Zillow's home-value estimates criticized
You've probably checked out Zillow.com's "Zestimates" at some point -- those free estimates of a home's value -- so you might have given some thought to how accurate they are. That's the subject of much debate, actually.
The latest volley comes from professors Daniel R. Hollas, Ronald C. Rutherford and Thomas A. Thomson, arguing in The Appraisal Journal that they're really Overzestimates:
The results indicate that Zillow overestimates value for approximately 80% of the houses in the sample by at least 1%. ... The average overestimation is 11.66% or $13,576, with a median of $9,717 or 7.92%. Zillow’s magnitude of overestimation is marginally higher than the value overestimation by recent homebuyers reported in the literature.
The authors -- from the University of Texas at San Antonio -- pitted Zestimates vs. 2006 sale prices of about 2,000 houses in Arlington, Texas. "Zillow indicates that this market is one where its data has its highest accuracy rating," they wrote, adding later: "The likelihood is that in [a] more ... volatile market with a lower accuracy rating Zillow would misprice at a higher rate and larger amount."
Zillow roared back with a response:
"In addition to being limited to only one city in the U.S., the study does not compare sales and Zestimate values during the same time period; it looks at sales in 2006 compared to Zestimate values in January and February 2007 – apples and oranges as it’s two separate periods of time," Zillow spokeswoman Jill Simmons wrote in an email to me. "It’s unfortunate that this study has been structured in such a misleading, and limited fashion."
Here's how the authors describe their methodology:
The sale price data, housing characteristics, and location data are obtained from the MLS for the city of Arlington, Texas (Tarrant County), for sales during the last six months of 2006. The sale price data was acquired in January 2007. Next, data from Zillow.com was obtained for each of the MLS sales to acquire Zestimates during the last week of January 2007 and the first two weeks of February 2007.
And here's how Zillow describes (for a non-mathematician audience) how it arrives at its figures:
When our statisticians developed the model to determine home values, they explored how homes in certain areas were similar (i.e., number of bedrooms and baths, and a myriad of other details) and then looked at the relationships between actual sale prices and those home details. These relationships form a pattern, and they used that pattern to develop a model to estimate a market value for a home.
How does the Zestimate of your home -- or homes near you -- match up with your sense of value?