For the week of August 29, the new release is here.

In the week ending Aug. 29, the advance figure for seasonally adjusted initial claims was 570,000, a decrease of 4,000 from the previous week's revised figure of 574,000. The 4-week moving average was 571,250, an increase of 4,000 from the previous week's revised average of 567,250.

The advance seasonally adjusted insured unemployment rate was 4.7 percent for the week ending Aug. 22, an increase of 0.1 percentage point from the prior week's unrevised rate of 4.6 percent.

The advance number for seasonally adjusted insured unemployment during the week ending Aug. 22 was 6,234,000, an increase of 92,000 from the preceding week's revised level of 6,142,000. The 4-week moving average was 6,216,750, a decrease of 27,250 from the preceding week's revised average of 6,244,000.

The fiscal year-to-date average for seasonally adjusted insured unemployment for all programs is 5.602 million.

We have a drop of 4,000 from last week. Basically a flat line of 570,000. But....we have a revised last week's number, of 574,000, magically increased from the reported 570,000 initial claims from August 22.

So, the 4 week moving average increased by 4,000 new initial unemployment claims and a revised last week's 4 week moving average to 571,250, from last week's reported 4 week moving average of 566,250.

Here is last week's initial claims. The 4 week moving average reported then was 566,250. So, they revised that number by +1000. Last week we had an initial weekly claims of 570,000. So that number was revised up by 4,000.

Wala, while it sure looks like initial claims are increasing, the initial media headlines can claim a drop.

Is your head spinning? I think it should.

Then we have seasonally adjusted, which we still should even out on a yr-to-yr basis.

In other words, a lot of noise in the data and to obtain a clearer picture one needs to look at trends with multiple data points. Amazing isn't it in 2009, we still cannot get accurate real time statistics. If VISA & MasterCard were running this show, they would know not only who just was laid off in 2 milliseconds from the event, but they could also report if that person went out and drank over it, threw themselves over a bridge or just canceled cable.

Oh yes, let's all remind ourselves that only 48% of the workforce can even quality for unemployment benefits.