Chart Year Breaks

Hello fellow members. I'm seeking your opinions on
something pertaining to a personal project I'm working
on.

The project itself is nothing spectacular. A typical
calculating of a year's biggest hits using my own point
system. What I want to ask your help on is where should
I place those records that crossover from the end of one
year to the beginning of the next. I've seen this done
various ways, and all have had their individual strengths
and weaknesses.

For instance, the Whitburn way is to place the track into
the year where it REACHES its peak position. In most
cases, that works just fine, but there are some things
that make this method questionable. The biggest example
is, of course, the Monkees' "I'm A Believer", which
becomes the #1 hit of 1966 using it. The issue is that
the track only spent 4 weeks in that year, with 6 of its
7 week run at #1 falling into 1967.

Another method is to place a record into the year that it
earns the majority of its point total. Again, sounds
good. But using that method, "Please Don't Go" by KC &
The Sunshine Band ends up in 1979, despite having its
single week at #1 falling in 1980.

My own method was to use a various of the Whitburn
method, which would take a look at a track's entire peak
position run to determine placement. Thus, "I'm A
Believer" moves into 1967 because of the 1 to 6 week
split. In cases where a track splits its peak run down
the middle, I'd use the rest of the Whitburn method
(weeks in top 10, weeks in top 40, week on chart) and
apply it to each year's separate chart run. That method
has worked for me for the most part, but one thing keeps
tripping me up.

(Now, I get to show my nitpicky nature to its fullest.
^_^ )

The thing that I keep questioning is the first week of
each year. I know it should be simple, right? If the
date is in January, it goes it that year. Except for
these are not set dates. These are week ENDING dates.
So, tell me. Does the week ENDING January 1, 1961 really
belong in 1961 when 6 of its days are in 1960? (I told
you it was nitpicky.)

I keep going back and forth with this. "Ignore the
ENDING and just use the date." "No, place that week in
the year that has the most days in it." "That's not
right. You should split that week's totals in ratio with
the number of days in each."

I know this is all rather trivial in the grand scheme of
things. And EVERY method is going to create anomalies
that are going to make me question them. But I would
still like to hear from all of you. Because I know
you've all done your own versions of this project at one
time or another. How did you handle it?

Thanks in advance for your takes on this.

IMO, if you're using a point system, then I'd say just go with the year it accumulated the most points. I do that with my own
personal projects.

As for chart dates, yes, they are for "the week ending", which means a chart dated January 1st really only has one day in the new
year. Also keep in mind that the data used to compile the published chart (especially in the "old" days) was likely at least a
week or two behind, thus you'd often get Christmas songs peaking in January, when they really peaked in December, etc.

Paul Haney wrote: IMO, if you're using a point system, then I'd say just go with the year it accumulated the most points. I do that with my own personal projects. As for chart dates, yes, they are for "the week ending", which means a chart dated January 1st really only has one day in the new year. Also keep in mind that the data used to compile the published chart (especially in the "old" days) was likely at least a week or two behind, thus you'd often get Christmas songs peaking in January, when they really peaked in December, etc.

Most points is a good way to go, to be sure. Especially
as later I intend to include points from Cash Box, and at
times they and Billboard disagree about what year a song
peaked.

The time lag is something I'd considered actually.
Thought about moving each chart back by a couple of
weeks. Finally decided to go with the published dates in
the end, though. Seemed less likely to screw it up that
way.

However, because of the lag, I had decided to continue
Joel's Christmas single year placement practice.

LunarLaugh wrote: I actually prefer the way Bronson's Billboards Hottest Hits book handles the ranking year by year; the song's entire chart run is considered but sorted from full year from January to December instead of the traditional November to November ranking of Billboard year-end charts.

Which is what I'm doing (Jan-Dec), but using my own point
system. And I could be wrong, but I thought Bronson only
used weeks in the Top 30 to better compile an overall Top
3000 or something like that. I'm not (at this point)
planning to do that. Therefore, I'm using a track's
entire Hot 100 run.

Though, to be fair, my point system does reward weeks in
the top 30 a fair bit more than lower positions, so a
track really has to pile up a lot of extra weeks to make
an impact overall, so maybe Bronson had the right idea.

This may not help with interpreting chart data to assign a "year"
to a particular song, but here's my two cents on this topic as a
radio programmer.

It became very clear early in my radio career how long it takes
average listeners to become aware and familiar with most brand
new songs. With a few exceptions (novelty songs or songs by
an extremely "hot" current act) during the first few weeks of
airplay just a small fraction of the audience is even aware that
the song exists. High familiarity kicks in around the time a song
is peaking on the charts and even more so in the weeks
following. When programmers realized that factor it led to the
creation of "recurrent" categories that continued to give those
songs significant spins rather than move them to "hold"
categories that "rested" recent hits before adding them to active
oldies rotations. Today those recent hits form the bedrock of
playlists for most current-based radio stations.

My point is that if you want to assign a "year" designation to a
song released late in the year look at the actual date the song
peaked on the chart then add several weeks thereafter. If a song
spent one week at #1 during the last week of December and
several more during January, the subsequent year is when most
listeners would likely place that song in their memory.

Also listening patterns become disrupted during the final weeks
of the year as stations play holiday music and run year-end
countdowns and special programming. So airplay for the regular
playlists were often disrupted. Also due to the holidays listeners
may have spent less time listening as they had other things to
do.

Unlike music & chart geeks the average radio listener is not
hyper-aware of brand new songs immediately upon release. It's
not the date when the song was recorded or when it was
released or when it first charted. It's the time frame when the
average person became fully aware of the song and has
memories corresponding with hearing it frequently.

Edited by CountryPD