Which Ontario Cities Have the Most Restaurants?

I’ve been thinking that restaurants are an interesting indicator of a community’s character.  Having a lot of restaurants can indicate a lot of different things – strong tourism sector, a highly mobile population, a role as a regional centre for the food and drink sector.  Kind of an interesting question to wonder if Toronto’s urbanity actually translates into more restaurants, or whether that’s just an assumption made by urbanites…

Being pathological about data and oversimplification, I decided to measure it by creating a really simple metric – restaurants per 1,000 population, using 2016 census data for the 27 cities in Ontario with populations over 100,000.  Call it the Urban Foodie Index.  To measure the number of restaurants – this could be the contentious part – I used TripAdvisor.com and simply captured the number of restaurants listed in that City.  Read below if you’re interested in my take on possible shortcomings of that data source. Long story short, I think the data is good but not perfect.

All that said…the results were interesting. Turns out, Toronto’s number of restaurants per capita is below that of many other cities.  Topping the table…Windsor, with the most restaurants per capita of all large cities, followed by Kingston, Waterloo, and St. Catherine’s.  Barrie rings in at #5, just ahead of Toronto at 6th.  Interestingly, the three cities with the least restaurants per capita were Milton, Ajax, and Brampton.  Here’s a nifty bar chart and the table with the data.

UFI Chart July 2017UFI Table July 2017

Many reasons I could speculate for these results.  Cities that have strong tourism sectors did predictably well.  Windsor has a casino and is adjacent to a US city of 2M people.  Kingston and Ottawa have strong tourism sectors, whereas many of the cities in the lower half of the table are less of a tourism destination.  In defense of that assumption, I checked Niagara Falls (439 restos, pop 88,071 = UFI of 4.98) and Collingwood (77 restos, pop 21,793=UFI of 3.53).  Both big tourist towns, both had high scores.

Also, I think central place theory comes into effect.  Even some of the more prosperous suburban GTA cities did not have a high score (Oakville, Markham for e.g.).  But cities like London, Kingston, and Barrie did well.  These are cities that are the urban centres in relatively populous surrounding small town/rural areas; the obvious theory here would be that suburbs tend to “leak” customers to neighbouring larger urban centres, whereas standalone urban centres are home to stronger regional clusters.  That said – Vaughan bucked the trend as the only GTA city with an above-average UFI outside of Toronto itself, although Burlington scored at exactly the Ontario average.

Anyway – this post really tells you two things:  1.  The cities we might assume are restaurant sector powerhouses may not actually be, and 2. I had too much spare time this weekend.

 

NOTES ON METHODOLOGY:

I stress tested the data a little, using my home city of Barrie.  Checking the maps, it appears the listings corresponded roughly to the municipal boundary, although there were a few exceptions.  Four restaurants out of 365 were outside the city boundary, one by a matter of meters, three by several km.  That said, I would assume these exceptions would likely distort results similarly for other cities on the list (I could well be wrong).  I also found one restaurant that is closed (very recently) and one that seems like it never opened.  That’s a pretty low error rate.

That said, there are definitely restaurants missing from Tripadvisor.  Most chains like Tim Horton’s seemed to only have some of their locations listed (Barrie has four AvocoBar outlets for example, Tripadvisor only lists three).  Again, one has to assume since the information is basically crowdsourced, this underrepresentation would apply equally to all cities.  But clearly another source of potential error.

Last disclaimer – there are probably lots of other sources of error I haven’t mentioned.  That said, I claim no rigour or scientific method to any of this, it is mostly for amusement.