The only really interesting thing I saw in this post was the discussion with realtors. It's significant that Zillow uses long-term averages, and therefore can't

The only really interesting thing I saw in this post was the discussion with realtors. It's significant that Zillow uses long-term averages, and therefore can't price in abrupt market shifts even once they're common knowledge. If nothing else, it implies that "Zestimates" are a bit dubious in tight urban markets where high variance is more possible. Of course, it also looks like Zillow is aware of that challenge. The ostensibly data-driven part was by far the least convincing section - as soon as I saw the sample size I decided it was nothing more than a starting spot-check.