrats, their jobs secure no matter who is in the White House, would manipulate the unemployment data to help President Obama, is ludicrous. Jack Welch knows it, too; when I called him Friday afternoon, he quickly backpedaled. “I’m not accusing anybody of anything,” he protested. But he went on to add that everything he’s seen suggests that the economy remains in the doldrums, and it just didn’t seem possible that the unemployment rate could have dropped so drastically, and so quickly.

Hence, Point No. 2: there is, indeed, something a little strange about the way the country derives its employment statistics. It turns out that the statistics the bureau releases each month are generated by two different reports. One, called the establishment report, is a survey of businesses. That’s where the 114,000 additional jobs comes from.

The second is a survey of 55,000 households, where people are asked about their employment status. Extrapolating from the survey, the bureau concluded that an additional 873,000 people had found work in September. It is that number that brought the unemployment rate from 8.1 percent to 7.8 percent.

When I asked a bureau spokeswoman why there was such divergence between the two numbers, she said she had no idea. “The reports are totally separate,” she said.

When I put the same question to economists, they shrugged. Maybe it was because an additional 582,000 Americans were working part time, which doesn’t show up in payroll statistics. Maybe it was because of increased government employment. For some unexplained reason, there is always an uptick in September. (“Maybe it has something to do with going back to school,” said Mark Zandi of Moody’s Analytics, who quickly added, “I’m just guessing here.”) In any case, it wasn’t anything economists hadn’t seen before. Sometimes the two surveys delink, though over the long term they tend to reinforce each other. In the short term, however, the household survey is considered the more volatile — and less reliable — of the two numbers.

Which leads to Point No. 3: there is something truly absurd about having the presidential race hinge on the unemployment rate. Even putting aside the reliability of the short-term numbers, the harsh reality is that no president has much control over the economy. That is especially true of President Obama, whose every effort to boost the economy these past two years has been stymied by Republicans. Again and again, they have shown that they would rather see the country suffer than do anything that might help Obama’s re-election.