BC Liquor Store Open Data
I came across an interesting open data set that has information on products BC Liquor Stores (BCLS) carry.
Being based in BC and often frequenting the liquor store with a poorly thought out idea of what to buy, I figured I’d have a look at the data to see if I can discover some new products, and/or simplify things for myself next time I go shopping for booze.
Value for Your Money?
I work as an economist in my professional life.
Naturally, this makes the concept of price vs. quantity something ingrained so deep in my brain it’s hard to recall ever not thinking about it.
So, whenever I’m at the liquor store (or any store, really), I’m always interested in how prices relate to the items before me in terms of quantity, quality, aesthetics, taste, etc.
Thankfully, I’ve reached an age and income level where my tastes for alcoholic beverages aren’t predicated by a university student budget constraint and a penchant for getting as intoxicated as possible.
With that said, I can’t deny the question never crosses my mind about what’s the best ‘bang for your buck’ at the liquor store, if one were (hypothetically) trying to maximize the amount of booze per dollar spent.
Best ‘Value’
After calculating price and quantity of pure alcohol per milliliter one receives for each product the BCLS carry, here’s the top 25 products if you’re trying to get as drunk as possible on the cheap (not advised):
Looking over the list, it seems like no matter what you choose, you’re in for one of the most painful hangovers of your life if you go down this path.
You might save a few bucks in your wallet, but you’ll probably shave at least a few years off your life expectancy.
Worst ‘Value’
What if money were no object, and one was looking to spend as much money as possible while receiving the least amount of pure alcohol per dollar spent.
What would that list look like?
If you don’t care about ‘value’, these might be some options: 
Topping the list is Macallan Genesis 72 Year Old (whiskey), coming in at a whopping $82,500 per bottle and an exorbitant price of $261.90 per milliliter of pure alcohol.
Interestingly, (but predictably, on reflection), most of the products topping this list are in the scotch/whiskey sub-category, which seems pretty sensible.
I probably won’t be picking up a bottle of Macallan Genesis 72 Year Old anytime soon, as that might set off a series of events ending in a quick divorce.
On the other end of the spectrum however, the economist in me is very intrigued what the best ‘value’ tastes like.
I’ve had my share of Colt 45 in my younger days, but I’ve actually never tried Pacific Western Cariboo Malt beer, and since it’s literally the cheapest product in the entire BCLS catalogue by price per milliliter of alcohol content, it’s not much of a loss if it’s absolutely terrible.
Which, I anticipate it will be.
But who knows!? Might get lucky! 🍻
Cooking with Wine?
I’m no Michelin starred chef, but I do like to cook, and I really like to cook with wine whenever the opportunity arises.
Living on the West Coast, we get some pretty amazing seafood out here, and there’s almost nothing better than some mussels or prawns sauteed in some form of buttery-garlicky-winey sauce.
Since seafood can be pretty pricey, and since one should probably only eat so much of it, I don’t regularly find myself shopping for cooking wine.
So, when it comes to picking up a new bottle of cooking wine, it’s always a pain looking for that cheap (but half decent) bottle to cook with1.
It seems like these data might be useful to find a good/cheap cooking wine, and in this instance, I’m actually not really interested in the cheapest price per milliliter of alcohol, since the alcohol is just going to be cooked off anyway.
So, I’ll simply be ranking the data by cheapest price per milliliter, while filtering on wine.
Here’s a list of the 25 ‘best value’ (1 Litre or less) bottles of wine that might work as cooking wines: 
There’s probably some oaky wines in that list, so one should check the oakiness before buying, but it looks like Pato Criollo, or Bianchi New Age from Argentina might be good snags.
Or, maybe even the Caviste brand of wines from California.
Wine Price Distributions
It seems like it’s just a ‘known’ thing that French wines are expensive, but in my experience, I find there’s actually a lot of decent bottles of French wine for pretty reasonable prices (for my taste).
Since I’ve got data on the price per bottle of every wine in the BCLS, I thought it’d be interesting to look at the price distributions of wine to see if the data backs up the notion that French wine is among the most expensive (at least, in the BCLS inventory).
Here’s a box plot2 of the price distributions of wines by country of origin, sorted by price:
Looks like there’s actually some truth to the notion that French wines can be pretty expensive. The median price of a bottle is around $110, but of course, there’s still plenty to choose from around the $50 mark as well (which isn’t exactly ‘cheap’).
One country that sticks out as being kind of interesting is China - can’t say I’ve ever tried Chinese wine, other than Shaoxing cooking wine.
Digging into the data manually a little bit, it turns out most of the Chinese wines are actually of a sub-category called ‘Asian Spirits’, so, it’s not really the kind of wine one might have a glass of with dinner.
Canada also stands out as a country with quite a lot of wines in the ‘good value’ range. Whether these wines are actually good or not is a different question, but I might shop around the Canadian wine section a little more on future visits to the liquor store!
Explore the Data
Maybe you’d like to poke around the data yourself? Have a look around the interactive plot below:
Looking to explore these data even further? Here’s a link to the Github repo containing the code to pull the data and generate all the plots.
Cheers! 🥂
Footnotes
Side note: I actually wasn’t too sure what wines are actually ‘good’ to cook with, and after a quick Google, apparently one should steer clear of oaky wines. Good to know!↩︎
Note that this plot is interactive. You can zoom in and filter variables on and off. To reset the chart quickly just double click anywhere inside the chart.↩︎