Why the two numbers disagree
The average selling price is exactly what it sounds like: total sale value divided by number of sales. It moves whenever the mix of homes that sold changes. A month heavy with detached sales pushes it up; a month heavy with condos pulls it down, even if no individual home changed value.
The MLS® HPI benchmark is built differently. The Canadian Real Estate Association describes the benchmark home in its MLS® HPI methodology backgrounder as a virtual, middle-of-the-road home for a given property type in a given neighbourhood, assembled from the typical features of homes that have sold there. It does not necessarily correspond to any real house. Because the benchmark holds those features constant, it isolates price change from mix change. That is why the industry treats it as the better gauge of trend.
October's own numbers illustrate the difference neatly. In the City of Toronto, the average selling price rose slightly year over year, from $1,069,683 to $1,075,627. The City of Toronto benchmark fell 3.6% over the same period. Prices did not rise. The mix of what sold shifted.
What neither number can tell you
Here is the limit that matters most for owners. Neither measure is your home's value.
The benchmark tells you how a typical home of your type in your area has moved. Your home has its own lot, condition, layout, updates and street. Two semi-detached houses on the same block can sell tens of thousands of dollars apart. Use the benchmark as a direction and a rough magnitude, not as an appraisal.
If you need a defensible number — for a refinance, a lender switch or a listing decision — the answer comes from an appraisal or a comparative market analysis of recent nearby sales, not from an index.
Readers who want the longer view of this measurement debate will find it in Canadian Home Prices Sit 3% Below Last Year Even as Average Sale Price Edges Up 0.6%.