Stan Humphries's Q&A

Hi mkgkrause. Wetdawgs' comment below has the correct link. Give us the address and some background and somebody from my team will look into it to see if it's within the expected tolerance or whether there is something systematic going on in your area. Regards, Stan

Hi psnoyes.  Can you provide the address for your property?  Just saw your other query about seasonality in your valuation.  You can send me a direct message via Zillow as well if you don't feel comfortable sharing your address on this public thread.  Thanks. Stan

Hi Richard.  Can you give the address for a particular example of this occurring?  Thanks. Stan

Hi jrtpapa.  We've had two major versions of the Zestimate algorithm now with the third to debut in the coming months (don't worry, we'll let you know when it hits the site).  As Pasadenan said, we've re-estimated complete history once previously (and will do so again) and make periodic tweaks to the base algorithm, sometimes globally, sometimes in just a few counties. The models are self-learning and designed to correct for systematic error (or what you refer to as "consistent bias") so, in general, one should find the pattern seen in Pasadenan's charts, assuming one is looking at a sufficiently large sample of homes.  Specifically that means actual sales that are equally above and below their estimated value.  When looking at a smaller region of homes or a narrow, non-random subset of homes, it is conceivable that one will find a biased pattern such as you report.  As wetdawgs notes, we regularly update our accuracy stats but these only contain information on absolute error (how far from the actual sale price the estimate is regardless of the direction of the error), not raw error (which indicates systematic error).  Since the latter is nominally zero and is fairly confusing to most people (not anyone on this current thread apparently), it is not reported. While we don't often discuss too many details about our models, I can't help but reveal that linear regression is not among the techniques we utilize and almost all of our models are trained at the sub-county level (and we use a lot of them; 334,000 created each night while producing Zestimates).  Hope this helps.

Hi dolphintnks.  If you're referring to listing information on for-sale or rental properties, then details about supplying us with a feed of that data can be found at: http://www.zillow.com/feeds/Feeds.htmIf you encounter problems, please contact feeds@zillow.com. Hope this helps. Thanks. Stan

Hi Nathan.  You can contact me. Most of our publicly available data is located on our Real Estate Market Report pages and is freely downloadable.  Hope this helps.

Ideally, the definition that a user should use when editing their home facts is that which the county used in computing the number that is found on Zillow's home detail page.  Unfortunately, we don't have a way of specifying the myriad definitions used by different counties so it's left up to users right now.  Don't worry if the sqft/bedroom is low when using your county's definition.  If recently sold homes around your home use the same definition of sqft, then the models will learn the relationship that exists between sqft and bedrooms, on the one hand, and sale price, on the other hand.

Hi Robert.  My contact info is on my profile.  Drop me a note and we'll see if we can be of help.  Stan

How long ago did you notice the $498K Zestimate?  In early 2008, when we released a new algorithm, we re-computed historical time series for all homes in the U.S. so the peak value might have changed before and after this re-computation.

Hi crpaulk.  Looks like you changed the square footage for 2013 H Ave on 9/29/09 and the square footage for 2011 H Ave on 1/28/10.  In both cases, the square footage changed from 1,023 to 2,000, a large change that is pushing up the Zestimate on both homes.  However, because you changed 2013 four months before 2011, the former has moved up more than the latter (because it started increasing sooner).  Both are still moving up though and I would expect both to reach approximately the same level, with perhaps with 2011 being higher since it has a higher tax assessment.  I don't believe we currently consider garage spaces in our local models in that area. Hope this helps explain what's happening. /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Table Normal"; mso-style-parent:""; font-size:10.0pt; font-family:"Times New Roman","serif";}