What 121215 active listings say about Ontario

On July 26, 2026 we froze a snapshot of every active, published TRREB listing in our production database — 121,215 of them — and spent some time just reading the data. Everything below is asking-side only: sold prices live behind VOW rules and belong on registered-visitor surfaces, not in a public essay. Asking data alone turns out to be plenty revealing.

Toronto is a rental market wearing a sales market’s clothes

The Toronto rollup held 26,038 active listings that day — and 13,460 of them were leases against 12,578 sales. More than half of what was publicly available in the city was for rent. No other large market in the snapshot tips that way so decisively; Mississauga’s 5,938 actives split roughly 55/45 toward sale. If you build websites for agents, as we do, this is not trivia: a Toronto agent’s site that treats rentals as an afterthought is hiding half the city’s live inventory from the visitors most likely to become future buyers.

The price ladder is really a geography lesson

Median asking prices in the snapshot draw the Greater Toronto map in one column of numbers: London at $619,900, Barrie at $720,000, Toronto at $779,900 — held down by its condo-heavy mix — Mississauga at $839,900, Brampton at $879,900, Markham at $988,000, Richmond Hill at $1,199,000, and Oakville out at $1,349,900. Vaughan’s median landed on exactly $999,999, which looked like a bug and turned out to be something more interesting — that one got its own post.

Nothing sits for long, anywhere

Computed days-on-market medians across the fourteen largest markets ran in a tight band from 34 days (Kitchener) to 46 (London), with most cities clustered in the high 30s. For all the regional price spread, the tempo of Ontario inventory that week was remarkably uniform — a listing priced to its market moved on roughly the same clock at Oshawa’s $719,000 median as at Oakville’s $1.35 million.

The data is messier than the dashboards admit

Two things the snapshot taught us about MLS data itself. First: those 121,215 listings carried 515 distinct city values. Toronto is stored as 35 district codes plus a handful of plain “Toronto” rows; London exists only as London South, East, and North; Ottawa arrives as some 37 neighbourhood-level values — Kanata, Barrhaven, and Orleans are each their own “city” as far as the feed is concerned. Hamilton cuts the other way: the bare Hamilton value held 3,914 listings while Stoney Creek, Ancaster, Waterdown, and Dundas sat beside it as separate values, so a naive count understates the real metro considerably. Any city page — like our Toronto page — is quietly a rollup someone had to define, and two platforms can disagree about a city’s inventory while both being technically faithful to the feed. Second: the feed’s days-on-market field was populated on under 2% of active rows, so every DOM figure above is computed from the listing date instead. When a vendor shows you a statistic, the honest question is not “what does it say” but “what did they have to do to the data to say it.”

What an agent should take from this

Publishing market statistics on your own site is legitimate and powerful, but the rules and the pitfalls are real — we wrote up what you may publish separately. Our own takeaway is narrower: aggregate numbers are only trustworthy when someone owns the unglamorous plumbing underneath them, from city-value rollups to recomputed DOM. That plumbing is most of the work, which is precisely why it is worth doing once, properly, at the platform layer.