How the area fit score is calculated

1. Each component is turned into a percentile — where this area sits among the postal areas that publish that component.
2. Percentiles for components where less is more fit (industrial releases, rent) are flipped, so every component runs the same way before anything is added.
3. Your weights are renormalised over the components this area actually has, so they add to 100%. A component nobody publishes for the area is left out, not scored zero.
4. The score is the weighted sum, out of 100. Below 3 available components there is no score at all.

• Everyday amenities within walking distance. Grocery, pharmacy, parks, primary and secondary schools, child care and libraries — averaged across the seven, weighted by where people live. Proximity only: nothing here says anything about any of these places, only how far away they are.
• Transit access. Residents with a transit stop within a 1 km walk, weighted by how many trips serve it. Kept separate from the other amenities because for most people it is a decision on its own rather than one of seven.
• Reported industrial releases nearby. Everything facilities within 10 km reported releasing to air in the most recent National Pollutant Release Inventory year. Reporting is threshold-based, so a low figure means little was reported and not that nothing is emitted.
• Rent being paid. CMHC’s surveyed average for a two-bedroom, across the survey zones covering this area and weighted by population. Rents being paid by sitting tenants included, so it is not what a unit would be advertised at today.
• Dwellings per square kilometre. How built-up the area is. This is the one component with no direction of its own — wanting a dense neighbourhood and wanting space are both ordinary reasons to move, so you choose which way it counts.

That is not caution for its own sake. A single “fit” figure blending income and tenure, published beside a breakdown of who lives where, is a tool for sorting people by neighbourhood. Every provincial human rights code prohibits exactly that in housing, and the fact that a methodology section exists does not change what the number does.

Crime is excluded too, and for a second reason on top of the first. It is published for whole metropolitan areas, not postal areas — so for most of the country the nearest available figure describes a region containing hundreds of thousands of people. Rolling that into a postal-area score would invent precision that does not exist, on the single most consequential variable on the site. Crime is shown on each area page, at the geography it actually describes, and it is never scored.

Air quality is excluded for a duller reason: the Air Quality Health Index is a live reading, republished hourly, and there is no stored history behind it. A fit score that moved every hour — and that fell for a whole province during a week of wildfire smoke — would not be measuring the place. Reported industrial releases cover the same ground with an annual figure, and that is the component you see instead.

There is no grade, no letter, no star rating and no ranked list of best areas anywhere on this site. Those are the same number wearing a costume, and the reader did not choose their weights.

Each component is normalised to its rank among the postal areas that publish it, not scaled between the highest and lowest values. That is not a stylistic choice: one postal area reports 194,374 tonnes of industrial releases to air in a year. Scaled between the minimum and that maximum, every other area in Canada would land in the top few percent of that component, and it would contribute nothing to anybody’s score while appearing to contribute a fifth of it.

The denominators are different for each component and the page says so beside every row. Amenity measures are published for 1,629 postal areas; CMHC surveys rents in 918. A rent percentile is a rank among the areas CMHC surveys — which are the larger centres — and not a rank among all 1,646.

A consequence worth stating plainly: two areas scored on different components are not comparable, even though both numbers run 0 to 100. Renormalising is what makes a partial score honest, and it is also what makes the difference invisible, so every score built on fewer than the full set says which ones were missing.

Four of the five components have a direction almost nobody argues with. Density does not. Wanting a dense, built-up neighbourhood and wanting space around you are both ordinary reasons to move, and picking one as the correct answer would be exactly the editorial verdict the rest of this page refuses to make. So the direction is yours, and the score changes when you change it.

Every component opens weighted the same. That is a starting point and not a recommendation — there is no research behind equal weights and no claim that they describe anyone. The controls run from “ignore” to “a lot”, mapping to 0, 1, 2 and 3, and are renormalised so that whatever you pick adds to 100% across the components the area has.

A component set to ignore is removed from the renormalisation entirely rather than multiplied by zero, so ignoring rent does not quietly change what the other components are divided by.