Our Method

Accurate real estate valuations for 100 million homes

Graphic of HouseCanary line chart
Current HouseCanary
What is the HouseCanary MdAPE?
MdAPE is the “Median Absolute Prediction Error”. It means that HouseCanary’s predicted value is within 2.7% of the actual sale value half of the time.
At HouseCanary, our constantly improving machine learning models generate the most accurate real estate valuations available. Our current Median Absolute Prediction Error (MdAPE) is an industry leading 2.7%, meaning our middle most projected price is just 2.7% from the actual sales price — and half of our projections are even better.
HouseCanary Monthly MdAPE
updated July 11th 2018

How accurate is HouseCanary in your area?

Regardless of where you get your real estate valuation estimates the accuracy will vary across markets due to availability of data
Availability of data
Availability of data refers to both knowledge of salient property characteristics, and also availability of recent sales transactions of comparable properties.
and different factors that affect prices
Variability in prices
Variability in prices refers to the distribution of prices within a given market, i.e., it is more difficult to value properties in a neighborhood where prices range from 100k to 2m than in another neighborhood where prices range from 100k to 300k.
. Click on a state below to see how accurate HouseCanary's property valuations are in your area.

A lower MdAPE indicates a more accurate property valuation.

*Sample size indicates the number of arms length sales over the last 6 months in which we had an AVM estimate prior to the sale price being known. results updated August 08, 2018

Hundreds of data series are taken into account

Residential real estate sales prices are influenced by much more than recent sales prices of nearby homes.

What factors influence housing valuation?

Home with value bubble

We pull sales and listing data from MLS and county assessor records, and pull insights from multiple other sources such as macroeconomic data, capital markets data, mortgage records, search and social data, and house/parcel data.

Visual of HouseCanary pulling data from a range of services
Visual of real estate price trends

Real estate price trends

Visual of comparable housing sales prices

Comparable housing sales prices

Visual of property attributes compared to those of other properties in the area

Property attributes compared to those of other properties in the area

Visual of home improvement history

Home improvement history

Visual of housing inventory within the local market

Housing inventory within the local market

Visual of local economic and employment trends

Local economic and employment trends

Visual of consumer demand and buying behavior

Consumer demand and buying behavior

We use machine learning and data science to run thousands of simulations

We run 8 individual AVMs, each of which executes different property valuation models. Every day, we use our unique ensemble methodology to run thousands of simulations for each property we track. We customize the weight we assign to various data sources to account for regional differences in market trends, economic factors, and demographics.

Animation visual of AVM's

Feedback loops are used to confirm and improve accuracy

On a monthly basis, we compare our previous property valuations against the actual reported sales prices of homes. This allows us to continually evaluate the accuracy of our property valuations, identify changes in local and regional markets, and tune our models and algorithms to account for local and regional changes.

Feedback loop icon

Need More Information?

Contact us if you have questions or would like to schedule a brief demo.