This methodological explanation is provided by rockgolf, former Sporcle Music editor and creator of the Canadian Top 150 quiz.


I like big lists and I cannot lie.

With July 1st 2017 being the 150th birthday of Canada, I wanted to do something special. Since music is my forte in Sporcle I determined I would create a list of the 150 biggest songs by Canadian artists in the past 49 years. Why not 50? Well, there weren’t any significant Canadians on the charts in 1967, but in 1968 the CRTC (the Canadian equivalent of the FCC) introduced a rule that Canadian radio stations had to play 30% Canadian content: either the artist performing the song or the songwriter had to be Canadian. This had a massive effect on Canadian music, and lots of them even became successful south of our border, in the US.

I’d have loved to use Canadian charts to compile this information, but there are major gaps in getting detailed info for most of that period. So instead, and probably to the benefit of most Sporcle players, I’ve gone with the US Billboard Hot 100 charts as my source of info.

The starting point was getting a week-by-week chart listing throughout the history of the Billboard charts. Joel Whitburn’s Record Research books like TOP POP SINGLES are great for summary information, and I’ll keep buying them as long as they keep making them, but I needed more detail. Years back I found a website that had the level of detail I needed in a downloadable mammoth spreadsheet, with song listings dating back to 1901! Sadly, the website that provided and maintained that data went offline about 3 years ago. So I’ve had to resort to more complex methods of getting the data, including referencing the archived Hot 100 charts at itself.

After a lot of work, I had a complete list of chart runs of every single on the chart in the past 50 years. Now I had to analyze it. The simplest method is to assign points to each chart position, for example #100 gets 1 point, and #1 gets 100 points. But that would mean a #1 song would only get twice the points of the #50 song that week. That doesn’t reflect reality. If you look at, for example, any Spotify chart there’s a huge difference between the spins for a #1 vs a #50 track – about 4 or 5 to 1 ratio in the number of plays.

So I added bonus points for higher chart positions: 10 bonus points for any song in the top 40, an additional 10 for the top 30, with even bigger bonuses for getting into top 20 and top 10, then even more bonus points at each position from #5 to #1. The end result is that a week at #1 is worth about 6 times the value of a week at #50. It sounds complex, but once I used Excel to develop the formula for one song, I could easily copy it to all the others.

Now I had a point total for every song, and could sort that as a single value. But there was a further problem.

Over 50 years, the turnover rate of the Hot 100 has varied wildly. In the 1960’s, a #1 song might only be on the charts for 12 weeks. (Can’t Buy Me Love by The Beatles reached #1 and was only on the Hot 100 for 10 weeks.) But by the 2010’s, the turnover rate on hit records slowed waaaay down. It was not uncommon for big hits to be on the chart for over a year and minor hits to last 30 weeks. Unless there was some sort of balancing to account for this, my list would be almost entirely from the last few years. A 2015 song that never got into the top ten is likely to have more raw points than a #1 song from the 1970’s.

Therefore, I applied some statistical analysis. I determined the total points of the top 10 songs from each year. Then I got a 3-year running average points for each year, its previous year and its following year. That 3-year average was compared to the average across all years to determine a “handicap” factor. As an example, that meant that the songs from 1972 got a handicap of 142.5%, while songs from 2014 got a handicap of 59.2%. So a song that in 1972 got 1,000 points would be inflated to 1,425 adjusted points, but a 2014 song with the raw score would be reduced to 592.

By taking the raw points from the chart run and multiplying it by the adjustment factor, songs from every year were playing on a level field. I know this sounds ridiculously complex, but thanks to Excel know-how, I was able to go from the raw chart numbers to the adjusted chart numbers list in a few hours.

That was the easy part.

The trickier part, because it had to be done manually, was trying to exhaustively identify every Canadian act. It helped here to have been a Canadian music fan for about 50 years, and to own the fantastic Canadian CD compilations Oh What a Feeling! There were 3 volumes each containing 4 hit-packed CDs released between 1996 and 2006 with just about every significant Canadian pop act whether they had American success or not. So one CD could have Leonard Cohen, The Tragically Hip and Shania Twain. Another would have k d lang, Neil Young and Blue Rodeo. And every song was great!

After that I had to add in the more recent Canadian invasion: Bieber, Drake, Weeknd, Cara, Mendes, etc. And scour the artists for any I missed. I hope I got ’em all, but if you spot an exception, please, please add a comment to the quiz to let me know and I’ll add them in if they qualify.

Enjoy the quiz! And Happy 150th Canada Day!