Losing a million dollars has never been so easy. Or so painless.
Consider this. In the past week, the value of my house has dropped from $6.6 million to $5.5 million.
Should I worry? Sink into spiraling depression? Cry, perhaps? Or just forget about it?
I’m going with the last option. The fact is I really don’t care.
The current estimate is still millions of dollars more than I ever imagined. And, unfortunately, it’s millions–and millions–more than I’ll ever get if I actually sell the house.
For some unfathomable reason, my house hit the jackpot on Zillow. In the virtual world of real estate, where statistics and data are easier to obtain than a real life referral to the best broker in town, the 1875 Victorian in suburban New York is worth as much as a beachfront estate in Southern California.
The popular real estate website currently Zestimates–values–the property at $1,538 per sf, significantly higher than the $348 per sf value it gives comparable properties. Last week, when Zillow valued the house at $6.6 million, the sf cost was even higher.
Technology makes data ubiquitous, but data can be deceptive.
The issue is balancing usability with accuracy. To get good results from a scientific perspective, site developers have to incorporate complex algorithmic solutions. But they also have to keep things simple, just to encourage the target audience to use the site. In the end, does usability trump accuracy?
One way or another, it’s something to keep in mind. I know, for example, the estimated value of my own home is significantly off. But would I recognize an apparent error as quickly if I were using online data to get ballpark figures on investment properties in another state?
Zillow spokesperson Amanda Hoffman said the Zestimate is computed through an algorithm that uses public records to estimate the value of homes. “Nationwide our median margin of error is 8.8%, and we measure our accuracy based on comparing historical sales prices to what the Zestimate was at that point in time,” she says. “To look at it another way, across the nation, Zestimates come within 5% of the sales price 32% of the time, within 10% of the sales price 54% of the time, and within 20% of the sales price 77% of the time.”
The accuracy varies by area. Where I live, Hoffman said the median margin of error is 11.1%, and the Zestimate comes within 5% of the sales price 26% of the time, within 10% of the sales price 45% of the time, and within 20% of the sales price 72% of the time.
“I’m not sure what you think the true value of your home is, or how far off the Zestimate is, but if you think it’s a lot more than 20% off, we’d be happy to do a little investigative work to figure out why that is,” she said.
So that’s what I’m doing. I asked Hoffman to check the discrepancy…unless she knows of anyone who wants to buy the house for $5.5 million.