Zillow vs. Reality: All Tri-Cities Zestimates®

Between January 16, 2019, and June 14, 2019, we recorded 532 home sales in the Tri-Cities. The data below reflect all Tri-Cities home sales that we tracked.

Median Error Rate

Median error rate is the primary metric that Zillow uses to gauge the accuracy of Zestimates®. In late January 2019, Zillow said its median error rate nationwide is 4.5 percent. That means half of all Zestimates® are within 4.5 percent of the actual sales price.

Zestimates® in our area are more inaccurate than they are across the country as a whole. Based on our research, the median error rate of Zestimates® in the Tri-Cities is 5.6 percent. In other words, half of the Zestimates® here are wrong by more than 5.6 percent of the actual selling price and half are wrong by less than 5.6 percent.

Zestimates® for Tri-Cities homes are also worse than in our neighboring big cities: Seattle’s median error rate is 2.2 percent and Portland’s is 1.5 percent, according to Zillow’s own tracking.

Among the four cities we studied, Pasco had the most accurate Zestimates® with a median error rate of just 2 percent.

Are Zestimates® Too High or Too Low?

In the Tri-Cities, Zestimates® are usually higher than the actual sales price. Of our 532 homes tracked, the Zestimates® of 420 homes (79%) were higher than the sales price. The Zestimates® of 112 homes (21%) were lower than the actual sales price.

Range of Zestimate® Inaccuracies

On its website, Zillow measures how many Zestimates® are more than 5, 10 and 20 percent wrong. We’ve done the same thing with our Tri-Cities data.

5 Percent Wrong

52 percent of all Tri-Cities Zestimates® were off by 5 percent or more.

10 Percent Wrong

30 percent of all Tri-Cities Zestimates® were at least 10 percent off.

20 Percent Wrong

9 percent of all Tri-Cities Zestimates® were at least 20 percent off.

Zestimate® Accuracy & Inaccuracy in Dollars

We can also look at the accuracy/inaccuracy of Zestimates® in terms of how many dollars they were off when compared to the actual sales price.

• 85 homes (16 percent) had a Zestimate® that was at least $50,000 off. That’s about one of every six homes sold.
• 35 homes (7 percent) had a Zestimate® that was at least $75,000 off.
• 10 homes (2 percent) had a Zestimate® that was wrong by $100,000 or more.

On the flip side:

• 131 homes (25 percent) sold with a Zestimate® that was within $5,000 of the sales price. That’s about one of every four homes sold.
• 60 homes (11 percent) had a Zestimate® that was within $2,500 of the sales price.
• 25 Tri-Cities homes (5 percent) sold with a Zestimate® that was within $1,000 of the sales price.

Most & Least Inaccurate Zestimates®

Among all 532 home sales/Zestimate® we recorded, the most egregiously wrong Zestimates® were:

• A Kennewick home sold in May with a Zestimate® that was $208,335 more than the actual sales price — the most inaccurate Zestimate® by dollar amount.
• A West Richland home sold in February with a Zestimate® that was 392 percent higher than the actual sales price — the most inaccurate Zestimate® by percent.

The least inaccurate (or most accurate) Zestimate® was a Richland home that sold in May for just $13 more than the Zestimate®.

Monthly Trends

Over the course of our study, Zestimates® improved at each level of inaccuracy. The percentage of Zestimates® that were wrong by at least 5 percent dropped from 53 percent in February to 50 percent in May. The greatest improvement was seen in Zestimates® that were wrong by at least 10 percent, which dropped from 32 percent to 24 percent.

(Note: For monthly trends, we’re not including January and June because we don’t have a full month of data for either.)

Navigation

1. Zillow vs Reality: Introduction
2. Zillow vs. Reality: Key Takeaways
3. Zillow vs. Reality: All Tri-Cities Zestimates® (this page)
4. Zillow vs. Reality: Kennewick Zestimates®
5. Zillow vs. Reality: Pasco Zestimates®
6. Zillow vs. Reality: Richland Zestimates®
7. Zillow vs. Reality: West Richland Zestimates®
8. Zillow vs. Reality: Home Value Comparisons
9. Our Methodology