Which songs have most weeks in the charts? |
|
One of the users asked "which songs spent the most weeks on the
charts, whether they reached number one or not". If you just want
to see the results without having to understand how we got them here
they are:
This is a really interesting question. The first component of which is to clarify
which charts, if one is just interested in the US and UK, for example, the obvious
thing would be to obtain both those charts (see the
Song Charts page for details of where to get them
from) and do some simple calculations. Of course you have to take a view on how
to weigh those two charts (are weeks in the official UK chart equivalent to weeks
in Billboard?). The other curveball is the odd handling that Billboard does for
songs that spend more than 20 weeks in the chart.
The fact that this user asked tsort.info this
question suggests that they don't want to just look at a couple of national charts, they
want to get an impression of which songs spent most weeks in the "charts of the
world". So a simple bit of processing on a few "notable" charts isn't
enough. We can just find each chart that notes the number of weeks and add them up,
right? Well we can do that, of course some of our source charts have months noted, but
for the sake of simplicity we can just multiply them by 4 to make the numbers about correct.
All time raw
When we do that we get the following results:
Which are an "interesting" selection of songs (something in there
to aggravate almost anyone we feel). But there are a few things to note. First
the total reflect the charts available, that have a note of duration, so the
total is fairly meaningless, if we had twice the number of charts the total would
be twice as big (the order could well be similar though). Second there is an
emphasis on songs that were big in Europe, that's because each country in Europe
has its own chart, so the US Billboard chart gets 52 weeks in the year and so
does Belgium. Finally there are a lot of entries from the 2000s,
because we have a lot of charts from the 2000s.
So taking all those into account can we make the results a bit less dependent on
the particular charts we have included? Well for one thing we can aggregate charts
by region. For various reasons it makes sense to group the charts available into
four buckets: North America; Other English speaking countries; the rest of Europe;
and everywhere else. The final group doesn't have enough charts in it, but here
are the all-time results for the first three:
All time North America
In the US and Canada songs seem to have had long runs in the 1950s
and 2000s and not inbetween.
All time other English speaking countries
Snow Patrol? Would not have guessed that.
All time rest of Europe
All time best results
When we combine those we get the following, our guess as to the longest running songs in charts
history across the whole world (well the bits we have charts for).
Longest Runs by Decade
We've done the same calculation for each decade from the 1950s to
2000s.
1950s worldwide longest runs
1950s North America
1950s other English speaking countries
1950s rest of Europe
1950s raw longest runs
1960s worldwide longest runs
1960s North America
1960s other English speaking countries
1960s rest of Europe
1960s raw longest runs
1970s worldwide longest runs
1970s North America
1970s other English speaking countries
1970s rest of Europe
1970s raw longest runs
1980s worldwide longest runs
1980s North America
1980s other English speaking countries
1980s rest of Europe
1980s raw longest runs
1990s worldwide longest runs
1990s North America
1990s other English speaking countries
1990s rest of Europe
1990s raw longest runs
2000s worldwide longest runs
2000s North America
2000s other English speaking countries
2000s rest of Europe
2000s raw longest runs
The process of calculating these results is a bit complex, for that
reason the data shown is not recalculated each time the site is
regenerated. So the information may not exactly match the current
version of the data. These results were calculated from the
2.3.0057 data.
Back to Introduction
|