Gallup, Rasmussen, and Ipsos/Reuters were the first pollsters to cover the Wednesday debate because they do rolling tracking (they poll every day and average the results over a set span). Gallup shows a single point gain for Obama, Rasmussen and Ipsos/Reuters show no change at all. All of these polls include results from before the debate.

These are very small data points subject to lots of noise, but added (not literally added, more like averaged) together, the margin of error drops significantly. With all covering at least one full day of post-debate sentiment, there has been no change in the 2012 general election. Most people believe that Mitt Romney "won" the debate, and almost nobody changed their mind because of it.

Barack Obama remains a big favorite to win based on state polling and electoral math.

Gallup (7-day)

9/24 - 9/30: Obama +4
9/25 - 10/1: Obama +6
9/26 - 10/2: Obama +4
9/27 - 10/3: Obama +4
9/28 - 10/4: Obama +5**

Rasmussen (3-day)

9/30 - 10/2: Obama +2
10/2 - 10/4: Obama +2**

Ipsos/Reuters (4-day)

9/23 - 9/27: Obama +7
9/24 - 9/28: Obama +5
9/27 - 10/1: Obama +5
9/28 - 10/2: Obama +5
9/29 - 10/3: Obama +5**

**Covers debate.

Jobs

As for the September job report, the topline number of jobs added was disappointing, well below what the economy needs to generate just to keep up with population growth. But it wasn't all bad. July and August were revised upwards. BLS regularly does this for the previous two months as they attain more survey data and can generate more accurate reports. It's not a con game or fudging. It's like running a pubic poll and releasing results when you have 1,000 people responding (one month), and then releasing them again when you have 3,000 people (two months) and 5,000 (three months).

Sometimes these an older report gets revised down, which is what happened in September when June and July were revised down. Funny how nobody (read: conservatives) complain about rigging when that happens.

These reports, folks, are created by what are called career civil servants. They can't be fired whenever a new administration comes into office and replaced by political appointees. The people who made these reports are largely the same people who made them when George W. Bush was president, and will make next year if Mitt Romney wins. And the data, while not peer reviewed before release, is peer reviewed after. Private sector analysts pour over this data every time it comes out and if it were cooked, everyone would know about it very quickly. That has never happened.

Moreover, as TPM notes, the BLS report was very close to private market projections released on Thursday and Friday morning. Before the report came out, via CBS:

For much of the week, financial markets have been sluggish, with a raft of economic data failing to alter market expectations that the data will show the U.S. economy generated some 110,000 jobs during September.

From ADP survey (WaPo):

A private survey shows that U.S. businesses added fewer workers in September than August, a sign that slow growth may be holding back hiring.

Payroll processor ADP said Wednesday that companies added 162,000 jobs last month. That's below August's total of 189,000, which was revised lower.

And then from BLS today:

The economy added 114,000 jobs in September and unemployment declined to 7.8 percent.

The consensus report from CBS only missed the mark by 4,000 jobs.

As for the unemployment rate dropping, it's just simple economics. The "jobs added" topline are full-time jobs. Part-time jobs are counted in U3, the unemployment rate that everyone uses. That's why the rate dropped. The economy only added 114,000 full-time jobs, but it seems to have added quite a few more part-time jobs. Not ideal, but it's more people back to work regardless.

On another note, it shouldn't be surprising that the team that denies evolution, climate change, Obama's citizenship, and recently public polling, is now denying job reports. This is getting very, very bad. A country and a society cannot survive when half the population denies reality.