For the past two years, over 3,800 teams have been competing for the $1 million prize from Zillow to beat the current Zestimate, Zillow’s tool for measuring the value of houses priced on the popular real estate marketplace. The winner has now been named as Team ChaNJestimate, a team comprised of US and international data scientists and engineers who have built a system that will push Zestimate’s current nationwide error rate from 4.5 percent to below 4 percent.
Originally working separately as two separate teams, Jordan Meyer, from the United States, joined Nima Shahbazi from Canada and Chahhou Mohamed from Morocco for the final round. Together, the team developed an algorithm, using machine learning techniques, eliminating outlier data points, and incorporating necessary external factors such as rental rates, commute times, and home prices.
While most home valuation tools can never be perfect, Zestimate’s current estimations are, on average, off by about $10,000. The improvements made by using Team ChaNJestimate, as well as other competing team’s insights, enables Zestimates to be about $1,300 closer to the most accurate sale price.
Zillow also chose two winners during the qualifying rounds, Team Silogram-2, who won the second-place prize of $100,000 and Team Zensemble, who won $50,000.
“We’ve been on a 13-year journey making the Zestimate more accurate, and hosting Zillow Prize allowed us to invite thousands of brilliant data scientists from around the world to join us on this journey. We’re so proud that the winning team’s huge achievement, and the work of all the teams in the competition, will provide millions of homeowners with a better understanding of one of their biggest life investments,” said Stan Humphries, chief analytics officer, and creator of the Zestimate.
