It just so happens I wrote a bit yesterday about a Zillow poll that ranked Richmond, Virginia as the 7th best place in the United States to buy a home.  And in that post, I alluded to the fact that I typically disdain Zillow data because I have found it to be ridiculously inaccurate in certain sub-markets, and especially with historic properties.  Here's a blog post I wrote on the topic back in March 2009, "The Zillow Rant OR Why Zillow Zucks."

Well, d*mned if an e-mail from Zillow didn't turn up in my e-mail inbox today.  The e-mail header is "How Accurate Are Zillow's Zestimates?"  Then, within the e-mail there is a whole bunch of different information to dig through, including videos, website information, charts, etc.

I wish I could figure out how to attach the entire e-mail, but that's beyond my technological skill set.  I'll settle right now for attaching a link to the web page within the Zillow site titled "Data Coverage and Zestimate Accuracy."  I'm going to take a little time to play around with the information and see what I can see, but in light of yesterday's post, I wanted to share it sooner rather than later.

So, more on this will come later, but here's some basic information.  If you click on "States and Counties," and then "Virginia" and then finally get to the specific data on the City of Richmond's Zestimate accuracy, what you find is only 61% of the Zestimates are within 20% of the ultimate sales price of properties.  I think that's a pretty cruddy record.  And this accuracy data includes those areas, such as subdivisions, where the Zestimate tends to be more accurate.  In other words, I bet in certain City sub-markets, the Zestimate isn't even within 20% of the sales price anywhere NEAR 61% of the time.

But again, more on this topic later.  Lord knows, websites like Zillow and Trulia are not going away anytime soon.  So we're all going to have to learn to play nice together.  As much as I'd like to take all my toys and go home.