There is a quote we often reference on the team, and it goes like this.
“There are three kinds of lies: Lies, Damned Lies, and Statistics”.
It has been attributed to Mark Twain, who himself attributed it to British Prime Minister Benjamin Disraeli, who might never have said it in the first place.
Regardless of the source, we periodically quote it, particularly around the start of the month, when the latest round of real estate statistics is published by our friends at TRREB, the Toronto Regional Real Estate Board. We absolutely believe that the data that is published is accurate, so we’re not questioning that aspect.
The reason we reference that quote is more to do with the headlines that start appearing around this time. TRREB provides tons of data each month in the form of a Market Watch Report, which is disseminated to media outlets, who then decide what data points make an interesting story.
These headlines often take the complicated market stats and come up with a simple statement that, when you dig into it, is either inaccurate or flat wrong. Sometimes this is because the media wants a more interesting story to push, and other times it is because the stats themselves can be misunderstood or wrongly interpreted.
It’s early January, 2025, so we now have the stats for the December, 2024 market, which means we also have the full stats for last year. Let’s go over the three biggest lies – uh, we mean, flawed interpretations of the data – and what it actually means.
It’s just a bit off, by that we mean, the exact opposite.
Let’s talk about prices for a bit.
The default standard for talking about real estate prices is the average price, which is simply the number of sales divided by the total dollar value of sales. If you had 10 sales that equal ten million dollars, the average sale price is $1M. If you had 10 sales that totals twenty million dollars, the average sale price is $2M, and so forth.
If you look at any headlines out there in January, 2025, they will be saying that the average price in the GTA went down 1.6% in the past year. It was approximately $1,084,000 in December, 2023 and it was about $1,067,000 in December, 2024. That means, on average, prices went down in the GTA by about $17,000. Got it.
Now, the average is a good measure of the total scope of data. It’s the most commonly used measure but it can be misleading when data is skewed or has outliers. For example, if a few multi-millionaires are included in a set of household incomes, the average income will be higher than reality.
So, is our housing data skewed? Well, if we look at what sales took place in different price bands in the data, it definitely isn’t equally distributed. In both December of 2023 and December of 2024, about 54% of sales were under $1 million, about 39% were between $1 million and $2 million and about 7% were over $2 million. That means, our real estate data is skewed to a moderate extent, and about 70% of the total sales take place in a price range between $600K to $1.5M.
In instances where data is skewed, the median, or the middle number of the data set, is a better measure to use. Going back to our example of 10 sales that equal $10M, for a $1M average, what if we looked and found that nine of the sales were for $800K and one was for $2.8M. That averages to $1M, but the data is skewed with a high outlier at the top end for that $2.8M sale. The median, or medium number would be $800K, which is more representative of the actual typical price.
If the average sale price in the GTA has gone down 1.6% in the last year, losing about $17,500 in value, what about the median sale price? It turns out, that is a very different story, in fact, pretty much the opposite.
The median sale price in December, 2024 was $930,000, which is about 13% lower than the average price. If we look back in December, 2023, the median sale price was $908,750, which means that prices went up by about 2.3% in the last year.
This shows how it absolutely matters how statistics are interpreted and which measures are used. In this case, prices have gone down on average by $17,500 in the past year, but our dataset is skewed, which means the median is a better measure to use, which shows prices have gone up by over $21,000 in the past year. Damn statistics.
It just takes two weeks longer than we say it does. What’s the big deal?
Our next example of how statistics can be misleading comes to us in the form of the days on market. Don’t worry, we’re not going to talk about average or median in this case.
Instead, we’re going to point out one of our favourite gripes about our market data, which is that the length of time it takes for properties to sell uses what is called Listing Days on Market, or LDOM. If a house goes up for sale on January 1st and sells January 10th, that means the listing was on the market for 10 days and it took 10 days to sell. Pretty clear, right? Take all of the properties that have sold in a month or year and you get the average days on market, or how long it takes real estate to get sold.
In the GTA, it took, on average, 25 days for a property to sell last year. That’s the average for all housing types, all price points, across all parts of the GTA, but it does give us an idea of how quickly (or slowly) the market is moving. We can compare the 25 days it took on average to sell in 2024 with the 19 days it took on average in 2023 and say pretty firmly that it was a slower year.
The problem with the Listing Days on Market has to do with the fact that the Toronto Regional Real Estate Board allows properties to be listed, terminated, and then relisted. This can be done as many times as a seller and their agent want to, so it radically skews the accuracy of our days on market stats.
In our previous example of a home that took 10 days to sell based on the listing going up January 1st and selling January 10th, what if that wasn’t their first listing of the home? What if it was listed October 15th for a higher price and it didn’t sell? What if it was relisted November 24th for a lower price and still didn’t sell? If that home sold January 10th after two previous listings, the 10 days on market isn’t actually accurate. In fact, if we look at the property days on market (or PDOM), it was actually on the market for 87 days. This PDOM stat takes all of the recent listings and combines them to get a days on market that is reflective of how long it actually took the property to sell.
A large number of listings in the GTA have more than one listing attempt, which means that the listing days on market, which is the stat that is used in the media to describe how quick or slow a market is these days, is often wildly inaccurate.
To prove that, the average listing days on market in the GTA was 25 days in 2024, but the average property days on market was 38 days. That’s 13 days longer, or almost two weeks longer. Back in 2023, the average listing days on market in the GTA was 19 days and that went up to 28 days for the PDOM. While it took longer in 2024 to sell than it did in 2023 by both metrics, if we compare the LDOM, it was 6 days slower, versus the PDOM, where it was 10 days slower. When we’re talking days, an additional four days is a significant slow down, so the media stories about the slow real estate market last year are actually understating how much slower it was to transact in 2024. Say it with us, damn statistics.
Tale of Two Townhouse Territories
The final example we’ll give about how statistics can be misleading has to do with the danger of taking an overall stat and applying it to any given subset of the data.
While we previously challenged the accuracy of the average sale price that is used in headlines, let’s look at how even if it was accurate in general, how egregiously wrong it can become when you apply it to specific housing types and market areas.
Here’s the chart from the Market Watch report with the sales and average price by major home type, including the year over year change.
You’ll note we’ve circled the Townhouse statistics, as it gives us a great example of how using the average statistics can be plain wrong when it comes to a specific subset of the data.
The average price (not the median) went down in the GTA by 1.6% in 2024 when you compare December 2024 to December 2023. If you owned a townhouse, however, the average price dropped by just 0.3% in the past year. Go deeper into the stats though, and you’ll see that in Toronto (the 416), the average price for a townhouse went down 18.2% in the past year, whereas the average price outside of Toronto (the 905) went up by 4.7%!
The moral of the story is that before you celebrate or shake your head in disgust at the average increase or decrease in real estate prices according to the latest headlines, think about whether that average is at all accurate when it comes to your home. Odds are pretty good that it will be a very different story for your home compared to what is happening on average!
We love data, but we’re also very aware that statistics can be misleading. Whenever we work with buyer or seller clients, we make sure to understand what is going on in the specific subset of the market they are transacting in – and we use that data to get our clients the best results. If you like the sound of that, get in touch to discuss how we can help move you forward.
