October 2017: Detached, Richmond Hill


  • Richmond Hill detached home sales down 54.8%, listings up 144.0% YoY
  • Sales to active listings ratio (SALR) is trending downward
  • Average Richmond Hill detached home price is down 12.1% YoY
  • Median Richmond Hill detached home price is down 13.6% YoY

You can read the original October 2017 TREB Market Watch publication here.


  1. Great charts and info Hanny, thanks! I’m curious, I see varying figures used from different sources. For instance, the TREB site is showing $1,319,500 as the average detached home price in Oct. for Richmond Hill and $1,555,400 as the peak in May … for a 15% drop to date. The figures you have here ($1,345,898 in Oct. and $1,884,611 in April) suggest it’s been a 28% drop. Do you know why there’s a difference between the figures?

    • Hi Nick,

      Glad you love the charts, and thanks for the question. The numbers I use on my site are the average and the median prices published by TREB in their monthly Market Watch report. In your example above, $1,345,989 is the average price for a detached home in Richmond Hill in October 2017 and $1,884,611 is the average price for May 2017. You are correct, if you calculate the percentage change between these numbers, you arrive at -28.6%.

      Now here are some caveats:

      A shortcoming of using an average is that it can be influenced by unusually high or low values within the data set. Shifts in housing mix (ie. changes in the number of high-end homes versus lower-end homes) over time can make it difficult to compare averages over time. It’s not a perfect measure, but you will notice that it’s quoted by TREB and the media regularly.

      I also publish the median price for the different property types. The median is the middle point that divides a series into two equal parts (ie. 50% of the transactions occur below the median and 50% occur above). The advantage of the median is that it is not impacted as much by extreme numbers. However, major shifts in mix could still impact comparability of the median over time. I personally believe that it’s the best available number to use, and I am surprised that it’s not quoted more widely.

      As for the $1,319,500 number that you quote from TREB – this number is the “Benchmark” detached home price in Richmond Hill. The “Benchmark” is a proprietary number calculated by TREB and is supposed to account for differences between the individual properties sold such as number of bedrooms/bathrooms/age/proximity to schools/parking/etc. In theory it sounds very good, however there are some shortcomings with this number as well. As you mention, the May TREB Market Watch Report shows that the Richmond Hill detached Benchmark price is $1,555,400 in May and $1,319,500 in October. The October 2017 report also shows that the year-over-year change in the benchmark price is 1.37%.

      If you look at the October 2016 TREB Market Watch, it shows that the Richmond Hill detached benchmark price is 1,300,000. If you calculate the YoY change based on the October 2017 and October 2016 numbers, you will find that the change is 1.50%. Close to the change published by TREB, but not totally the same. Interesting, right?

      I’ve gone through this problem in a lot of detail (I know, I need to get out more). In the example we’re discussing, the difference is quite small. However, in some cases the changes that an analyst like me can calculate using the publicly available benchmark data are very very different from those published by TREB. When I contacted TREB to ask them about accessing all of the data needed to use their methodology paper to link benchmark prices over time, they said to me “That information is not publicly available”.

      That’s a long way of telling you that the Benchmarks published by TREB are a black box that the public is not allowed to look into. Because of this, I couldn’t make meaningful use of the benchmark numbers on my site. Trust me, I tried.

      One more thing I’ll add to this already-too-long response: I know of properties in my area that have had incorrect attributes entered into the MLS sold data set (example: a listing says it had parking when it does not). For this reason, I am a little bit suspect of using such a data set to “adjust”, particularly when the adjustments are not made public.



  2. Many thanks for your thoughtful and detailed reply Hanny. Very interesting and you’re right, the method of calculation can clearly make a big difference.

Leave a Reply

Your email address will not be published.