I try to keep the time that I get the info fairly consistent to be a true reflection of “7 days later” and not 6.3 days later. I will calculate them and post them tonight.

Ardell, you can do absorbtion rates many ways, but your point is a good one that sales volumes do have a seasonal variation in most areas. But I think the point is more to show how long it would take at existing sales levels (or existing pendings) to get an idea of the current market, not to project how long it will actually take for the inventory to sell.

Stated differently, ARs are more to show the strength of the current market, rather than to show anything about the future.

Likewise, the same would happen if posting in July or August and not accounting for the diminished expected activity for October, November and December.

The first quarter is usually 20% of the year. So when we have the 1st quarter of 2008 numbers in, we will be able to project volume for 2008.

And it doesn’t stack the numbers in favor of looking better. If there was one closing in a prior year that was not recorded in the system, the YOY would look better than it should (in an insignificant way though). Correcting the problem would make it look worse (except that I doubt the archived data gets updated).

Look at it this way. Over in the P-I blog Greg was pointing out that looking at pendings, area 710 is approaching having only a three month supply of inventory. When I looked, my numbers were something like 75 pending and 245 active. So if you’re comparing pendings to actives, having 10 off would affect the ratio greatly. Rather than approaching 3 months it would be approaching 4 months. But 710 only had one erroneous pending, which has been sitting there from 2004. Not a big deal in that instance.

But if you compare YOY and only clean up this year, assuming all years have aberrations, then you are stacking the deck in favor of making the numbers look better.